From David.Bear@asu.edu Thu Nov 1 17:21:03 2007 From: David.Bear@asu.edu (David Bear) Date: Thu, 1 Nov 2007 09:21:03 -0700 Subject: [OpenAFS] mac install instructions Message-ID: <20071101162103.GB3988@asu.edu> It's been 3+ years since I installed openafs on a mac. The experience then was less the possitive. I'm hoping to find an installation readme somewhere to go over. I am not finding anything like that on openafs.org. do I have to download the openafs for mac before I get the readme? also, I'm interested in best practices (or things to avoid) when using openafs client on mac. -- David Bear phone: 602-496-0424 fax: 602-496-0955 College of Public Programs/ASU University Center Rm 622 411 N Central Phoenix, AZ 85007-0685 "Beware the IP portfolio, everyone will be suspect of trespassing" From shadow@gmail.com Thu Nov 1 17:25:58 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 1 Nov 2007 12:25:58 -0400 Subject: [OpenAFS] mac install instructions In-Reply-To: <20071101162103.GB3988@asu.edu> References: <20071101162103.GB3988@asu.edu> Message-ID: ------=_Part_13129_85249.1193934358905 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On 11/1/07, David Bear wrote: > > It's been 3+ years since I installed openafs on a mac. The experience > then was less the possitive. I'm hoping to find an installation > readme somewhere to go over. I am not finding anything like that on > openafs.org. do I have to download the openafs for mac before I get > the readme? um. double-click the installer package in the dmg, optionally edit /var/db/openafs/etc/ThisCell, reboot, done? ------=_Part_13129_85249.1193934358905 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On 11/1/07, David Bear <David.Bear@asu.edu> wrote:
It's been 3+ years since I installed openafs on a mac. The experience
then was less the possitive. I'm hoping to find an installation
readme somewhere to go over. I am not finding anything like that on
openafs.org. do I have to download the openafs for mac before I get
the readme?

um. double-click the installer package in the dmg, optionally edit /var/db/openafs/etc/ThisCell, reboot, done?
 


------=_Part_13129_85249.1193934358905-- From rich@nd.edu Thu Nov 1 17:42:33 2007 From: rich@nd.edu (Rich Sudlow) Date: Thu, 01 Nov 2007 12:42:33 -0400 Subject: [OpenAFS] Mac OS X + aklog integration In-Reply-To: <0DAC3799-F505-4D7B-9BD1-AF25A5B03359@inf.ed.ac.uk> References: <47289205.5070804@kickflop.net> <0DAC3799-F505-4D7B-9BD1-AF25A5B03359@inf.ed.ac.uk> Message-ID: <472A01F9.4070009@nd.edu> Simon Wilkinson wrote: > > On 31 Oct 2007, at 14:32, Jeff Blaine wrote: > >> I don't see a native way to run aklog once krb5 creds >> are acquired. Is that true? > > Certainly with 10.4, you can use a loginLogout plugin to renew your AFS > tokens whenever your Kerberos credentials are renewed. There are a few > AFS loginLogout plugins available - although I didn't have much luck > with the one I tried (when dealing with credential renewal, it seemed to > try to use the credentials that had just expired to get AFS tokens). Could someone summarize the loginLogout plugins that are available? To have these shipped would be great (as Derrick mentioned) Would any of these methods happen to be "blessed" by Apple? Such that their updates wouldn't cause failures with them. Rich > > I keep meaning to look into this further, but haven't got round to it yet. > > Cheers, > > Simon. > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info -- Rich Sudlow University of Notre Dame Center for Research Computing 128 Information Technology Center PO Box 539 Notre Dame, IN 46556-0539 (574) 631-7258 office phone (574) 631-9283 office fax From David.Bear@asu.edu Thu Nov 1 18:20:54 2007 From: David.Bear@asu.edu (David Bear) Date: Thu, 1 Nov 2007 10:20:54 -0700 Subject: [OpenAFS] mac install instructions In-Reply-To: References: <20071101162103.GB3988@asu.edu> Message-ID: <20071101172054.GD3988@asu.edu> On Thu, Nov 01, 2007 at 12:25:58PM -0400, Derrick Brashear wrote: > On 11/1/07, David Bear wrote: > > > > It's been 3+ years since I installed openafs on a mac. The experience > > then was less the possitive. I'm hoping to find an installation > > readme somewhere to go over. I am not finding anything like that on > > openafs.org. do I have to download the openafs for mac before I get > > the readme? > > > um. double-click the installer package in the dmg, optionally edit > /var/db/openafs/etc/ThisCell, reboot, done? thanks. I was worried about other issues like 'integrated logon', ie. will the user need to klog from a term, or is there some kind of pam integration for afs -- or do I need to intall MIT kerb for mac ... etc. Also, last time we use openafs on a mac, finder would hang on occasion when accessing afs volumes so I assumed I had done something wrong on the install. last, since everyone has their own idea about where to put things, I was curious if there was something like transarc-paths or mac-paths or whatever .. I will just assume that need to look in /var/db/openafs/etc/ for configuration items -- also that openafs client with auto-start (when I install openafs on suse it does not auto-generate the start links in /etc/init.d/rcN.d ( or since mac is bsd would it be /usr/local/etc/rc?? ) -- David Bear phone: 602-496-0424 fax: 602-496-0955 College of Public Programs/ASU University Center Rm 622 411 N Central Phoenix, AZ 85007-0685 "Beware the IP portfolio, everyone will be suspect of trespassing" From botsch@cnf.cornell.edu Thu Nov 1 18:35:26 2007 From: botsch@cnf.cornell.edu (Dave Botsch) Date: Thu, 1 Nov 2007 13:35:26 -0400 Subject: [OpenAFS] mac install instructions In-Reply-To: <20071101172054.GD3988@asu.edu> References: <20071101162103.GB3988@asu.edu> <20071101172054.GD3988@asu.edu> Message-ID: <20071101173526.GB12461@puff.cnf.cornell.edu> See below... On Thu, Nov 01, 2007 at 10:20:54AM -0700, David Bear wrote: > On Thu, Nov 01, 2007 at 12:25:58PM -0400, Derrick Brashear wrote: > > On 11/1/07, David Bear wrote: > > > > > thanks. I was worried about other issues like 'integrated logon', ie. > will the user need to klog from a term, or is there some kind of pam > integration for afs -- or do I need to intall MIT kerb for mac ... > etc. Integrated login still needs to be configured separately. Depending on what version of OS X you tried to do it on, it's quite easy now. PAM is only used by ssh, in which case you would need a pam_krb auth module and a pam-openafs-session session module. For gui console logins, you edit /etc/authorization to enable Krb5 authentication (and make sure your kerberos realm is set up in edu.mit.Kerberos). To also get tokens on logging in, there are two main choices. The first is a Kerberos login_logout plugin. The second is to put aklog in the /etc/mach_init_per_user.d For working with AFS while logged in, you want to get two different tools: 1. Finder plugin which adds an AFS context menu for messing with things like ACLs on directories 2. AFSTokens gui app for obtaining/destroying tokens and working with group membership/owned groups. > Also, last time we use openafs on a mac, finder would hang on occasion > when accessing afs volumes so I assumed I had done something wrong on > the install. Finder still sometimes has issues... usually with either a firewall getting in the way or if you are switching ip addresses. > > last, since everyone has their own idea about where to put things, I > was curious if there was something like transarc-paths or mac-paths or > whatever .. I will just assume that need to look in > /var/db/openafs/etc/ for configuration items -- also that openafs > client with auto-start (when I install openafs on suse it does not > auto-generate the start links in /etc/init.d/rcN.d ( or since mac is > bsd would it be /usr/local/etc/rc?? ) > > Well, the openafs package for Mac puts stuff in /var/db/openafs and in /Library/OpenAFS .. the startup script should also go in the right place for Mac and start openafs on boot. Config stuff is /var/db/openafs/etc > -- > David Bear > phone: 602-496-0424 > fax: 602-496-0955 > College of Public Programs/ASU > University Center Rm 622 > 411 N Central > Phoenix, AZ 85007-0685 > "Beware the IP portfolio, everyone will be suspect of trespassing" > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > -- ******************************** David William Botsch Programmer/Analyst CNF Computing botsch@cnf.cornell.edu ******************************** From shadow@gmail.com Thu Nov 1 19:31:21 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 1 Nov 2007 14:31:21 -0400 Subject: [OpenAFS] mac install instructions In-Reply-To: <20071101172054.GD3988@asu.edu> References: <20071101162103.GB3988@asu.edu> <20071101172054.GD3988@asu.edu> Message-ID: ------=_Part_13563_6904227.1193941881653 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On 11/1/07, David Bear wrote: > > On Thu, Nov 01, 2007 at 12:25:58PM -0400, Derrick Brashear wrote: > > On 11/1/07, David Bear wrote: > > > > > > It's been 3+ years since I installed openafs on a mac. The experience > > > then was less the possitive. I'm hoping to find an installation > > > readme somewhere to go over. I am not finding anything like that on > > > openafs.org. do I have to download the openafs for mac before I get > > > the readme? > > > > > > um. double-click the installer package in the dmg, optionally edit > > /var/db/openafs/etc/ThisCell, reboot, done? > > thanks. I was worried about other issues like 'integrated logon', ie. > will the user need to klog from a term, or is there some kind of pam > integration for afs -- or do I need to intall MIT kerb for mac ... > etc. macos comes with kerberos. you may wish to integrate with login, but the current packages don't. if you want to add it, it's yours to add. Also, last time we use openafs on a mac, finder would hang on occasion > when accessing afs volumes so I assumed I had done something wrong on > the install. hang, or just be slow? remember it wants to stat everything. it should behave better now by default last, since everyone has their own idea about where to put things, I > was curious if there was something like transarc-paths or mac-paths or > whatever .. I will just assume that need to look in > /var/db/openafs/etc/ for configuration items -- also that openafs > client with auto-start (when I install openafs on suse it does not > auto-generate the start links in /etc/init.d/rcN.d ( or since mac is > bsd would it be /usr/local/etc/rc?? ) /Library/StartupItems/OpenAFS/OpenAFS is what gets run. It will start by default ------=_Part_13563_6904227.1193941881653 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On 11/1/07, David Bear <David.Bear@asu.edu> wrote:
On Thu, Nov 01, 2007 at 12:25:58PM -0400, Derrick Brashear wrote:
> On 11/1/07, David Bear <David.Bear@asu.edu> wrote:
> >
> > It's been 3+ years since I installed openafs on a mac. The experience
> > then was less the possitive. I'm hoping to find an installation
> > readme somewhere to go over. I am not finding anything like that on
> > openafs.org. do I have to download the openafs for mac before I get
> > the readme?
>
>
> um. double-click the installer package in the dmg, optionally edit
> /var/db/openafs/etc/ThisCell, reboot, done?

thanks. I was worried about other issues like 'integrated logon', ie.
will the user need to klog from a term, or is there some kind of pam
integration for afs -- or do I need to intall MIT kerb for mac ...
etc.

macos comes with kerberos. you may wish to integrate with login, but the current packages don't. if you want to add it, it's yours to add.

Also, last time we use openafs on a mac, finder would hang on occasion
when accessing afs volumes so I assumed I had done something wrong on
the install.

hang, or just be slow? remember it wants to stat everything.

it should behave better now by default

last, since everyone has their own idea about where to put things, I
was curious if there was something like transarc-paths or mac-paths or
whatever .. I will just assume that need to look in
/var/db/openafs/etc/ for configuration items -- also that openafs
client with auto-start (when I install openafs on suse it does not
auto-generate the start links in /etc/init.d/rcN.d ( or since mac is
bsd would it be /usr/local/etc/rc?? )

/Library/StartupItems/OpenAFS/OpenAFS

is what gets run. It will start by default

------=_Part_13563_6904227.1193941881653-- From l.schimmer@cgv.tugraz.at Fri Nov 2 13:45:04 2007 From: l.schimmer@cgv.tugraz.at (Lars Schimmer) Date: Fri, 02 Nov 2007 13:45:04 +0100 Subject: [OpenAFS] problem creating user in windows AD Message-ID: <472B1BD0.7030701@cgv.tugraz.at> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi! Not really OpenAFS problem, more kerberos, but maybe someone has a idea: OpenAFS 1.5.26 on windows XP SP2 kfw 3.2.2 on windows Windows 2003 server and AD on it. Probelm so far: new PC setup, I created a new user in AD and in OpenAFS. While trying to logging in (and create a new profile in OpenAFS filespace) I get only: access denied. Maybe its a bootstrapping problem, after creating a temp profile by windows, user got a token and ticket. But no profile is saved into AFS space. the option in kfw (obtain new credential) is NOT activated and I assume thats the problem. It is not save in profile but needed to access the profile, or? Logging in with a existent user/profile is fine. MfG, Lars Schimmer - -- - ------------------------------------------------------------- TU Graz, Institut f=FCr ComputerGraphik & WissensVisualisierung Tel: +43 316 873-5405 E-Mail: l.schimmer@cgv.tugraz.at Fax: +43 316 873-5402 PGP-Key-ID: 0x4A9B1723 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFHKxvPmWhuE0qbFyMRAsVzAKCPNU9OzJlc4cLOKmafEsODrq6ffACeLHPG 01Hcpfihn0WCSUl85uZtllw=3D =3DHgGl -----END PGP SIGNATURE----- From jaltman@secure-endpoints.com Fri Nov 2 14:08:03 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Fri, 02 Nov 2007 09:08:03 -0400 Subject: [OpenAFS] Re: problem creating user in windows AD In-Reply-To: <472B1BD0.7030701@cgv.tugraz.at> References: <472B1BD0.7030701@cgv.tugraz.at> Message-ID: <472B2133.6050006@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms090206070106020506010808 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Lars Schimmer wrote: > Hi! > > Not really OpenAFS problem, more kerberos, but maybe someone has a idea: > > OpenAFS 1.5.26 on windows XP SP2 > kfw 3.2.2 on windows > > Windows 2003 server and AD on it. > > Probelm so far: new PC setup, I created a new user in AD and in OpenAFS. > While trying to logging in (and create a new profile in OpenAFS > filespace) I get only: access denied. Tokens required to access profiles at logon are obtained by the OpenAFS Integrated Logon. NIM is not involved. > > Maybe its a bootstrapping problem, after creating a temp profile by > windows, user got a token and ticket. But no profile is saved into AFS > space. The profile will not be saved into AFS until logout. Does the user have a token at logout? > the option in kfw (obtain new credential) is NOT activated and I > assume thats the problem. > It is not save in profile but needed to access the profile, or? Profiles containing files or directories whose names include characters that cannot be represented in the OEM character set configured for the machine cannot be saved to AFS. The Windows CIFS client cannot speak to the AFS Client Service's SMB Server using a Unicode aware protocol. > > Logging in with a existent user/profile is fine. > > MfG, > Lars Schimmer ________________________________________________ Kerberos mailing list Kerberos@mit.edu https://mailman.mit.edu/mailman/listinfo/kerberos --------------ms090206070106020506010808 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMDIxMzA4MDNaMCMGCSqGSIb3DQEJBDEWBBSlkP0U PXArL0r73ryirsOOkDElOzBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAGJ919E8abv+vh2l+phZTf9OXG4OH0rOPAjIX3de6b/PFgNHPIvAj 54lKD/fTfQgLODwI9uZnBec+i0+p6Ygj1Qjwwd70z6EUwTOexTusf+k8eZjJujQLlaKKEl3y rulHjzFrqBjiUNtKsVUFI9gbFpY1leWBVxs8cy/U5eevarjkDDHxVzR/KSmCOitRdH6yPFrb I4kG8lYsDRESDHVk6e0fTxgrDjnq3mzA81MFxnqfT5LQAjSSOUPrDndAQKrY4f9ZFfqNg6Tw /nYYg2ApJqQWjwzKc5N1+6Fva+sfAnp8CShPJouXuM0vD2pmNCCVIMfDRuw5S2RW+Qgk0xAg MQAAAAAAAA== --------------ms090206070106020506010808-- From phalenor@gmail.com Fri Nov 2 14:18:35 2007 From: phalenor@gmail.com (Andrew Cobaugh) Date: Fri, 2 Nov 2007 09:18:35 -0400 Subject: [OpenAFS] UFS logging on Solaris - could this cause volumes to go offline and salvager to dump core Message-ID: <1b8d56200711020618h7c2aefebga148b58b76e62d2b@mail.gmail.com> I can't find any reference to UFS logging on Solaris past 2004 in the archives (unless I'm not looking hard enough). Is it safe to turn on logging on ufs under Solaris these days? Reason I ask is I have one or two volumes that will randomly go offline, or fail to vos backup or move. Just yesterday my homedir volume went offline, and subsequent attempts to salvage the vice partition resulted in salvager dumping core. After the 3rd attempt it finally succeeded in reattaching the volume. This seems to occur regardless of what vice partition the volume is on, ruling out disk issues (and my disks are mirrored across separate controllers), and I don't see anything to indicate disk problems to that end. I am running Solaris 10 update 2 with OpenAFS 1.4.4. I wish I could attribute the volume offline-ing and salvager problem to logging (which I will turn off if that turns out to be a problem). I have the core dump and the output from SalvageLog from one run where it dumps core, and the final run where it succeeds here: http://www.phys.psu.edu/~phalenor/salvager_problem/ -- Andy Cobaugh phalenor@[gmail.com|phys.psu.edu|psu.edu] atc135@[bmb].psu.edu andy@michelle.mne.psu.edu From shadow@dementia.org Fri Nov 2 14:50:24 2007 From: shadow@dementia.org (Derrick J Brashear) Date: Fri, 2 Nov 2007 09:50:24 -0400 (EDT) Subject: [OpenAFS] UFS logging on Solaris - could this cause volumes to go offline and salvager to dump core In-Reply-To: <1b8d56200711020618h7c2aefebga148b58b76e62d2b@mail.gmail.com> References: <1b8d56200711020618h7c2aefebga148b58b76e62d2b@mail.gmail.com> Message-ID: On Fri, 2 Nov 2007, Andrew Cobaugh wrote: > I can't find any reference to UFS logging on Solaris past 2004 in the > archives (unless I'm not looking hard enough). Is it safe to turn on > logging on ufs under Solaris these days? Not unless you use namei. And it never will be. Don't use logging ufs with the inode fileserver. Don't bother trying. It can't be fixed; the inode bits will never be extra again. It won't work. From deengert@anl.gov Fri Nov 2 15:25:34 2007 From: deengert@anl.gov (Douglas E. Engert) Date: Fri, 02 Nov 2007 09:25:34 -0500 Subject: [OpenAFS] UFS logging on Solaris - could this cause volumes to go offline and salvager to dump core In-Reply-To: <1b8d56200711020618h7c2aefebga148b58b76e62d2b@mail.gmail.com> References: <1b8d56200711020618h7c2aefebga148b58b76e62d2b@mail.gmail.com> Message-ID: <472B335E.6070608@anl.gov> Andrew Cobaugh wrote: > I can't find any reference to UFS logging on Solaris past 2004 in the > archives (unless I'm not looking hard enough). Is it safe to turn on > logging on ufs under Solaris these days? > Don't thing anything has changed. Logging should be off. We used to use UFS for vice, but have now switched to using the namei with ZFS. A few years ago we ran into problems with fsck, as AFS required its own fsck, because the AFS fileserver dinks around with UFS inodes. But Sun change the superblock for logging and the AFS fsck could not handle it, and did not do check the volume. The long term solution is to use the namei fileservers. > Reason I ask is I have one or two volumes that will randomly go > offline, or fail to vos backup or move. Just yesterday my homedir > volume went offline, and subsequent attempts to salvage the vice > partition resulted in salvager dumping core. After the 3rd attempt it > finally succeeded in reattaching the volume. > > This seems to occur regardless of what vice partition the volume is > on, ruling out disk issues (and my disks are mirrored across separate > controllers), and I don't see anything to indicate disk problems to > that end. > > I am running Solaris 10 update 2 with OpenAFS 1.4.4. I wish I could > attribute the volume offline-ing and salvager problem to logging > (which I will turn off if that turns out to be a problem). > > I have the core dump and the output from SalvageLog from one run where > it dumps core, and the final run where it succeeds here: > > http://www.phys.psu.edu/~phalenor/salvager_problem/ > -- Douglas E. Engert Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439 (630) 252-5444 From l.schimmer@cgv.tugraz.at Fri Nov 2 15:34:57 2007 From: l.schimmer@cgv.tugraz.at (Lars Schimmer) Date: Fri, 02 Nov 2007 15:34:57 +0100 Subject: [OpenAFS] Re: problem creating user in windows AD In-Reply-To: <472B2133.6050006@secure-endpoints.com> References: <472B1BD0.7030701@cgv.tugraz.at> <472B2133.6050006@secure-endpoints.com> Message-ID: <472B3591.2070102@cgv.tugraz.at> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Jeffrey Altman wrote: > Lars Schimmer wrote: >> Hi! >> >> Not really OpenAFS problem, more kerberos, but maybe someone has a ide= a: >> >> OpenAFS 1.5.26 on windows XP SP2 >> kfw 3.2.2 on windows >> >> Windows 2003 server and AD on it. >> >> Probelm so far: new PC setup, I created a new user in AD and in OpenAF= S. >> While trying to logging in (and create a new profile in OpenAFS >> filespace) I get only: access denied. > Tokens required to access profiles at logon are obtained by the OpenAFS > Integrated Logon. NIM is not involved. >> Maybe its a bootstrapping problem, after creating a temp profile by >> windows, user got a token and ticket. But no profile is saved into AFS >> space. > The profile will not be saved into AFS until logout. Does the user hav= e > a token at logout? >> the option in kfw (obtain new credential) is NOT activated and I >> assume thats the problem. >> It is not save in profile but needed to access the profile, or? > Profiles containing files or directories whose names include characters > that cannot be represented in the OEM character set configured for the > machine cannot be saved to AFS. The Windows CIFS client cannot speak t= o > the AFS Client Service's SMB Server using a Unicode aware protocol. >> Logging in with a existent user/profile is fine. >> Ok, thx, found the problem. Typos are really really BAD. MfG, Lars Schimmer - -- - ------------------------------------------------------------- TU Graz, Institut f=FCr ComputerGraphik & WissensVisualisierung Tel: +43 316 873-5405 E-Mail: l.schimmer@cgv.tugraz.at Fax: +43 316 873-5402 PGP-Key-ID: 0x4A9B1723 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFHKzWQmWhuE0qbFyMRAqKmAJ9+AFMzTRFnESpohk41/d0tfQfFXgCbBtEi TYIRHw+fUItvqR6m7oxUiy0=3D =3DU098 -----END PGP SIGNATURE----- From jaltman@secure-endpoints.com Fri Nov 2 15:49:03 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Fri, 02 Nov 2007 10:49:03 -0400 Subject: [OpenAFS] UFS logging on Solaris - could this cause volumes to go offline and salvager to dump core In-Reply-To: <472B335E.6070608@anl.gov> References: <1b8d56200711020618h7c2aefebga148b58b76e62d2b@mail.gmail.com> <472B335E.6070608@anl.gov> Message-ID: <472B38DF.3020001@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms060400080304090001020007 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Douglas E. Engert wrote: > The long term solution is to use the namei fileservers. The long term solution is to use Posix Extended Attributes to store the AFS metadata. A subject for openafs-devel. --------------ms060400080304090001020007 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMDIxNDQ5MDNaMCMGCSqGSIb3DQEJBDEWBBTuY5Gf ohbS3CEfXlDRyferhA2zZzBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEALwkh3V09jvksJAnwgDJuuVO9Lfm+mQufMTzP8rKCeAlq1sEXvDV1 lyGsW/sOhcl0hD3rQI3mXMufzFEPjRuJZiun3KHZlf8E5KaGlTZZap4U6MevKZPhs6+7DbeA WEQAyTrtc+KsL2Hc3F/V+uMIbHLqESMqcuqX6yqxXVt/Ca2GF5WqmjU8opqy/6kMBUUQinGi hlvRUZY746h0cQ0aDiucwg6ZVvIHXUCNtndR9UftuMvfweBl8Le6vJrLGyf9T3YILzQluRo+ Ig0cEhC5gwRnXvDk6vIkSxaAYBBwufM2A73UqqsksTGqAuFlUvsW+aesQ+T3P1uyUqu+AGg1 ZAAAAAAAAA== --------------ms060400080304090001020007-- From matt.smith@uconn.edu Fri Nov 2 20:56:18 2007 From: matt.smith@uconn.edu (Smith, Matt) Date: Fri, 2 Nov 2007 15:56:18 -0400 Subject: [OpenAFS] OpenAFS kernel module Linux/s390 In-Reply-To: References: <1193852486.7100.17.camel@d80h167.uconn.edu> Message-ID: <1194033378.9335.22.camel@d80h167.uconn.edu> --=-7KTE5EPK4Ed7ET6d6e4r Content-Type: multipart/mixed; boundary="=-D51MntE50K7mAWPJyIH5" --=-D51MntE50K7mAWPJyIH5 Content-Type: text/plain Content-Transfer-Encoding: quoted-printable All- I have successfully compiled the OpenAFS module 1.4.5 (final) for Linux s390x, against the Debian Etch (backports) kernel headers package "linux-headers-2.6.22-2-s390x". However, an insmod of the module causes a segmentation fault. uname -a: Linux lnxzvm82 2.6.22-2-s390x #1 SMP Fri Aug 31 00:23:26 UTC 2007 s390x GNU/Linux To compile, I am using the following: #> ./configure --enable-debug-kernel --with-linux-kernel-headers=3D/usr/src/linux-headers-2.6.22-2-s390x #> make only_libafs #> make install_only_libafs=20 Then, I issue the insmod: #> insmod /usr/local/lib/openafs/libafs-2.6.22-2-s390x.mp.ko This generates a segfault. Attached (snip_kern.log) is the relevant snippet from syslog. And, at this point, lsmod shows the module is loaded, and I cannot rmmod it. A full reboot is needed. #> lsmod | grep libafs libafs 5430876 1=20 #>rmmod -f libafs ERROR: Removing 'libafs': Device or resource busy Any suggestions? Is there better debugging info I can supply? Thank you all, -Matt On Wed, 2007-10-31 at 14:07 -0400, Derrick Brashear wrote: >=20 >=20 > On 10/31/07, Smith, Matt wrote: > All- > Can anyone recommend an OpenAFS module version and Linux > kernel > version that are known to work on Linux/s390x? I am using > Debian Etch > on s390x, and have tried both the 2.6.18 and 2.6.22 (from > etch-backports) kernels, with the OpenAFS module 1.4.2-6 > package and > 1.4.5pre2 from source, with no success. > =20 > To avoid excessive debugging info until requested, I'll > simply state > that the above combos fail with references to conflicting > types for > 'lockIdcmp2', 'HandleFlock' and 'lockIdSet', within > afs_vnop_flock.c .=20 > =20 > If it is useful, I can send whatever output is requested. I > just > didn't want to flood the list if it is known that it "just > won't work". >=20 > that should have been fixed in 1.4.4; it hints that configure isn't > working right detecting features in your kernel. >=20 > you might put config.log somewhere, as well as the log messages. you > don't have to mail it. AFS or on the web is fine. >=20 --=20 Matt Smith matt.smith@uconn.edu University Information Technology Services (UITS) University of Connecticut PGP Key ID: 0xE9C5244E --=-D51MntE50K7mAWPJyIH5 Content-Disposition: attachment; filename=snip_kern.log Content-Type: text/x-log; name=snip_kern.log; charset=utf-8 Content-Transfer-Encoding: base64 Tm92ICAyIDE1OjMzOjA0IGxueHp2bTgyIGtlcm5lbDogbGliYWZzOiBtb2R1bGUgbGljZW5zZSAn aHR0cDovL3d3dy5vcGVuYWZzLm9yZy9kbC9saWNlbnNlMTAuaHRtbCcgdGFpbnRzIGtlcm5lbC4N Ck5vdiAgMiAxNTozMzowNCBsbnh6dm04MiBrZXJuZWw6IEZvdW5kIHN5c3RlbSBjYWxsIHRhYmxl IGF0IDB4MjUzNzkwIChwYXR0ZXJuIHNjYW4pDQpOb3YgIDIgMTU6MzM6MDQgbG54enZtODIga2Vy bmVsOiBVbmFibGUgdG8gaGFuZGxlIGtlcm5lbCBwb2ludGVyIGRlcmVmZXJlbmNlIGF0IHZpcnR1 YWwga2VybmVsIGFkZHJlc3MgMDAwMDAwMDAwMDI1MzAwMA0KTm92ICAyIDE1OjMzOjA0IGxueHp2 bTgyIGtlcm5lbDogT29wczogMDAwNCBbIzFdDQpOb3YgIDIgMTU6MzM6MDQgbG54enZtODIga2Vy bmVsOiBNb2R1bGVzIGxpbmtlZCBpbjogbGliYWZzKFApIGxvb3AgcWV0aCBjY3dncm91cCBkbV9t aXJyb3IgZG1fc25hcHNob3QgZG1fbW9kIGRhc2RfZmJhX21vZCBkYXNkX2Vja2RfbW9kIGRhc2Rf bW9kDQpOb3YgIDIgMTU6MzM6MDQgbG54enZtODIga2VybmVsOiBDUFU6ICAgIDAgICAgVGFpbnRl ZDogUCAgICAgIA0KTm92ICAyIDE1OjMzOjA0IGxueHp2bTgyIGtlcm5lbDogUHJvY2VzcyBpbnNt b2QgKHBpZDogMTg0MjAsIHRhc2s6IDAwMDAwMDAwMTc5NGUzNDgsIGtzcDogMDAwMDAwMDAwYjU4 N2ExMCkNCk5vdiAgMiAxNTozMzowNCBsbnh6dm04MiBrZXJuZWw6IEtybmwgUFNXIDogMDcwNDEw MDE4MDAwMDAwMCAwMDAwMDAwMDIxZmJjMTJlIChvc2lfc3lzY2FsbF9pbml0KzB4MTc2LzB4MTk0 IFtsaWJhZnNdKQ0KTm92ICAyIDE1OjMzOjA0IGxueHp2bTgyIGtlcm5lbDogICAgICAgICAgICBS OjAgVDoxIElPOjEgRVg6MSBLZXk6MCBNOjEgVzowIFA6MCBBUzowIENDOjEgUE06MCBFQTozDQpO b3YgIDIgMTU6MzM6MDQgbG54enZtODIga2VybmVsOiBLcm5sIEdQUlM6IDAwMDAwMDAwMDAwMDBm ZDkgMDAwMDAwMDAyMjQ4NjdlMCAwMDAwMDAwMDAwMjUzNzkwIDAwMDAwMDAwMjFmYzVhZGMNCk5v diAgMiAxNTozMzowNCBsbnh6dm04MiBrZXJuZWw6ICAgICAgICAgICAgMDAwMDAwMDAyMWZiYmZl OCAwMDAwMDAwMDAwMDAwMDAwIDAwMDAwMDAwMDAwMDAwMzIgMDAwMDAwMDAyMTBiZTE4OA0KTm92 ICAyIDE1OjMzOjA0IGxueHp2bTgyIGtlcm5lbDogICAgICAgICAgICAwMDAwMDAwMDAwMDAwMDAw IDAwMDAwMDAwMGE0M2Y2NDggMDAwMDAwMDAwYTQzZjAwMCAwMDAwMDAwMDIyNDg2N2Y4DQpOb3Yg IDIgMTU6MzM6MDQgbG54enZtODIga2VybmVsOiAgICAgICAgICAgIDAwMDAwMDAwMjFmNTkwMDAg MDAwMDAwMDAyMWZkMWY0MCAwMDAwMDAwMDIxZmJiZmU4IDAwMDAwMDAwMGI1ODdiZTANCk5vdiAg MiAxNTozMzowNCBsbnh6dm04MiBrZXJuZWw6IEtybmwgQ29kZTogMDAwMDAwMDAyMWZiYzEyMDog ZTMyMDEwMDAwMDA0XklsZ15JJSVyMiwwKCUlcjEpDQpOb3YgIDIgMTU6MzM6MDQgbG54enZtODIg a2VybmVsOiAgICAgICAgICAgIDAwMDAwMDAwMjFmYmMxMjY6IDUwMzAyMzM4XkleSXN0XkklJXIz LDgyNCglJXIyKQ0KTm92ICAyIDE1OjMzOjA0IGxueHp2bTgyIGtlcm5lbDogICAgICAgICAgICAw MDAwMDAwMDIxZmJjMTJhOiBhN2Y0ZmY5N15JXklicmNeSTE1LDEyMWZiYzA1OA0KTm92ICAyIDE1 OjMzOjA0IGxueHp2bTgyIGtlcm5lbDogICAgICAgICAgID4wMDAwMDAwMDIxZmJjMTJlOiA1MDMw MjIyNF5JXklzdF5JJSVyMyw1NDgoJSVyMikNCk5vdiAgMiAxNTozMzowNCBsbnh6dm04MiBrZXJu ZWw6ICAgICAgICAgICAgMDAwMDAwMDAyMWZiYzEzMjogYTdmNGZmYmFeSV5JYnJjXkkxNSwxMjFm YmMwYTYNCk5vdiAgMiAxNTozMzowNCBsbnh6dm04MiBrZXJuZWw6ICAgICAgICAgICAgMDAwMDAw MDAyMWZiYzEzNjogYzAyMDAwMDBkNjgzXklsYXJsXkklJXIyLDIxZmQ2ZTNjDQpOb3YgIDIgMTU6 MzM6MDQgbG54enZtODIga2VybmVsOiAgICAgICAgICAgIDAwMDAwMDAwMjFmYmMxM2M6IGMwZTVm ZmZjZWRjNl5JYnJhc2xeSSUlcjE0LDEyMWY1OWNjOA0KTm92ICAyIDE1OjMzOjA0IGxueHp2bTgy IGtlcm5lbDogICAgICAgICAgICAwMDAwMDAwMDIxZmJjMTQyOiBhNzQ4ZmZmMF5JXklsaGleSSUl cjQsLTE2DQpOb3YgIDIgMTU6MzM6MDQgbG54enZtODIga2VybmVsOiBDYWxsIFRyYWNlOg0KTm92 ICAyIDE1OjMzOjA0IGxueHp2bTgyIGtlcm5lbDogKFs8MDAwMDAwMDAyMWZiYmZlOD5dIG9zaV9z eXNjYWxsX2luaXQrMHgzMC8weDE5NCBbbGliYWZzXSkNCk5vdiAgMiAxNTozMzowNCBsbnh6dm04 MiBrZXJuZWw6ICBbPDAwMDAwMDAwMjA4MTMwMzI+XSBpbml0X21vZHVsZSsweDMyLzB4MTEwIFts aWJhZnNdDQpOb3YgIDIgMTU6MzM6MDQgbG54enZtODIga2VybmVsOiAgWzwwMDAwMDAwMDAwMDYw YjgwPl0gc3lzX2luaXRfbW9kdWxlKzB4MTk4NC8weDFhZTQNCk5vdiAgMiAxNTozMzowNCBsbnh6 dm04MiBrZXJuZWw6ICBbPDAwMDAwMDAwMDAwMjE3NzA+XSBzeXNjX25vZW11KzB4MTAvMHgxNg0K Tm92ICAyIDE1OjMzOjA0IGxueHp2bTgyIGtlcm5lbDogIFs8MDAwMDAwMDA3N2Y2NTFlZT5dIDB4 NzdmNjUxZWUNCg0K --=-D51MntE50K7mAWPJyIH5-- --=-7KTE5EPK4Ed7ET6d6e4r Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQBHK4DiGP63pOnFJE4RAsLxAKCthNO2z4UWlMBnl9Yg0xim5EPemwCePWwv dP7ETOLzz2NnPr5XwjTzHSA= =Zqz1 -----END PGP SIGNATURE----- --=-7KTE5EPK4Ed7ET6d6e4r-- From megacz@hcoop.net Fri Nov 2 20:01:56 2007 From: megacz@hcoop.net (Adam Megacz) Date: Fri, 02 Nov 2007 12:01:56 -0700 Subject: [OpenAFS] where does volserver deposit its core dumps? Message-ID: Hello, Could anybody tell me where volserver leaves its core dumps? (the answer is not /var/lib/openafs/cores/) I have to honestly admit I've never debugged a program via core dumps before. Always used printf() or [last resort] gdb. Unfortunately in my current situation, attaching gdb to any of the volserver pids causes volserver to become unresponsive (yes, even after typing "continue" in gdb). After detaching gdb, volserver remains unresponsive. Kinda frustrating. - a -- PGP/GPG: 5C9F F366 C9CF 2145 E770 B1B8 EFB1 462D A146 C380 From cclausen@acm.org Sat Nov 3 19:40:08 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Sat, 3 Nov 2007 13:40:08 -0500 Subject: [OpenAFS] where does volserver deposit its core dumps? References: Message-ID: <480EA6AC81DE4CC381FC037700ADFFB3@CDCHOME> Adam Megacz wrote: > Could anybody tell me where volserver leaves its core dumps? (the > answer is not /var/lib/openafs/cores/) > > I have to honestly admit I've never debugged a program via core dumps > before. Always used printf() or [last resort] gdb. > > Unfortunately in my current situation, attaching gdb to any of the > volserver pids causes volserver to become unresponsive (yes, even > after typing "continue" in gdb). After detaching gdb, volserver > remains unresponsive. Kinda frustrating. Probably depends on the platform. On sun4x_510, I think cores ended up in /usr/afs/logs < <480EA6AC81DE4CC381FC037700ADFFB3@CDCHOME> Message-ID: "Christopher D. Clausen" writes: >> Could anybody tell me where volserver leaves its core dumps? (the >> answer is not /var/lib/openafs/cores/) > Probably depends on the platform. On sun4x_510, I think cores ended up > in /usr/afs/logs Thanks, but no luck. This is linux-sparc64, by the way. - a -- PGP/GPG: 5C9F F366 C9CF 2145 E770 B1B8 EFB1 462D A146 C380 From megacz@hcoop.net Sun Nov 4 01:35:49 2007 From: megacz@hcoop.net (Adam Megacz) Date: Sat, 03 Nov 2007 17:35:49 -0700 Subject: [OpenAFS] Re: linux sparc64 volserver crashes (w/ gcc 4.1.2) References: <4728A292.7010300@secure-endpoints.com> Message-ID: Jeffrey Altman writes: > Stack trace from core? Got it. Recompiling with gcc-3.4 (rather than debian's 4.1.2) made the problem go away. Go figure. - a ) at volprocs.c:257 #1 0x00016c34 in SAFSVolCreateVolume (acid=0xf7cd3c4c, apart=1534752, aname=0x176a48 "krunk899", atype=1, aparent=-137545022, avolid=0x20005be7, atrans=0xf7cd3e6c) at volprocs.c:376 #2 0x0001f748 in AFSVolExecuteRequest (z_call=0x176d00) at volint.ss.c:24 #3 0x00047fbc in rxi_ServerProc (threadID=1535232, newcall=, socketp=0xf7cd3f5c) at rx.c:1413 #4 0x00040120 in rx_ServerProc () at rx_lwp.c:371 #5 0x0004f5d4 in Create_Process_Part2 () #6 0x0004fce8 in savecontext () #7 0xfcfdff0b in ?? () #8 0xfcfdff0b in ?? () Previous frame identical to this frame (corrupt stack?) (gdb) -- PGP/GPG: 5C9F F366 C9CF 2145 E770 B1B8 EFB1 462D A146 C380 From coy.hile@coyhile.com Sun Nov 4 02:42:49 2007 From: coy.hile@coyhile.com (Coy Hile) Date: Sat, 3 Nov 2007 22:42:49 -0400 (EDT) Subject: [OpenAFS] Re: where does volserver deposit its core dumps? In-Reply-To: References: <480EA6AC81DE4CC381FC037700ADFFB3@CDCHOME> Message-ID: On Sat, 3 Nov 2007, Adam Megacz wrote: > > "Christopher D. Clausen" writes: >>> Could anybody tell me where volserver leaves its core dumps? (the >>> answer is not /var/lib/openafs/cores/) > >> Probably depends on the platform. On sun4x_510, I think cores ended up >> in /usr/afs/logs > > Thanks, but no luck. This is linux-sparc64, by the way. > Does linux have something akin to coreadm(1M) or syscorepath (on AIX)? If it does, then I think the answer would be "where you configured core dumps to be dropped." -- Coy Hile coy.hile@coyhile.com From rra@stanford.edu Sun Nov 4 02:49:32 2007 From: rra@stanford.edu (Russ Allbery) Date: Sat, 03 Nov 2007 19:49:32 -0700 Subject: [OpenAFS] where does volserver deposit its core dumps? In-Reply-To: (Adam Megacz's message of "Fri, 02 Nov 2007 12:01:56 -0700") References: Message-ID: <87d4uqn2rn.fsf@windlord.stanford.edu> Adam Megacz writes: > Could anybody tell me where volserver leaves its core dumps? (the > answer is not /var/lib/openafs/cores/) /var/log/openafs. The cores directory is a misunderstanding in the Debian package and is no longer created by the latest version. It wasn't used by anything. -- Russ Allbery (rra@stanford.edu) From shadow@gmail.com Sun Nov 4 04:32:09 2007 From: shadow@gmail.com (Derrick Brashear) Date: Sun, 4 Nov 2007 00:32:09 -0400 Subject: [OpenAFS] Re: linux sparc64 volserver crashes (w/ gcc 4.1.2) In-Reply-To: References: <4728A292.7010300@secure-endpoints.com> Message-ID: ------=_Part_19642_6621096.1194150729713 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline in my source that's vnode->dataVersion = 1; (on line 257) What's it in yours? And really, don't cc openafs-bugs in mail to the list. Every reply doesn't need its own ticket. On 11/3/07, Adam Megacz wrote: > > > Jeffrey Altman writes: > > Stack trace from core? > > Got it. Recompiling with gcc-3.4 (rather than debian's 4.1.2) made > the problem go away. Go figure. > > - a > > ) at volprocs.c:257 > #1 0x00016c34 in SAFSVolCreateVolume (acid=0xf7cd3c4c, apart=1534752, > aname=0x176a48 "krunk899", atype=1, aparent=-137545022, > avolid=0x20005be7, > atrans=0xf7cd3e6c) at volprocs.c:376 > #2 0x0001f748 in AFSVolExecuteRequest (z_call=0x176d00) at volint.ss.c:24 > #3 0x00047fbc in rxi_ServerProc (threadID=1535232, newcall= optimized out>, > socketp=0xf7cd3f5c) at rx.c:1413 > #4 0x00040120 in rx_ServerProc () at rx_lwp.c:371 > #5 0x0004f5d4 in Create_Process_Part2 () > #6 0x0004fce8 in savecontext () > #7 0xfcfdff0b in ?? () > #8 0xfcfdff0b in ?? () > Previous frame identical to this frame (corrupt stack?) > (gdb) > > > -- > PGP/GPG: 5C9F F366 C9CF 2145 E770 B1B8 EFB1 462D A146 C380 > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > ------=_Part_19642_6621096.1194150729713 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline in my source that's
    vnode->dataVersion = 1;
(on line 257)

What's it in yours?

And really, don't cc openafs-bugs in mail to the list. Every reply doesn't need its own ticket.

On 11/3/07, Adam Megacz <megacz@hcoop.net> wrote:

Jeffrey Altman <jaltman@secure-endpoints.com> writes:
> Stack trace from core?

Got it.  Recompiling with gcc-3.4 (rather than debian's 4.1.2) made
the problem go away.  Go figure.

  - a

) at volprocs.c:257
#1  0x00016c34 in SAFSVolCreateVolume (acid=0xf7cd3c4c, apart=1534752,
    aname=0x176a48 "krunk899", atype=1, aparent=-137545022, avolid=0x20005be7,
    atrans=0xf7cd3e6c) at volprocs.c:376
#2  0x0001f748 in AFSVolExecuteRequest (z_call=0x176d00) at volint.ss.c:24
#3  0x00047fbc in rxi_ServerProc (threadID=1535232, newcall=<value optimized out>,
    socketp=0xf7cd3f5c) at rx.c:1413
#4  0x00040120 in rx_ServerProc () at rx_lwp.c:371
#5  0x0004f5d4 in Create_Process_Part2 ()
#6  0x0004fce8 in savecontext ()
#7  0xfcfdff0b in ?? ()
#8  0xfcfdff0b in ?? ()
Previous frame identical to this frame (corrupt stack?)
(gdb)


--
PGP/GPG: 5C9F F366 C9CF 2145 E770  B1B8 EFB1 462D A146 C380

_______________________________________________
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

------=_Part_19642_6621096.1194150729713-- From megacz@cs.berkeley.edu Sun Nov 4 04:50:36 2007 From: megacz@cs.berkeley.edu (Adam Megacz) Date: Sat, 03 Nov 2007 21:50:36 -0700 Subject: [OpenAFS] Re: linux sparc64 volserver crashes (w/ gcc 4.1.2) References: <4728A292.7010300@secure-endpoints.com> Message-ID: "Derrick Brashear" writes: > in my source that's > vnode->dataVersion = 1; > (on line 257) > > What's it in yours? Same. - a From jdreed@MIT.EDU Mon Nov 5 17:54:30 2007 From: jdreed@MIT.EDU (jdreed@MIT.EDU) Date: Mon, 05 Nov 2007 12:54:30 -0500 Subject: [OpenAFS] RHEL5 kmod packages require explicit kernel version? Message-ID: <20071105125430.yd8q6t7njuzo8osk@webmail.mit.edu> I've been experimenting with the RHEL5 yum repository, and came across the following unfortunatel situation: If you attempt to install openafs-client, and a kmod package does not exist for your kernel version, it will attempt to install a kernel for which it has a supported kmod. This is a fine assumption if your current kernel is older than the one the kmod requires, but not if it's newer. Attempting to install openafs-client on a machine running 2.6.18-8.1.15 fails, since yum won't let you install an older kernel. --- [root@arr-heck jdreed]# yum install openafs-client *snip* ============================================================================= Package Arch Version Repository Size ============================================================================= Installing: openafs-client x86_64 1.4.5-el5.1 openafs 613 k Removing: kernel x86_64 2.6.18-8.el5 installed 72 M Installing for dependencies: kernel x86_64 2.6.18-8.1.14.el5 rhel-x86_64-client-5 14 M kmod-openafs x86_64 1.4.5-1.2.6.18_8.1.14.el5 openafs 270 k *snip* Transaction Check Error: package kernel-2.6.18-8.1.15.el5 (which is newer than kernel-2.6.18-8.1.14.el5) is already installed --- I realize that the workaround is probably to install openafs-kernel-source, which seems to satisfy the requirements of openafs-client, but is it necessary for the kmod packages to require a kernel version? While it's true that the kmod won't work in other kernel versions, since it installs in a separate directory, it seems like it would be better to drop the explicit requirement to avoid failure modes like this one. Presumably there will always be some skew between when Redhat releases a new kernel and when openafs releases the kmod for it, but I can also see the case for enforcing requirements to prevent users from thinking they have installed a working kmod when they haven't. -Jon From dhowells@redhat.com Mon Nov 5 18:07:36 2007 From: dhowells@redhat.com (David Howells) Date: Mon, 05 Nov 2007 18:07:36 +0000 Subject: [OpenAFS] RHEL5 kmod packages require explicit kernel version? In-Reply-To: <20071105125430.yd8q6t7njuzo8osk@webmail.mit.edu> References: <20071105125430.yd8q6t7njuzo8osk@webmail.mit.edu> Message-ID: <31164.1194286056@redhat.com> jdreed@MIT.EDU wrote: > I realize that the workaround is probably to install openafs-kernel-source, > which seems to satisfy the requirements of openafs-client, but is it > necessary for the kmod packages to require a kernel version? While it's > true that the kmod won't work in other kernel versions, since it installs in > a separate directory, it seems like it would be better to drop the explicit > requirement to avoid failure modes like this one. If you don't have the explicit requirement, then you can get the situation where someone does an upgrade and reboots and then their box doesn't work any more because they don't have an AFS module for their newly running kernel and, say, /usr is stored on AFS. David From jdreed@MIT.EDU Mon Nov 5 18:28:58 2007 From: jdreed@MIT.EDU (Jonathan Reed) Date: Mon, 5 Nov 2007 13:28:58 -0500 Subject: [OpenAFS] RHEL5 kmod packages require explicit kernel version? In-Reply-To: <31164.1194286056@redhat.com> References: <20071105125430.yd8q6t7njuzo8osk@webmail.mit.edu> <31164.1194286056@redhat.com> Message-ID: I suppose that's true, but I'd wonder how many sites actually have that setup these days, and of those sites, how many have users who have root on the workstations thus allowing them to shoot themselves in the foot in such a manner. I'd speculate that the number is lower than the number of people who might try to install openafs in the time between when RHEL releases a kernel update and Openafs releases the corresponding kmod. I could be wrong, however. The right answer I suppose would be for yum to install openafs-kernel- source if a kmod package is not available, but I suspect that goes beyond the abilities of yum, at least in its standard configuration. -Jon On Nov 5, 2007, at 1:07 PM, David Howells wrote: > If you don't have the explicit requirement, then you can get the > situation > where someone does an upgrade and reboots and then their box > doesn't work any > more because they don't have an AFS module for their newly running > kernel and, > say, /usr is stored on AFS. > > David From sxw@inf.ed.ac.uk Mon Nov 5 18:36:51 2007 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Mon, 5 Nov 2007 18:36:51 +0000 Subject: [OpenAFS] RHEL5 kmod packages require explicit kernel version? In-Reply-To: References: <20071105125430.yd8q6t7njuzo8osk@webmail.mit.edu> <31164.1194286056@redhat.com> Message-ID: <575F8C86-7EF3-461D-BE21-7C63E530228D@inf.ed.ac.uk> On 5 Nov 2007, at 18:28, Jonathan Reed wrote: > I suppose that's true, but I'd wonder how many sites actually have > that setup these days, and of those sites, how many have users who > have root on the workstations thus allowing them to shoot > themselves in the foot in such a manner. I'd speculate that the > number is lower than the number of people who might try to install > openafs in the time between when RHEL releases a kernel update and > Openafs releases the corresponding kmod. I could be wrong, however. I'd hope to be be able to get into a situation where I can build and make available kernel modules for the current release on all of our 'supported' Fedora and RHEL platforms pretty much immediately upon a new kernel-devel package appearing upstream. This is now pending the availability of suitable build hardware, and some time to finish the required automation. Simon. From Dan Hyde Mon Nov 5 19:11:35 2007 From: Dan Hyde (Dan Hyde) Date: Mon, 05 Nov 2007 14:11:35 -0500 Subject: [OpenAFS] Callback issues Message-ID: <29004.1194289895@block.ifs.umich.edu> We're seeing problems with our (patched) OpenAFS-1.4.2 threaded fileservers. Here's what I know (those who want less detail should skip to "Core files"): Structures: Inside callback.c there are CallBack's and FileEntry's. There used to be a limit of 65535 of shared space for both, but the two are now split out and limited by available memory. There are lists which link these two structs, including the timeout lists, host lists, CallBack lists, FileEntry lists, and free lists. In-use FileEntry's contain single-linked lists for FileEntry's and CallBack's; these contain index values into the FileEntry and CallBack arrays, terminated by a 0. CallBack's contain single-linked lists for CallBack's and FileEntry's, and double-linked lists (including heads of list) for timeouts and host entries; these are index values, with the single-linked lists terminated by a 0, and the double-linked lists terminated by links to themselves. The first thing in a struct CallBack is the link to the next CallBack; the first thing in a struct FileEntry used to be the link to the next FileEntry, but that is no longer true today. The two free lists, CBfree and FEfree, are single-linked lists of CallBack's and FileEntry's using the first word of their struct cast as pointers; they are null-terminated. Removing a CallBack from an in-use list and putting it on the free chain overwrites the link to the next CallBack; this guarentee is no longer true for FileEntry's. Locks: There used to be a CBFElock used when adding and removing CallBack's and FileEntry's from their free chains; this lock is no longer there. There is now a host_glock_mutex that is set with H_LOCK. There is a host-specific lock that used to be set with h_Lock but is now set with h_Lock_r, which does an H_UNLOCK, h_Lock, and H_LOCK -- h_Lock_r will allow any thread blocking on an H_LOCK to get into the critical region protected by the host_glock_mutex. There is also a host-specific hold that "prevents this structure and inferior structures from disappearing". It is not clear if a host lock or hold is needed when adding or removing the host entries of a CallBack. Two routines do not: CleanupTimedOutCallBacks and DeleteFileCallBacks (aka DeleteFile), and the comments for DeleteFileCallBacks are unclear on the issue: "This call doesn't set all of the various host locks, but it shouldn't really matter since we're not adding callbacks, but deleting them. I'm not sure why it doesn't set the lock, however; perhaps it should." Core files: In two of the three core files, in iGetFE, FEfree is bad; it's not a pointer. In the other core file, in CleanupTimedOutCallBacks_r, a timeout chain of CallBack's contains a FileEntry firstcb that leads to another CallBack with a cnext index which is a pointer. In all three cases, we are ignoring the resulting segfault, repeatedly, while holding the H_LOCK global mutex, blocking most other threads. Log files: On two of our fileservers, there are FileLog entries that contain 'Request for unallocated vnode' followed by a pointer. In one FileLog, there are 41 references to 0x40000000 as a vnode, and 1 for 0xffffffff; on another FileLog, there are 16 references to 0x40000000. History: We have access to Transarc AFS as far back as 1995. There have been concerns regarding the manipulation of these lists all along. Next steps: - check the three core files we have to see if the CBfree chain is bad, too. - find a general solution to removing the contents of a KeyFile from the data of a core file so we can share it We hope the above information proves useful. From shadow@gmail.com Mon Nov 5 19:15:21 2007 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 5 Nov 2007 14:15:21 -0500 Subject: [OpenAFS] Callback issues In-Reply-To: <29004.1194289895@block.ifs.umich.edu> References: <29004.1194289895@block.ifs.umich.edu> Message-ID: ------=_Part_23574_18157973.1194290121776 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On 11/5/07, Dan Hyde wrote: > > We're seeing problems with our (patched) OpenAFS-1.4.2 threaded > fileservers. Here's what I know (those who want less detail should > skip to "Core files"): There's an open ticket in RT, 74708, which is this issue. I don't have the answer yet. ------=_Part_23574_18157973.1194290121776 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On 11/5/07, Dan Hyde <drh@umich.edu> wrote:
We're seeing problems with our (patched) OpenAFS-1.4.2 threaded
fileservers.  Here's what I know (those who want less detail should
skip to "Core files"):

There's an open ticket in RT, 74708, which is this issue. I don't have the answer yet.


------=_Part_23574_18157973.1194290121776-- From dlg@dsrw.org Mon Nov 5 21:51:23 2007 From: dlg@dsrw.org (david l goodrich) Date: Mon, 5 Nov 2007 15:51:23 -0600 Subject: [OpenAFS] OpenAFS 1.4.5 on OSX 10.5 Message-ID: <20071105215123.GB5757@aether.dsrw.org> --bp/iNruPH9dso1Pn Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Is anyone else seeing this behavior with OpenAFS on Leopard? Every time I aklog, I get a permission denied, but I still get tokens. Any advice would be great. --david elektra:~ dlg$ unlog elektra:~ dlg$ tokens Tokens held by the Cache Manager: --End of list-- elektra:~ dlg$ aklog aklog: Permission denied so unable to create remote PTS user dlg@dsrw.org in cell dsrw.org (status: 267269). elektra:~ dlg$ tokens Tokens held by the Cache Manager: Tokens for afs@dsrw.org [Expires Dec 5 15:45] --End of list-- elektra:~ dlg$ uname -a Darwin elektra.dsrw.org 9.0.0 Darwin Kernel Version 9.0.0: Tue Oct 9 21:35:55 PDT 2007; root:xnu-1228~1/RELEASE_I386 i386 elektra:~ dlg$ bos version openafs 1.4.5 elektra:~ dlg$=20 --bp/iNruPH9dso1Pn Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHL5BbHDmo5jqnP4QRAg6DAJ9G00WcOGH6TUGyM4OokU0kPi/GyQCeKFpa RYcDi8ae9Zmgu4Ya3mUs0W8= =hTwl -----END PGP SIGNATURE----- --bp/iNruPH9dso1Pn-- From jaltman@secure-endpoints.com Mon Nov 5 22:00:16 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Mon, 05 Nov 2007 17:00:16 -0500 Subject: [OpenAFS] OpenAFS 1.4.5 on OSX 10.5 In-Reply-To: <20071105215123.GB5757@aether.dsrw.org> References: <20071105215123.GB5757@aether.dsrw.org> Message-ID: <472F9270.8060004@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms000008030604030106010508 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit You are not the only one. It is a change in the way the Kerberos libraries on Leopard behave when there is no [domain_realm] mapping specified in the krb5.conf file for the AFS volume server hostnames. Jeffrey Altman david l goodrich wrote: > Is anyone else seeing this behavior with OpenAFS on Leopard? > > Every time I aklog, I get a permission denied, but I still get > tokens. Any advice would be great. > --david > > elektra:~ dlg$ unlog > elektra:~ dlg$ tokens > > Tokens held by the Cache Manager: > > --End of list-- > elektra:~ dlg$ aklog > aklog: Permission denied so unable to create remote PTS user > dlg@dsrw.org in cell dsrw.org (status: 267269). > elektra:~ dlg$ tokens > > Tokens held by the Cache Manager: > > Tokens for afs@dsrw.org [Expires Dec 5 15:45] > --End of list-- > elektra:~ dlg$ uname -a > Darwin elektra.dsrw.org 9.0.0 Darwin Kernel Version 9.0.0: Tue > Oct 9 21:35:55 PDT 2007; root:xnu-1228~1/RELEASE_I386 i386 > elektra:~ dlg$ bos version > openafs 1.4.5 > elektra:~ dlg$ > > > --------------ms000008030604030106010508 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMDUyMjAwMTZaMCMGCSqGSIb3DQEJBDEWBBRApApr iGfC2yga7siw4vNa/gqeMzBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAh9ltk0coC9dTeFOuLiXuh+ws5vejBgwRCnkha8P3FkHoTtiXAYVE pQuRaUnhXPSV3vpDXqnCUdkMdRo1JFQ3NTWIci4llaBqvkye3Hmkd6U2WtEKK5de3fKlkhep tgPnsZJZPDGg6f+QF4v3/C15BU6TJTvqKtsfGiN+IhCgUKv1NGRqs4DaakS9qctDCJY4b6L3 +BIbb7+0dU6pMjFQSkPr3MGPyKupFoe+m/v9hNXoJhh1lK/e69FYW/GsQVi5SBqMBvXjAijD X/DQsUvKjZZ7KtAHh/vB0Bv74susUzbq4YFZV8SL3rvRGfI5CAFrUCV1N5p1y6zaGBYXXncl 3gAAAAAAAA== --------------ms000008030604030106010508-- From dlg@dsrw.org Mon Nov 5 22:30:03 2007 From: dlg@dsrw.org (david l goodrich) Date: Mon, 5 Nov 2007 16:30:03 -0600 Subject: [OpenAFS] OpenAFS 1.4.5 on OSX 10.5 In-Reply-To: <472F9270.8060004@secure-endpoints.com> References: <20071105215123.GB5757@aether.dsrw.org> <472F9270.8060004@secure-endpoints.com> Message-ID: <20071105223003.GC5757@aether.dsrw.org> --wq9mPyueHGvFACwf Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Nov 05, 2007 at 05:00:16PM -0500, Jeffrey Altman wrote: > You are not the only one. It is a change in the way the Kerberos > libraries on Leopard behave when there is no [domain_realm] mapping > specified in the krb5.conf file for the AFS volume server hostnames. Left unsaid here is that adding this: [domain_realm] dsrw.org =3D DSRW.ORG .dsrw.org =3D DSRW.ORG to /Library/Preferences/edu.mit.Kerberos made the problem go away. --david >=20 > Jeffrey Altman >=20 >=20 > david l goodrich wrote: > > Is anyone else seeing this behavior with OpenAFS on Leopard? > >=20 > > Every time I aklog, I get a permission denied, but I still get > > tokens. Any advice would be great. > > --david > >=20 > > elektra:~ dlg$ unlog > > elektra:~ dlg$ tokens > >=20 > > Tokens held by the Cache Manager: > >=20 > > --End of list-- > > elektra:~ dlg$ aklog > > aklog: Permission denied so unable to create remote PTS user > > dlg@dsrw.org in cell dsrw.org (status: 267269). > > elektra:~ dlg$ tokens > >=20 > > Tokens held by the Cache Manager: > >=20 > > Tokens for afs@dsrw.org [Expires Dec 5 15:45] > > --End of list-- > > elektra:~ dlg$ uname -a > > Darwin elektra.dsrw.org 9.0.0 Darwin Kernel Version 9.0.0: Tue > > Oct 9 21:35:55 PDT 2007; root:xnu-1228~1/RELEASE_I386 i386 > > elektra:~ dlg$ bos version > > openafs 1.4.5 > > elektra:~ dlg$=20 > >=20 > >=20 > >=20 --wq9mPyueHGvFACwf Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHL5lrHDmo5jqnP4QRAmdDAJ4lr7vatvN5r5wK8U3nLvT6N2+EggCfcWOq gaQCsoBcB5zTvhq+zRw1v2s= =POwN -----END PGP SIGNATURE----- --wq9mPyueHGvFACwf-- From keith@cs.auckland.ac.nz Tue Nov 6 00:17:08 2007 From: keith@cs.auckland.ac.nz (Keith Johnston) Date: Tue, 6 Nov 2007 13:17:08 +1300 Subject: [OpenAFS] OpenAFS 1.4.5 on OSX 10.5 In-Reply-To: <20071105223003.GC5757@aether.dsrw.org> References: <20071105215123.GB5757@aether.dsrw.org> <472F9270.8060004@secure-endpoints.com> <20071105223003.GC5757@aether.dsrw.org> Message-ID: I have added the domain realm to my edu.mit.Kerberos file but still get the error message and I see that it is using a ID number that is not my UID. But it is still getting me tokens. kjoh001$ aklog -d Authenticating to cell ec.auckland.ac.nz (server afs- db1.ec.auckland.ac.nz). We've deduced that we need to authenticate using referrals. Getting tickets: afs/ec.auckland.ac.nz@ Using Kerberos V5 ticket natively About to resolve name kjoh001@EC.AUCKLAND.AC.NZ to id in cell ec.auckland.ac.nz. Id 32766 doing first-time registration of kjoh001@ec.auckland.ac.nz at ec.auckland.ac.nz aklog: Permission denied so unable to create remote PTS user kjoh001@ec.auckland.ac.nz in cell ec.auckland.ac.nz (status: 267269). Set username to kjoh001@ec.auckland.ac.nz Setting tokens. kjoh001@ec.auckland.ac.nz / @ EC.AUCKLAND.AC.NZ kjoh001$ klist Kerberos 5 ticket cache: 'API:Initial default ccache' Default principal: kjoh001@EC.AUCKLAND.AC.NZ Valid Starting Expires Service Principal 11/06/07 12:25:45 11/06/07 22:25:45 krbtgt/EC.AUCKLAND.AC.NZ@EC.AUCKLAND.AC.NZ 11/06/07 12:25:57 11/06/07 22:25:45 afs/ec.auckland.ac.nz@ kjoh001$ tokens Tokens held by the Cache Manager: Tokens for afs@ec.auckland.ac.nz [Expires Nov 6 22:25] --End of list-- Keith On 6/11/2007, at 11:30 AM, david l goodrich wrote: > On Mon, Nov 05, 2007 at 05:00:16PM -0500, Jeffrey Altman wrote: >> You are not the only one. It is a change in the way the Kerberos >> libraries on Leopard behave when there is no [domain_realm] mapping >> specified in the krb5.conf file for the AFS volume server hostnames. > > Left unsaid here is that adding this: > > [domain_realm] > dsrw.org = DSRW.ORG > .dsrw.org = DSRW.ORG > > to /Library/Preferences/edu.mit.Kerberos made the problem go away. > --david >> >> Jeffrey Altman >> >> >> david l goodrich wrote: >>> Is anyone else seeing this behavior with OpenAFS on Leopard? >>> >>> Every time I aklog, I get a permission denied, but I still get >>> tokens. Any advice would be great. >>> --david >>> >>> elektra:~ dlg$ unlog >>> elektra:~ dlg$ tokens >>> >>> Tokens held by the Cache Manager: >>> >>> --End of list-- >>> elektra:~ dlg$ aklog >>> aklog: Permission denied so unable to create remote PTS user >>> dlg@dsrw.org in cell dsrw.org (status: 267269). >>> elektra:~ dlg$ tokens >>> >>> Tokens held by the Cache Manager: >>> >>> Tokens for afs@dsrw.org [Expires Dec 5 15:45] >>> --End of list-- >>> elektra:~ dlg$ uname -a >>> Darwin elektra.dsrw.org 9.0.0 Darwin Kernel Version 9.0.0: Tue >>> Oct 9 21:35:55 PDT 2007; root:xnu-1228~1/RELEASE_I386 i386 >>> elektra:~ dlg$ bos version >>> openafs 1.4.5 >>> elektra:~ dlg$ >>> >>> >>> > > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Keith Johnston xtn: 87977 Computer Support Computer Science Department Rm 395 This email is brought to you by the letters OS X and the number 10 and 5 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From jaltman@secure-endpoints.com Tue Nov 6 00:31:46 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Mon, 05 Nov 2007 19:31:46 -0500 Subject: [OpenAFS] OpenAFS 1.4.5 on OSX 10.5 In-Reply-To: References: <20071105215123.GB5757@aether.dsrw.org> <472F9270.8060004@secure-endpoints.com> <20071105223003.GC5757@aether.dsrw.org> Message-ID: <472FB5F2.7010308@secure-endpoints.com> Keith Johnston wrote: > I have added the domain realm to my edu.mit.Kerberos file but still get > the error message and I see that it is using a ID number that is not my > UID. But it is still getting me tokens. > > kjoh001$ aklog -d > Authenticating to cell ec.auckland.ac.nz (server > afs-db1.ec.auckland.ac.nz). > We've deduced that we need to authenticate using referrals. > Getting tickets: afs/ec.auckland.ac.nz@ This indicates that there is no domain_realm mapping specified for .ec.auckland.ac.nz in the krb5 configuration file. As a result, the Kerberos v5 library provided a referrals principal name (one without a realm). As a result it cannot determine that your Kerberos v5 principal name should have the realm removed before querying the Protection service. > Using Kerberos V5 ticket natively > About to resolve name kjoh001@EC.AUCKLAND.AC.NZ to id in cell > ec.auckland.ac.nz. > Id 32766 As a result, it gets the anonymous ID number because the name kjoh001@EC.AUCKLAND.AC.NZ does not exist in the database. > doing first-time registration of kjoh001@ec.auckland.ac.nz at > ec.auckland.ac.nz > aklog: Permission denied so unable to create remote PTS user aklog therefore tries to create a PTS entry and fails. > kjoh001@ec.auckland.ac.nz in cell ec.auckland.ac.nz (status: 267269). You can disable the pts registration by using the -noprdb flag. Jeffrey Altman From ronc@depauw.edu Tue Nov 6 04:06:52 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Mon, 05 Nov 2007 23:06:52 -0500 Subject: [OpenAFS] number of files Message-ID: <472FE85C.8010206@depauw.edu> Hello, I have a cluster that mounts OpenAFS volumes. That all seems to work. Things is that I have an MPI app that creates quite a few (a lot actually) files. After I was done and wanted to delete those files I got: -bash: /bin/rm: Argument list too long Nothing wrong AFS wise, but made me wonder. How many files does a directory on an OpenAFS volume "hold" ? Ron -- ================================================================= Ron Croonenberg | | Phone: 1 765 658 4761 Lab Instructor & | Fax: 1 765 658 4732 Technology Coordinator | | Department of Computer Science | e-mail: ronc@DePauw.edu DePauw University | 275 Julian Science & Math Center | 602 South College Ave. | Greencastle, IN 46135 | ================================================================= From rra@stanford.edu Tue Nov 6 04:17:55 2007 From: rra@stanford.edu (Russ Allbery) Date: Mon, 05 Nov 2007 20:17:55 -0800 Subject: [OpenAFS] number of files In-Reply-To: <472FE85C.8010206@depauw.edu> (Ron Croonenberg's message of "Mon, 05 Nov 2007 23:06:52 -0500") References: <472FE85C.8010206@depauw.edu> Message-ID: <87ir4gvwgc.fsf@windlord.stanford.edu> Ron Croonenberg writes: > I have a cluster that mounts OpenAFS volumes. That all seems to work. > Things is that I have an MPI app that creates quite a few (a lot > actually) files. After I was done and wanted to delete those files I > got: -bash: /bin/rm: Argument list too long Yeah, you often have to resort to xargs at that point. > Nothing wrong AFS wise, but made me wonder. How many files does a > directory on an OpenAFS volume "hold" ? 50,000 as long as the file names are short. Longer file names consume two or more directory blocks, hence reducing the total number. -- Russ Allbery (rra@stanford.edu) From warlord@MIT.EDU Tue Nov 6 04:21:10 2007 From: warlord@MIT.EDU (Derek Atkins) Date: Mon, 05 Nov 2007 23:21:10 -0500 Subject: [OpenAFS] number of files In-Reply-To: <472FE85C.8010206@depauw.edu> References: <472FE85C.8010206@depauw.edu> Message-ID: <20071105232110.zphzjrdwjnnoccg4@webmail.mit.edu> Quoting Ron Croonenberg : > Hello, > > > I have a cluster that mounts OpenAFS volumes. That all seems to work. > Things is that I have an MPI app that creates quite a few (a lot > actually) files. After I was done and wanted to delete those files I > got: -bash: /bin/rm: Argument list too long > > Nothing wrong AFS wise, but made me wonder. How many files does a > directory on an OpenAFS volume "hold" ? This has nothing to do with AFS and everything to do with your bash line-length limits. Maybe try something like: rm -ri . > Ron -derek -- Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory Member, MIT Student Information Processing Board (SIPB) URL: http://web.mit.edu/warlord/ PP-ASEL-IA N1NWH warlord@MIT.EDU PGP key available From ronc@depauw.edu Tue Nov 6 05:01:16 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Tue, 06 Nov 2007 00:01:16 -0500 Subject: [OpenAFS] number of files In-Reply-To: <20071105232110.zphzjrdwjnnoccg4@webmail.mit.edu> References: <472FE85C.8010206@depauw.edu> <20071105232110.zphzjrdwjnnoccg4@webmail.mit.edu> Message-ID: <472FF51C.2030604@depauw.edu> I know it has nothing to do with AFS, that's what I said. I was just wondering if there was a max to the number of files in a dir on an AFS volume Derek Atkins wrote: > Quoting Ron Croonenberg : > >> Hello, >> >> >> I have a cluster that mounts OpenAFS volumes. That all seems to work. >> Things is that I have an MPI app that creates quite a few (a lot >> actually) files. After I was done and wanted to delete those files I >> got: -bash: /bin/rm: Argument list too long >> >> Nothing wrong AFS wise, but made me wonder. How many files does a >> directory on an OpenAFS volume "hold" ? > > This has nothing to do with AFS and everything to do with your > bash line-length limits. Maybe try something like: > > rm -ri . > >> Ron > > -derek > From ronc@depauw.edu Tue Nov 6 05:08:26 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Tue, 06 Nov 2007 00:08:26 -0500 Subject: [OpenAFS] number of files In-Reply-To: <87ir4gvwgc.fsf@windlord.stanford.edu> References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> Message-ID: <472FF6CA.6040307@depauw.edu> thanks.! I am@ 13,000 partial data files now. so when I go over that limit, do things break I or do the files just not get created any more? thanks, > 50,000 as long as the file names are short. Longer file names consume two > or more directory blocks, hence reducing the total number. > > From jaltman@secure-endpoints.com Tue Nov 6 05:18:46 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Tue, 06 Nov 2007 00:18:46 -0500 Subject: [OpenAFS] number of files In-Reply-To: <472FF6CA.6040307@depauw.edu> References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> <472FF6CA.6040307@depauw.edu> Message-ID: <472FF936.4050808@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms010403070102070907090303 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Ron Croonenberg wrote: > thanks.! > > I am@ 13,000 partial data files now. > so when I go over that limit, do things break I or do the files just not > get created any more? when there is no room to create more files/directories, you will get a create failure. --------------ms010403070102070907090303 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMDYwNTE4NDZaMCMGCSqGSIb3DQEJBDEWBBQj2J1+ JeKuFqxlLwyk1Ifr6VDwiTBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAVzkwprfvaBy45QXgeoOKTsh7CkYWdDoGW7Ue403W15RuCW2HvFkT cpVP7Lue0LR/NApKEE6RrayKcIF+5/2C+36fGWEub29rU4JBiTiwDJlES8plRX1cRk1PuOJy URcYpxkI0sumwg86nlsLEfP3m8wKJZwSKSHOOhjymU7Q7lqV3pRiuv2Uty6XSqLhaw6OBNOe 2d3dLsrGRLtkZIj/138bxSw0tPXmIJPCMPcoN/BMEuLmBo6qJvlqe0CEeMKA9lNJTX2GZHIJ HbRRLnypnshuKkkRoWPbMf6h55t87oJcLKVotYfOGBTTfANOktz021EF30K7dnzF3yGG9zlM uwAAAAAAAA== --------------ms010403070102070907090303-- From carson@taltos.org Tue Nov 6 11:58:18 2007 From: carson@taltos.org (Carson Gaspar) Date: Tue, 06 Nov 2007 03:58:18 -0800 Subject: [OpenAFS] RHEL5 kmod packages require explicit kernel version? In-Reply-To: <31164.1194286056@redhat.com> References: <20071105125430.yd8q6t7njuzo8osk@webmail.mit.edu> <31164.1194286056@redhat.com> Message-ID: <473056DA.5030204@taltos.org> David Howells wrote: > jdreed@MIT.EDU wrote: > >> I realize that the workaround is probably to install openafs-kernel-source, >> which seems to satisfy the requirements of openafs-client, but is it >> necessary for the kmod packages to require a kernel version? While it's >> true that the kmod won't work in other kernel versions, since it installs in >> a separate directory, it seems like it would be better to drop the explicit >> requirement to avoid failure modes like this one. > > If you don't have the explicit requirement, then you can get the situation > where someone does an upgrade and reboots and then their box doesn't work any > more because they don't have an AFS module for their newly running kernel and, > say, /usr is stored on AFS. Exactly. The correct solution is "maintain a sane repository". Which may, sadly, mean "maintain your own repository"... yum is absolutely correct in refusing to install a version of OpenAFS that won't actually _work_. -- Carson From andrej.filipcic@ijs.si Tue Nov 6 16:56:35 2007 From: andrej.filipcic@ijs.si (Andrej Filipcic) Date: Tue, 6 Nov 2007 17:56:35 +0100 Subject: [OpenAFS] osx 10.5 panic on umount Message-ID: <200711061756.35422.andrej.filipcic@ijs.si> Hi, openafs 1.5.26 for OSX 10.5 panics when stopping the afs service: umount -f /afs Is 1.4.5 a better choice? Cheers, Andrej The generated report: Tue Nov 6 17:46:11 2007 panic(cpu 0 caller 0x27D576D0): msg@/Users/shadow/Source/openafs-devel-1_5_x/src/rx/rx_kcommon.c:139 Latest stack backtrace for cpu 0: Backtrace: 0x0009AD18 0x0009B6BC 0x00029DC4 0x27D576D0 0x27D6F100 0x27CDA004 0x27CEB760 0x27D7ACA8 0x27D8AAA8 0x00105490 0x000F4F10 0x000F537C 0x0030924C 0x000B24C8 0x00000000 Kernel loadable modules in backtrace (with dependencies): org.openafs.filesystems.afs(1.5.26)@0x27cc5000->0x27d9ffff Proceeding back via exception chain: Exception state (sv=0x16fe4a00) PC=0x925FFC8C; MSR=0x0000D030; DAR=0xE0800000; DSISR=0x42000000; LR=0x00002288; R1=0xBFFFF6F0; XCP=0x00000030 (0xC00 - System call) BSD process name corresponding to current thread: umount Mac OS version: 9A581 Kernel version: Darwin Kernel Version 9.0.0: Tue Oct 9 21:37:58 PDT 2007; root:xnu-1228~1/RELEASE_PPC -- _____________________________________________________________ doc. dr. Andrej Filipcic, E-mail: Andrej.Filipcic@ijs.si Department of Experimental High Energy Physics - F9 Jozef Stefan Institute, Jamova 39, P.o.Box 3000 SI-1001 Ljubljana, Slovenia Tel.: +386-1-477-3674 Fax: +386-1-425-7074 ------------------------------------------------------------- From matt@slackers.net Tue Nov 6 17:18:47 2007 From: matt@slackers.net (Matthew Andrews) Date: Tue, 06 Nov 2007 09:18:47 -0800 Subject: [OpenAFS] Re: limit amount of uncommitted cache manager data? In-Reply-To: References: <87ejgo2c53.fsf@windlord.stanford.edu> Message-ID: <4730A1F7.6000309@slackers.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Derrick J Brashear wrote: > On Mon, 24 Sep 2007, Adam Megacz wrote: > >> >> "Derrick Brashear" writes: >>> it would involve a semantic change but we could start flushing >>> changes in >>> the background before fsync. there are of course potential issues. >> >> Wouldn't it be the same semantics as if the client cache were rather >> small? > > largely. but this forces the change for everyone. > >> I can get the app to fsync(), though, so it looks like I'll go with that. > > good. Hi, I looked at this a long time ago, and IIRC the "cache full" behavior was a synchronous one. I.E. cache data was pushed to the server during the write calls, and only enough data was written so that the write in question could succeed. Ideally this activity would occur as a background task starting when the cache hit some high watermark of dirty data, and run until some low watermark had been reached. There is a thread in the cachemanager which handles asynchronous writes when the cachemanager is configured to write after close(I don't remember the option offhand) One of the issues with operating in that mode is that you can have the write fail after the file is closed if for example your tokens expire before all of the dirty data is flushed, but I don't think that would be an issue if you did the background writes while the file was still open, and required everything to be written before the close succeeded. I'm inclined to agree with Adam. I'm not sure this is really a change of semantics for the client. The client does not guarantee that changes will not be sent to the server until a close() or fsync() is called, but rather only that those calls will guarantee that all previously written data is sent to the server. Implementing this is one of the things that's been sitting on my rainy day list for several years now, but somehow there don't seem to be as many rainy days here in California as I used to get back in Pittsburgh ;-) Adam, if you're interested in working on this, feel free to send me an e-mail off list, and we can talk about it more. - -Matt > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > > -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFHMKH3pLF3UzlwZVgRAlhTAJ0dHA+HEU/gTmWoNOLUEhrlAXJE/gCeKREF bn+v2Kpuj6ZjcKSkw/1q8Ps= =C6sT -----END PGP SIGNATURE----- From shadow@gmail.com Tue Nov 6 17:48:45 2007 From: shadow@gmail.com (Derrick Brashear) Date: Tue, 6 Nov 2007 12:48:45 -0500 Subject: [OpenAFS] osx 10.5 panic on umount In-Reply-To: <200711061756.35422.andrej.filipcic@ijs.si> References: <200711061756.35422.andrej.filipcic@ijs.si> Message-ID: On Nov 6, 2007 11:56 AM, Andrej Filipcic wrote: > > Hi, > > openafs 1.5.26 for OSX 10.5 panics when stopping the afs service: > umount -f /afs > > Is 1.4.5 a better choice? Well, it's "recommended". Is that not good enough? ;) http://www.openafs.org/macos.html From pauljohn32@gmail.com Tue Nov 6 22:14:17 2007 From: pauljohn32@gmail.com (Paul Johnson) Date: Tue, 6 Nov 2007 16:14:17 -0600 Subject: [OpenAFS] Please explain RPM packaging of config files like ThisCell and CellServeDB Message-ID: <13e802630711061414k2c1779fdm623012208fafd08a@mail.gmail.com> I have a working Linux system with an openafs client. I'm an RPM builder/user . I want guidance on how users are supposed to protect their site configurations against RPM upgrades of openafs. Basically, I don't want any RPM install to change the configurations. At all. A long time ago, probably 3 or 4 years ago, I installed a new openafs RPM and it replaced my config files (ThisCell, etc). That's no good. So I worked around that by taking the openafs SRPM file, editing these files in the SOURCE directory ThisCell CellServeDB cacheinfo and rebuilding RPMS. When a new kernel would be released, I'd just rebuild to get a new openafs-kernel file. The config I wanted would just get installed. Now, in openafs 1.4.5, I notice some changes/complications. The SRPM does not have ThisCell and cacheinfo, now those are tucked down in the source code for openafs. But there is a file called CellServDB.dist What is that? What's it for? In the install directory, I find CellServDB CellServDB.local CellServDB.dist What do the suffixes mean ".dist" and ".local" for config files? Does OpenAfS read all these when It starts? I also have CellServDB.rpmorig The RPM install moved my configuration out of the way. I don't want that to happen. pj -- Paul E. Johnson Professor, Political Science 1541 Lilac Lane, Room 504 University of Kansas From sxw@inf.ed.ac.uk Tue Nov 6 23:43:25 2007 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Tue, 6 Nov 2007 23:43:25 +0000 Subject: [OpenAFS] Please explain RPM packaging of config files like ThisCell and CellServeDB In-Reply-To: <13e802630711061414k2c1779fdm623012208fafd08a@mail.gmail.com> References: <13e802630711061414k2c1779fdm623012208fafd08a@mail.gmail.com> Message-ID: <5F9479EB-25DA-4ABB-924E-A676793D0DCC@inf.ed.ac.uk> Here's what currently happens with the RPMs from openafs.org (note that there are other OpenAFS builds available, which will have different behaviour). We don't distribute a 'CellServDB' file as such. Instead, the RPM has a script that runs immediately following installation which combines the contents of CellServDB.dist (which comes from the RPM) and CellServDB.local (in which you can make local additions) to produce CellServDB. CellServDB is the only file that is read by OpenAFS itself. 'ThisCell' and 'cacheinfo' are marked as configuration files in the RPM spec. This means that if we don't change them between releases, then any changes you make will remain in place. However, if something in those files changes between releases, your changes will be moved to a '.rpmorig' file, and the ones from the RPM installed. This is probably suboptimal - I'll look at fixing this. If you have a CellServDB.rpmorig, it suggests that you're not using the current OpenAFS RPMs. How recently was that file created? Cheers, Simon. From rns@unimelb.edu.au Wed Nov 7 01:33:10 2007 From: rns@unimelb.edu.au (Robert Sturrock) Date: Wed, 07 Nov 2007 12:33:10 +1100 Subject: [OpenAFS] OpenAFS and Oracle (Support/Certification)? Message-ID: <20071107013310.GE306@unimelb.edu.au> Hi All. This question is perhaps slightly off-topic for this list, but I'm very interested to hear experiences from others about this. We have spent quite a bit of time here setting up some new infrastructure to provide a single-sign-on (Kerberos/LDAP/PAM) and single home directory (OpenAFS) framework. We had found our old method of managing *NIX accounts individually (on our now 200+ Linux and Solaris machines) was not scaling at all well, so the improvement for us in moving to this framework was substantial. OpenAFS was chosen primarily because it seemed very feature rich, mature and had excellent platform support for both servers and desktops. However, a problem has arisen around the use of Oracle. In addition to our stock of general purpose systems (Web, mail etc), we have many machines that run Oracle databases and applications, ranging from systems with a single Oracle DB instance to full Oracle App/DB (RAC) suites. Now apparently Oracle will only "certify" kernel modules that are provided by the O/S vendors directly (for us, this is RedHat and Sun), plus a handful of special modules for (I'm told) certains SANs and the like. Since the use of AFS relies on loading a kernel module, we are concerned that there will be support problems with Oracle. They regard a kernel with an unsupported kernel module as "tainted". Whilst we will endeavour to work through this with Oracle, I was wondering if other sites are running OpenAFS on systems that also run Oracle? If so, have there been any support issues with Oracle, or indeed any issues with stability on Oracle hosts that could be pinned-down specifically to AFS? Has anyone been able to obtain certification from Oracle to run AFS? (Note that we're only planning on using OpenAFS to provide home directories and some non-critical shared areas, not in any way to interact directly with Oracle). We run Redhat Linux 4+5 (32/64 bit), and Solaris x86 and SPARC. One possible workaround would be to the use the AFS/NFS translator, so that the Oracle hosts only need NFS, but I would really only want to do that as a last resort. I'd be interested in hearing any experiences or ideas that people may have on this subject. Regards, Robert. From rra@stanford.edu Wed Nov 7 02:10:52 2007 From: rra@stanford.edu (Russ Allbery) Date: Tue, 06 Nov 2007 18:10:52 -0800 Subject: [OpenAFS] OpenAFS and Oracle (Support/Certification)? In-Reply-To: <20071107013310.GE306@unimelb.edu.au> (Robert Sturrock's message of "Wed, 07 Nov 2007 12:33:10 +1100") References: <20071107013310.GE306@unimelb.edu.au> Message-ID: <87y7darej7.fsf@windlord.stanford.edu> Robert Sturrock writes: > Since the use of AFS relies on loading a kernel module, we are concerned > that there will be support problems with Oracle. They regard a kernel > with an unsupported kernel module as "tainted". Whilst we will > endeavour to work through this with Oracle, I was wondering if other > sites are running OpenAFS on systems that also run Oracle? If so, have > there been any support issues with Oracle, or indeed any issues with > stability on Oracle hosts that could be pinned-down specifically to AFS? Yup, we run AFS on all of our systems, including all of our Oracle database servers, both Linux and Solaris. We haven't had any trouble with stability since Oracle fixed their installer bug that would freak out about the AFS file system (probably more than a decade ago, and that was only an installer bug due to bad code to detect available file systems), and we've not had any trouble with Oracle supporting us. -- Russ Allbery (rra@stanford.edu) From gendalia@gmail.com Wed Nov 7 05:51:10 2007 From: gendalia@gmail.com (Tracy Di Marco White) Date: Tue, 6 Nov 2007 23:51:10 -0600 Subject: [OpenAFS] Automatic move of volumes In-Reply-To: References: <1193225691.11351.34.camel@thor> <471F34AB.4030601@msu.edu> Message-ID: <7024c8c80711062151v3a5bca77h268a44f2ea0d8c0f@mail.gmail.com> On 10/24/07, Steven Jenkins wrote: > On 10/24/07, Derrick Brashear wrote: > ... > > perl scripts exist to do it and I think have been posted here in the past; > > they may even deal with the "RO already exists" case. > > > It would be nice if there were a repository of publically available > contrib stuff like that. http://www.eyrie.org/~eagle/software/ is one of my favorite sources of Russ's software... mvto is quite useful for moving RO volumes around, although I'd already written all my own scripts for moving volumes when I found it. Also, I still use balance. -Tracy From rra@stanford.edu Wed Nov 7 05:56:09 2007 From: rra@stanford.edu (Russ Allbery) Date: Tue, 06 Nov 2007 21:56:09 -0800 Subject: [OpenAFS] Automatic move of volumes In-Reply-To: (Derrick Brashear's message of "Wed, 24 Oct 2007 09:29:02 -0400") References: <1193225691.11351.34.camel@thor> <471F34AB.4030601@msu.edu> Message-ID: <87k5ouppja.fsf@windlord.stanford.edu> "Derrick Brashear" writes: > On 10/24/07, Steven Jenkins wrote: >> Assuming your new location does not already have the ROs on it: >> 1 vos listvol to get the list of volumes to move; if the old location >> is not empty, then, in this order: >> 2 For RO volumes: vos addsite to the new location, vos release, vos >> remove old ROs >> 3 For RW volumes: vos move >> 4 For .backup volumes: vos backup (or backupsys, depending on how you >> made them in the first place) once the RW volumes are moved >> 5 Goto 1 > perl scripts exist to do it and I think have been posted here in the past; > they may even deal with the "RO already exists" case. Yeah, mvto can deal with that case. It still has a few edge cases that it doesn't deal with completely happily, but it aborts rather than attempting them. > the interesting case is where the RW has unreleased changes and you want > to recreate the ROs as they are now. i don't know of distributed tools > to do this. frak and bundle can revert changes while storing them in a side directory and then reapply them later, but that's not really the same thing. It would be nice if vos move could just move ROs. -- Russ Allbery (rra@stanford.edu) From finkej@rpi.edu Wed Nov 7 14:16:46 2007 From: finkej@rpi.edu (Finke, Jon E) Date: Wed, 7 Nov 2007 09:16:46 -0500 Subject: [OpenAFS] OpenAFS and Oracle (Support/Certification)? In-Reply-To: <20071107013310.GE306@unimelb.edu.au> References: <20071107013310.GE306@unimelb.edu.au> Message-ID: <525A866D97B60F4A9DCFD9842EE4790401A1AD98@troy-be-ex2.win.rpi.edu> We have run Oracle on AFS clients for the past 15 years or so without incident (Oracle data files and binaries were all on local disk or SAN volumes, NOT in AFS space). The Oracle instances did not interact directly with AFS files, although many oracle applications did (were stored in AFS space, and/or read and wrote files in AFS space.) =20 We recently upgraded Oracle, and moved it to a non AFS machine - mostly for perceived reduction in dependencies, but mostly that the system build for the new server is NOT afs based (prior machines were maintained using AFS/Package, so AFS was a requirement for system management - organizational evolution moved this Oracle service into a less AFS centric department.) The Oracle database continues to be used to manage all of our computing accounts, including provisioning the AFS accounts automatically. The vast majority of the management system source code (SQL scripts, PL/SQL programs, etc) still lives and is maintained out of AFS space. This has proven to be an interesting exercise for me when I moved to a windows desktop and got to stress test the windows OpenAFS client (it certainly has given me a lot of stress at times...) Jon Finke - Senior Systems Programmer - CMT - RPI 518 276 8185 (voice) - 518 276 2809 (fax) - http://www.rpi.edu/~finkej =20 -----Original Message----- From: openafs-info-admin@openafs.org [mailto:openafs-info-admin@openafs.org] On Behalf Of Robert Sturrock Sent: Tuesday, November 06, 2007 8:33 PM To: openafs-info@openafs.org Subject: [OpenAFS] OpenAFS and Oracle (Support/Certification)? Hi All. This question is perhaps slightly off-topic for this list, but I'm very interested to hear experiences from others about this. We have spent quite a bit of time here setting up some new infrastructure to provide a single-sign-on (Kerberos/LDAP/PAM) and single home directory (OpenAFS) framework. We had found our old method of managing *NIX accounts individually (on our now 200+ Linux and Solaris machines) was not scaling at all well, so the improvement for us in moving to this framework was substantial. OpenAFS was chosen primarily because it seemed very feature rich, mature and had excellent platform support for both servers and desktops. However, a problem has arisen around the use of Oracle. In addition to our stock of general purpose systems (Web, mail etc), we have many machines that run Oracle databases and applications, ranging from systems with a single Oracle DB instance to full Oracle App/DB (RAC) suites. Now apparently Oracle will only "certify" kernel modules that are provided by the O/S vendors directly (for us, this is RedHat and Sun), plus a handful of special modules for (I'm told) certains SANs and the like. Since the use of AFS relies on loading a kernel module, we are concerned that there will be support problems with Oracle. They regard a kernel with an unsupported kernel module as "tainted".=20 Whilst we will endeavour to work through this with Oracle, I was wondering if other sites are running OpenAFS on systems that also run Oracle? If so, have there been any support issues with Oracle, or indeed any issues with stability on Oracle hosts that could be pinned-down specifically to AFS? Has anyone been able to obtain certification from Oracle to run AFS? (Note that we're only planning on using OpenAFS to provide home directories and some non-critical shared areas, not in any way to interact directly with Oracle). We run Redhat Linux 4+5 (32/64 bit), and Solaris x86 and SPARC. One possible workaround would be to the use the AFS/NFS translator, so that the Oracle hosts only need NFS, but I would really only want to do that as a last resort. I'd be interested in hearing any experiences or ideas that people may have on this subject. Regards, Robert. _______________________________________________ OpenAFS-info mailing list OpenAFS-info@openafs.org https://lists.openafs.org/mailman/listinfo/openafs-info From jblaine@kickflop.net Wed Nov 7 18:24:07 2007 From: jblaine@kickflop.net (Jeff Blaine) Date: Wed, 07 Nov 2007 13:24:07 -0500 Subject: [OpenAFS] CVS, GSSAPI, and AFS tokens Message-ID: <473202C7.9010900@kickflop.net> How are people handling krb5 auth with CVS and also getting tokens for gserver connections (GSSAPI/krb5)? From knape@njit.edu Wed Nov 7 18:26:57 2007 From: knape@njit.edu (Dean Knape) Date: Wed, 07 Nov 2007 13:26:57 -0500 Subject: [OpenAFS] installing loopback adapter after sysprep Message-ID: <47320371.1040409@njit.edu> Hello,

I have a sysprep'd W2K3 R2 server VM with OpenAFS 1.5.26.  I've added the necessary GuiRunOnce entry in the sysprep.inf using the instloop.exe extracted from this version's MSI to have the loopback adapter reinstalled.

Instloop does reinstall the loopback adapter but I end up with my previously DHCP enabled local area connection getting a static 10. ... address and my AFS adapter doing DHCP.

Any ideas?


Thanks,
dean
From rra@stanford.edu Wed Nov 7 18:30:42 2007 From: rra@stanford.edu (Russ Allbery) Date: Wed, 07 Nov 2007 10:30:42 -0800 Subject: [OpenAFS] CVS, GSSAPI, and AFS tokens In-Reply-To: <473202C7.9010900@kickflop.net> (Jeff Blaine's message of "Wed, 07 Nov 2007 13:24:07 -0500") References: <473202C7.9010900@kickflop.net> Message-ID: <87r6j1oqlp.fsf@windlord.stanford.edu> Jeff Blaine writes: > How are people handling krb5 auth with CVS and also getting > tokens for gserver connections (GSSAPI/krb5)? CVS's network protocol terrifies me. Where we're still using CVS, we just put the repositories directly in AFS and use AFS ACLs to control access. It's a bit slower, but it works. You have to play some games with the LockDir configuration parameter if you want to provide read-only access via AFS, but it's doable. (We've never needed this.) Subversion is a lot nicer. -- Russ Allbery (rra@stanford.edu) From cclausen@acm.org Wed Nov 7 18:57:57 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Wed, 7 Nov 2007 12:57:57 -0600 Subject: [OpenAFS] installing loopback adapter after sysprep References: <47320371.1040409@njit.edu> Message-ID: <7FA4DB06BE994619833AFECBA850F8D2@CDCHOME> Dean Knape wrote: > I have a sysprep'd W2K3 R2 server VM with OpenAFS 1.5.26. I've added > the necessary GuiRunOnce entry in the sysprep.inf using the > instloop.exe extracted from this version's MSI to have the loopback > adapter reinstalled. > > Instloop does reinstall the loopback adapter but I end up with my > previously DHCP enabled local area connection getting a static 10. > ... address and my AFS adapter doing DHCP. I too have seen this happen. Usually happens only when there are multiple NICs in a machine, either real ethernet ones or FireWire. But since you are using a VM, I bet you do not have firewire. Try this: Before running sysprep on the image, disable the network adapter at the VM level. Shutdown the AFS service. start -> run -> cmd set DEVMGR_SHOW_DETAILS=1 set DEVMGR_SHOW_NONPRESENT_DEVICES=1 %SYSTEMROOT%\system32\devmgmt.msc View -> Show hidden devices then find and delete all NICs, including the loopback adapter This should remove pre-configured network adapters from your syspreped image. Hopefully this will allow newly detected ones to be correctly setup and the loopback adapter install to work as desired. < References: <473202C7.9010900@kickflop.net> Message-ID: <20071107191623.GA23211@citi.umich.edu> Jeff Blaine wrote: How are people handling krb5 auth with CVS and also getting tokens for gserver connections (GSSAPI/krb5)? I assume you're using cvs over ssh? I started putting together a web page that explains how to do this. I never published it because it's not really complete. Anyway, here's what I've written up. Contributions welcome. I should probably put it in the wiki. http://www.citi.umich.edu/u/rees/openbsd/afs-ssh.html From rees@umich.edu Wed Nov 7 19:25:14 2007 From: rees@umich.edu (Jim Rees) Date: Wed, 7 Nov 2007 14:25:14 -0500 Subject: [OpenAFS] CVS, GSSAPI, and AFS tokens In-Reply-To: <20071107191623.GA23211@citi.umich.edu> References: <473202C7.9010900@kickflop.net> <20071107191623.GA23211@citi.umich.edu> Message-ID: <20071107192514.GB23211@citi.umich.edu> Never mind, I missed the part about gserver. I think gserver forwards tickets automatically but you still have to arrange for aklog (or afslog) to run on the server, and I don't know how to do that. It might actually be easier to switch to ssh. From deengert@anl.gov Wed Nov 7 19:26:43 2007 From: deengert@anl.gov (Douglas E. Engert) Date: Wed, 07 Nov 2007 13:26:43 -0600 Subject: [OpenAFS] CVS, GSSAPI, and AFS tokens In-Reply-To: <87r6j1oqlp.fsf@windlord.stanford.edu> References: <473202C7.9010900@kickflop.net> <87r6j1oqlp.fsf@windlord.stanford.edu> Message-ID: <47321173.1070400@anl.gov> Russ Allbery wrote: > Jeff Blaine writes: > >> How are people handling krb5 auth with CVS and also getting >> tokens for gserver connections (GSSAPI/krb5)? > > CVS's network protocol terrifies me. Where we're still using CVS, we just > put the repositories directly in AFS and use AFS ACLs to control access. > It's a bit slower, but it works. We are doing just the opposite, use GSSAPI/Kerberos for authentication and have CVS repositories on local disk. Its a small CVS, so don't read much into this. CVS is on the way out, SVN looks much better. The same executable can function as client or server. It can be started from inetd as a pserver which also responds to a gserver. To do this, add to /etc/services: cvspserver 2401/tcp # CVS remote server, GSSAPI or PW And to inetd.conf something like (all on one line): cvspserver stream tcp nowait root /usr/sbin/in.tcpd /krb5/bin/cvs -f --allow-root=/opt/cvsroot pserver The CVSROOT/passwrd file should be empty to force the use of GSSAPI with kerberos only. The host needs a service principal of cvs/@ When you built cvs, add the --with-gssapi=/path/to/your/krb5 Never looked at gss delegation to get AFS token as the user. You can also do :external with ssh which can use GSSAPI, but this requires the users to have accounts on the server. This might be easier to get AFS token. Even on Windows, WinCVS can use gserver or a PuTTY with gssapi. > > You have to play some games with the LockDir configuration parameter if > you want to provide read-only access via AFS, but it's doable. (We've > never needed this.) > > Subversion is a lot nicer. Yes. > -- Douglas E. Engert Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439 (630) 252-5444 From knape@njit.edu Wed Nov 7 20:13:46 2007 From: knape@njit.edu (Dean Knape) Date: Wed, 07 Nov 2007 15:13:46 -0500 Subject: [OpenAFS] installing loopback adapter after sysprep In-Reply-To: <7FA4DB06BE994619833AFECBA850F8D2@CDCHOME> References: <47320371.1040409@njit.edu> <7FA4DB06BE994619833AFECBA850F8D2@CDCHOME> Message-ID: <47321C7A.80801@njit.edu> Christopher D. Clausen wrote: > Dean Knape wrote: > >> I have a sysprep'd W2K3 R2 server VM with OpenAFS 1.5.26. I've added >> the necessary GuiRunOnce entry in the sysprep.inf using the >> instloop.exe extracted from this version's MSI to have the loopback >> adapter reinstalled. >> >> Instloop does reinstall the loopback adapter but I end up with my >> previously DHCP enabled local area connection getting a static 10. >> ... address and my AFS adapter doing DHCP. >> > > I too have seen this happen. Usually happens only when there are > multiple NICs in a machine, either real ethernet ones or FireWire. But > since you are using a VM, I bet you do not have firewire. > > Try this: > Before running sysprep on the image, disable the network adapter at the > VM level. > > Shutdown the AFS service. > start -> run -> cmd > set DEVMGR_SHOW_DETAILS=1 > set DEVMGR_SHOW_NONPRESENT_DEVICES=1 > %SYSTEMROOT%\system32\devmgmt.msc > > View -> Show hidden devices > then find and delete all NICs, including the loopback adapter > > This should remove pre-configured network adapters from your syspreped > image. Hopefully this will allow newly detected ones to be correctly > setup and the loopback adapter install to work as desired. > > < > That did the trick. There were a number of interfaces I wasn't able to remove but by stopping the service and removing my primary and loopback adapters instloop was able to add the loopback successfully. Thanks, dean From ckarney@sarnoff.com Thu Nov 8 16:51:13 2007 From: ckarney@sarnoff.com (Charles Karney) Date: Thu, 08 Nov 2007 11:51:13 -0500 Subject: [OpenAFS] Building kernel modules for Fedora 7 Message-ID: The 1.4.5 release of OpenAFS includes kernel modules for versions 2.6.22.4_65.fc7 2.6.22.7_85.fc7 However there are two more recent kernel versions 2.6.22.9-91.fc7 2.6.23.1-21.fc7 (1) Is it possible to include modules for these versions (i686) on the 1.4.5 release page. (2) I notice that there's an RPM-build-notes posted with instructions for how to build modules. However it leaves out a discussion of what I find to be the most common case: I've just installed an updated kernel and BEFORE I reboot I would like to build and install the corresponing OpenAFS module. How do I specify a kernel other that the current one with rpmbuild? Also the discussion in RPM-build-notes presupposes a familiarity with the process of building RPM files -- a familiarity which I, for one, don't have. It would be great if RPM-build-notes could be supplemented by a "cookbook recipe" to follow. Something along the lines of Ensure SRPM for openafs is installed, if not rpm -Uhv openafs-1.4.5-fc7.1.src.rpm yum update kernel kernel-devel cd rpmbuild ... openafs.spec rpm -ihv /kmod-openafs-... depmod xxx (?) I recall seeing this in some E-mail on this list a year or so ago. However, it would be better to include the information on the release page. Thanks. -- Charles Karney Sarnoff Corporation, Princeton, NJ 08543-5300 URL: http://charles.karney.info Tel: +1 609 734 2312 Fax: +1 609 734 2662 From knape@njit.edu Thu Nov 8 20:33:03 2007 From: knape@njit.edu (Dean Knape) Date: Thu, 08 Nov 2007 15:33:03 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging Message-ID: <4733727F.2040500@njit.edu> Hi, I have a W2K3 R2 server VM with OpenAFS 1.5.26 provisioned from a sysprep image. When I try to run a program out of AFS from explorer I get "Windows cannot access the specified device, path, or file. You may not have the appropriate permissions to access the item.". Running the same programs from a command line works correctly. ACLs are correct. I get the same error with and without token. Refreshing the AFScache file did not help. Any ideas? Thanks. dean From gwolosh@njit.edu Thu Nov 8 20:47:48 2007 From: gwolosh@njit.edu (Gedaliah Wolosh) Date: Thu, 8 Nov 2007 15:47:48 -0500 (EST) Subject: [OpenAFS] OpenAFS and Oracle (Support/Certification)? In-Reply-To: <20071107013310.GE306@unimelb.edu.au> References: <20071107013310.GE306@unimelb.edu.au> Message-ID: On Wed, the 26th of Cheshvan, 5768 (11/7/2007) Robert Sturrock wrote: > Hi All. > > This question is perhaps slightly off-topic for this list, but I'm very > interested to hear experiences from others about this. > > > > > (Note that we're only planning on using OpenAFS to provide home > directories and some non-critical shared areas, not in any way to > interact directly with Oracle). > > We run Redhat Linux 4+5 (32/64 bit), and Solaris x86 and SPARC. > This is exactly what I do. I have 4 servers with 5 instances over the past 6 years with no problems. Oracle has never asked nor commented when I have requested service. Gedaliah Wolosh University Computing System - IST New Jersey Institute of Technology > One possible workaround would be to the use the AFS/NFS translator, so > that the Oracle hosts only need NFS, but I would really only want to > do that as a last resort. > > I'd be interested in hearing any experiences or ideas that people may > have on this subject. > > Regards, > > Robert. > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > From cclausen@acm.org Thu Nov 8 22:01:47 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Thu, 8 Nov 2007 16:01:47 -0600 Subject: [OpenAFS] trouble running programs out of AFS after imaging References: <4733727F.2040500@njit.edu> Message-ID: Dean Knape wrote: > I have a W2K3 R2 server VM with OpenAFS 1.5.26 provisioned from a > sysprep image. > > When I try to run a program out of AFS from explorer I get "Windows > cannot access the specified device, path, or file. You may not have > the appropriate permissions to access the item.". Running the same > programs from a command line works correctly. ACLs are correct. I > get the same error with and without token. Refreshing the AFScache > file did not help. What application? Could you copy the application in question to local disk? (E.g. did you actually have read access to it?) fs checks & fs checkv & fs flusha and try again. < References: <4733727F.2040500@njit.edu> Message-ID: <47338F36.8010505@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms000905060704030700020307 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Dean Knape wrote: > Hi, > > I have a W2K3 R2 server VM with OpenAFS 1.5.26 provisioned from a > sysprep image. > > When I try to run a program out of AFS from explorer I get "Windows > cannot access the specified device, path, or file. You may not have the > appropriate permissions to access the item.". Running the same programs > from a command line works correctly. ACLs are correct. I get the same > error with and without token. Refreshing the AFScache file did not help. > > Any ideas? Read the release notes and use the SysInternal's tools to figure out which operation is in fact failing. Then file a bug report including a description of how to reproduce the problem. --------------ms000905060704030700020307 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMDgyMjM1MzRaMCMGCSqGSIb3DQEJBDEWBBQgYFFy 6QerCgkwsiVyEbxt3Xwm6DBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAcYRWVAGDp7v4Ufb0XGntCdKZHEuCa7W+DsgQlivEQqqcjRWoj+QG SFdjEN4CDriRrQS6hA8nunwTdM9UP0L0UfHgYLrtiWWibPxNoJqDelwBjj/XPEEnzR8ndDSf BZl6i38lCwOwyV5vZE3hTaPJIfEMWKkHIRqsKXJFYCEGV/2RUFQZ7SZBEVvPsrbEZ+hoovyb vJu7j8O6VXABXJREJstudVozvGamlYLRzGNH40a0VrpUQyMwPNr+KkjowImgK6no2E+9Wf2A 5pswICf7HX9aA3p2Jzbdn7g8rZDcm4l/sIzmb9JKuYwVnblxMx3hbUHZrwF911wXq+BghASJ ywAAAAAAAA== --------------ms000905060704030700020307-- From mgmonza@gmail.com Fri Nov 9 05:26:21 2007 From: mgmonza@gmail.com (MG) Date: Fri, 09 Nov 2007 00:26:21 -0500 Subject: [OpenAFS] Error 11862791 AFS service may not have started Message-ID: <4733EF7D.9000401@gmail.com> Hello, all, I started getting this several months ago: "Error 11862791 AFS service may not have started" now whenever I try to obtain tokens in AFS. This is on a Windows XP, SP2 machine. I don't know what I may have changed to cause this, but our installation was going through so many changes and upgrades and schedule slippages about that time that I was getting seasick just trying to stay on top of things. To try to cure the problem, I just installed 1.5.1800,"typical" configuration, and it's happening in this version too, but it just started in the previous installation after a couple of years of working without problems. I know next to nothing about registries, but looking at this page: http://www.nabble.com/Error-Number-11862791-t2225811.html, it seems that HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Lsa\MSV1_0] "BackConnectionHostNames" should equal "multi-sz"? If so, and if the TYPE in my machine's registery key equals REG_MULTI_SZ but the DATA equals AFS, could this be the problem? Should I change the DATA to equal MULTI_SZ? If anyone knows what else may be causing this, I'd appreciate knowing that too. I've already made sure my folders option for working offline is unset and my tcp/ip settings allow netbios, so don't what else to check. Thanks for any help - I can get along without AFS, but its a lot better with. Chris From jaltman@secure-endpoints.com Fri Nov 9 06:10:53 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Fri, 09 Nov 2007 01:10:53 -0500 Subject: [OpenAFS] Error 11862791 AFS service may not have started In-Reply-To: <4733EF7D.9000401@gmail.com> References: <4733EF7D.9000401@gmail.com> Message-ID: <4733F9ED.6040805@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms070602030307010402080803 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit MG wrote: > Hello, all, > > I started getting this several months ago: "Error 11862791 AFS service > may not have started" now whenever I try to obtain tokens in AFS. This > is on a Windows XP, SP2 machine. I don't know what I may have changed > to cause this, but our installation was going through so many changes > and upgrades and schedule slippages about that time that I was getting > seasick just trying to stay on top of things. > > To try to cure the problem, I just installed 1.5.1800,"typical" > configuration, and it's happening in this version too, but it just > started in the previous installation after a couple of years of working > without problems. The current release is 1.5.27. http://www.openafs.org/windows.html > I know next to nothing about registries, but looking at this page: > http://www.nabble.com/Error-Number-11862791-t2225811.html, it seems that > > HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Lsa\MSV1_0] > "BackConnectionHostNames" > should equal "multi-sz"? > If so, and if the TYPE in my machine's registery key equals REG_MULTI_SZ > but the DATA equals AFS, could this be the problem? Should I change the > DATA to equal MULTI_SZ? If you have any doubts about this key's contents, delete the key and let OpenAFS recreate it. > If anyone knows what else may be causing this, I'd appreciate knowing > that too. I've already made sure my folders option for working offline > is unset and my tcp/ip settings allow netbios, so don't what else to check. What the error indicates is that the clients cannot talk to the server over NetBIOS. There was a bug fixed in 1.5.26 that could have produced random results when trying to generate the Netbios name to be used. It should be "AFS" in all cases if you have the MS Loopback Adapter installed. There is extensive debugging information in the Release Notes. Please read them if you have not already done so. Jeffrey Altman --------------ms070602030307010402080803 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMDkwNjEwNTRaMCMGCSqGSIb3DQEJBDEWBBS2QtoI dyCTR9hE/SHsiCFGSg5SlzBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEApo1T8gZJ0cz1ywRTeYvGqhsQnMA/hNHZG4nI6rJNXGeDuGco7t7K 1lc4wTUARdYZV7xdKYCmma5wZNxz/e7dqODsdqMX1te5aTkktKbFt5tV4FzFV6o3BDuOaaQm k38hVrsA3rIAzySB09RRHtncOW3ay93D2OQPkYb70f0awYuXWMTGAeqhArj9LkbfFu40xG5N Zz6zqUIyuxnsD/BTdYEJZ4lYewLysEcKlKjFHpvvxpHk0Q8nplC0Q2YESlCHk8YeZ2OdhxSt aa50gHjtqRG7EzK5SNjjhyMPz4WGI28gbtQJcO9YeAqdIeMG07A1OtRgNCZwEWK/ECktfyfO kQAAAAAAAA== --------------ms070602030307010402080803-- From mgmonza@gmail.com Fri Nov 9 17:22:56 2007 From: mgmonza@gmail.com (MG) Date: Fri, 09 Nov 2007 12:22:56 -0500 Subject: [OpenAFS] Error 11862791 AFS service may not have started In-Reply-To: <4733F9ED.6040805@secure-endpoints.com> References: <4733EF7D.9000401@gmail.com> <4733F9ED.6040805@secure-endpoints.com> Message-ID: <47349770.7040704@gmail.com> Jeffrey Altman wrote: > MG wrote: > >> I started getting this several months ago: "Error 11862791 AFS service >> may not have started" now whenever I try to obtain tokens in AFS. This >> is on a Windows XP, SP2 machine. >> >> To try to cure the problem, I just installed 1.5.1800,"typical" >> configuration, and it's happening in this version too, but it just >> started in the previous installation after a couple of years of working >> without problems. >> > > The current release is 1.5.27. > > http://www.openafs.org/windows.html > > The one I installed was the one the installation's admins' page had the pointer to. I should add that there is next to no user support here, so I'm doing what I can by searching the Internet. I downloaded and installed 1.5.27 and got the same error. >> I know next to nothing about registries, but looking at this page: >> http://www.nabble.com/Error-Number-11862791-t2225811.html, it seems that >> >> HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Lsa\MSV1_0] >> "BackConnectionHostNames" >> should equal "multi-sz"? >> If so, and if the TYPE in my machine's registery key equals REG_MULTI_SZ >> but the DATA equals AFS, could this be the problem? Should I change the >> DATA to equal MULTI_SZ? >> > > If you have any doubts about this key's contents, delete the key and let > OpenAFS recreate it. > > I did, and the OpenAFS installation did not recreate it. > What the error indicates is that the clients cannot talk to the server > over NetBIOS. There was a bug fixed in 1.5.26 that could have produced > random results when trying to generate the Netbios name to be used. It > should be "AFS" in all cases if you have the MS Loopback Adapter > installed. > > ipconfig /all indicates that AFS is bound to the loopback adapter. The only anomalous setting is that DHCP enabled = "NO". > There is extensive debugging information in the Release Notes. Please > read them if you have not already done so. > > I did not see anything that addresses this persistent error, either in the release notes or in the documentation in general, on http://www.openafs.org/doc Chris From cclausen@acm.org Fri Nov 9 17:34:01 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Fri, 9 Nov 2007 11:34:01 -0600 Subject: [OpenAFS] Error 11862791 AFS service may not have started References: <4733EF7D.9000401@gmail.com> <4733F9ED.6040805@secure-endpoints.com> <47349770.7040704@gmail.com> Message-ID: <17FA51D141A34D4C9543048F007C3163@CDCHOME> MG wrote: > I downloaded and installed 1.5.27 and got the same error. > > ipconfig /all indicates that AFS is bound to the loopback adapter. The > only anomalous setting is that DHCP enabled = "NO". DHCP should not be enabled on the loopback adapter. By default, it has the hardcoded IP address of 10.254.254.253 >> There is extensive debugging information in the Release Notes. Please >> read them if you have not already done so. > > I did not see anything that addresses this persistent error, either in > the release notes or in the documentation in general, on > > http://www.openafs.org/doc Not your error specifically, but there is information in the release notes on how to debug general problems. For instance, what is in the %SYSTEMROOT%\Temp\afsd_init.log file? ----- If you join the #openafs IRC channel on Freenode there are useful people who can help you out. < References: <4733EF7D.9000401@gmail.com> <4733F9ED.6040805@secure-endpoints.com> <47349770.7040704@gmail.com> Message-ID: <47349FE6.7020100@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms060201090302010604020008 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit MG wrote: > I did not see anything that addresses this persistent error, either in > the release notes or in the documentation in general, on See section 4 of the release notes. What do you get when you DIR \\AFS\ALL ? Following the directions in section 4.1 of the release notes, what do you get when you execute the tokens.exe command after setting the IoctlDebug registry value? --------------ms060201090302010604020008 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMDkxNzU5MDJaMCMGCSqGSIb3DQEJBDEWBBQOM9gE kEB3Z1g9bo6/GYwV+KYeLjBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAfMEzSYjEJ/J4TJhGyCOKKdlw6PlOgbfDzdOznbOdmnfdHyPM/xSp juPtN2QZeFvy1hv9ZeyyoOauWiJrNeT3NXyAoDjFfE0O3jE84GPmcRjoi/EcCwuy6z0++NKK a919DZJjSgUZ0X+WkR53iXYC1Bb0t2lQrepvCyFEz0ikuZN5d48TNjon8I/7R+INggkcfdL5 DYcjRLsmlRbuWEJVIw4S7zUq5rxFA0jLiRfG/tX20Q2oGbzQwiMSmwJNWSn1IsxKyCLa1+6d gpZV6/y7Oi338nklM+am+5HEabqUCFQuAkTSs4r1tfqMXDBT/fyVnUWuQYofaOVMbX3IXFAV 5QAAAAAAAA== --------------ms060201090302010604020008-- From jellyfish4756@gmail.com Sun Nov 4 17:45:54 2007 From: jellyfish4756@gmail.com (=?BIG5?B?qKbqTaxJ?=) Date: Mon, 5 Nov 2007 01:45:54 +0800 Subject: [OpenAFS] [help] about crate a new cell In-Reply-To: <1f87420b0711021448v150fea79k3a7e9c44dbbccf61@mail.gmail.com> References: <1f87420b0711021448v150fea79k3a7e9c44dbbccf61@mail.gmail.com> Message-ID: <1f87420b0711040945x1587ddc6u7e5ea20190501857@mail.gmail.com> ------=_Part_22127_17040001.1194198354410 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline [help] about crate a new cell when I crating a new ASF cell , I came into some problems and I can't find discription about this problem show as follows --------------------------------------------------------- [root@root ~]# bos setcellname root.cycu.edu.tw cycu.edu.tw -noauth bos: failed to set cell (you are not authorized for this operation) -------------------------------------------------------- What can I do? ps.my hostname is "root.cycu.edu.tw" cellname is " cycu.edu.tw" centOS 4.5 Thanks for replay ------=_Part_22127_17040001.1194198354410 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline
[help] about crate a new cell

when I crating a new ASF cell , I came into some problems

and I can't find discription about this problem
 
show as follows


---------------------------------------------------------
[root@root ~]#  bos setcellname root.cycu.edu.tw cycu.edu.tw -noauth
bos: failed to set cell (you are not authorized for this operation)

--------------------------------------------------------
What can I do?

ps.my hostname is "root.cycu.edu.tw"
          cellname is   " cycu.edu.tw"
          centOS 4.5
                                Thanks for replay ------=_Part_22127_17040001.1194198354410-- From apaoluzzi@mac.com Tue Nov 6 08:15:18 2007 From: apaoluzzi@mac.com (Alberto Paoluzzi) Date: Tue, 6 Nov 2007 09:15:18 +0100 Subject: [OpenAFS] OpenAFS 1.4.5 on OSX 10.5 In-Reply-To: <20071105215123.GB5757@aether.dsrw.org> References: <20071105215123.GB5757@aether.dsrw.org> Message-ID: <51CF4EB1-0C71-49E2-9484-E739FE5B52A9@mac.com> --Apple-Mail-20--866130329 Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit On 05/nov/07, at 22:51, david l goodrich wrote: > Is anyone else seeing this behavior with OpenAFS on Leopard? Yes. I have the EXACTLY the same behavior with OpenAFS 1.4.5 (Leopard build). I may add that data streaming looks brooken. Music from an Openafs volume to iTunes does not work anymore. --alberto > Every time I aklog, I get a permission denied, but I still get > tokens. Any advice would be great. > --david > > elektra:~ dlg$ unlog > elektra:~ dlg$ tokens > > Tokens held by the Cache Manager: > > --End of list-- > elektra:~ dlg$ aklog > aklog: Permission denied so unable to create remote PTS user > dlg@dsrw.org in cell dsrw.org (status: 267269). > elektra:~ dlg$ tokens > > Tokens held by the Cache Manager: > > Tokens for afs@dsrw.org [Expires Dec 5 15:45] > --End of list-- --Apple-Mail-20--866130329 Content-Type: text/html; charset=US-ASCII Content-Transfer-Encoding: quoted-printable
On 05/nov/07, at 22:51, = david l goodrich wrote:

Is anyone else seeing this = behavior with OpenAFS on Leopard?

Yes. I have the EXACTLY = the same behavior with OpenAFS 1.4.5 (Leopard build). 
I = may add that data streaming looks = brooken. 
Music from an Openafs volume to iTunes = does not work anymore.

--alberto


Every time I aklog, I get a = permission denied, but I still get
tokens.  Any advice would be = great.
 --david

elektra:~ dlg$ unlog
elektra:~ dlg$ = tokens

Tokens held by the Cache Manager:

  --End = of list--
elektra:~ dlg$ aklog
aklog: Permission denied so unable = to create remote PTS user
dlg@dsrw.org in cell dsrw.org (status: = 267269).
elektra:~ dlg$ tokens

Tokens held by the Cache = Manager:

Tokens for afs@dsrw.org [Expires Dec  5 = 15:45]
  --End of = list--

= --Apple-Mail-20--866130329-- From apaoluzzi@mac.com Tue Nov 6 08:51:03 2007 From: apaoluzzi@mac.com (Alberto Paoluzzi) Date: Tue, 6 Nov 2007 09:51:03 +0100 Subject: [OpenAFS] OpenAFS 1.4.5 on OSX 10.5 In-Reply-To: <51CF4EB1-0C71-49E2-9484-E739FE5B52A9@mac.com> References: <20071105215123.GB5757@aether.dsrw.org> <51CF4EB1-0C71-49E2-9484-E739FE5B52A9@mac.com> Message-ID: On 06/nov/07, at 09:15, Alberto Paoluzzi wrote: >> Is anyone else seeing this behavior with OpenAFS on Leopard? > > Yes. I have the EXACTLY the same behavior with OpenAFS 1.4.5 > (Leopard build). > I may add that data streaming looks brooken. > Music from an Openafs volume to iTunes does not work anymore. seems to be OK after installing and configuring mit-kerberos- extras-10.5-b1a.dmg --a From jason@rampaginggeek.com Fri Nov 9 23:07:28 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Fri, 09 Nov 2007 18:07:28 -0500 Subject: [OpenAFS] [help] about crate a new cell In-Reply-To: <1f87420b0711040945x1587ddc6u7e5ea20190501857@mail.gmail.com> References: <1f87420b0711021448v150fea79k3a7e9c44dbbccf61@mail.gmail.com> <1f87420b0711040945x1587ddc6u7e5ea20190501857@mail.gmail.com> Message-ID: <4734E830.1090806@rampaginggeek.com> It looks like to forgot to add the "-noauth" option when starting boserver. Take a look at these instruction: http://www.dementia.org/twiki/bin/view/AFSLore/FedoraAFSInstall The instructions are for Fedora, but should apply to Centos as well. Jason 谷燁施 wrote: > > [help] about crate a new cell > > when I crating a new ASF cell , I came into some problems > > and I can't find discription about this problem > > show as follows > > > --------------------------------------------------------- > [root@root ~]# bos setcellname root.cycu.edu.tw > cycu.edu.tw -noauth > bos: failed to set cell (you are not authorized for this operation) > > -------------------------------------------------------- > What can I do? > > ps.my hostname is "root.cycu.edu.tw " > cellname is " cycu.edu.tw " > centOS 4.5 > Thanks for replay From ronc@depauw.edu Sun Nov 11 00:45:13 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Sat, 10 Nov 2007 19:45:13 -0500 Subject: [OpenAFS] openafs-1.4.5 install In-Reply-To: <87ir4gvwgc.fsf@windlord.stanford.edu> References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> Message-ID: <47365099.302@depauw.edu> Helo all, I am trying to install openafs-1.4.5 (client) on an FC6 machine but is complaining about a dependency.: # rpm -ivh openafs-1.4.5-fc6.1.x86_64.rpm Preparing... ########################################### [100%] 1:openafs ########################################### [100%] # rpm -ivh kmod-openafs-1.4.5-1.2.6.22.5_49.fc6.x86_64.rpmerror: Failed dependencies: openafs-kmod-common >= 1.4.5 is needed by kmod-openafs-1.4.5-1.2.6.22.5_49.fc6.x86_64 I looked in the Fedora 6 section, but didn't see a 'openafs-kmod-common' rpm ? thanks, Ron -- ================================================================= Ron Croonenberg | | Phone: 1 765 658 4761 Lab Instructor & | Fax: 1 765 658 4732 Technology Coordinator | | Department of Computer Science | e-mail: ronc@DePauw.edu DePauw University | 275 Julian Science & Math Center | 602 South College Ave. | Greencastle, IN 46135 | ================================================================= From stephen@physics.unc.edu Sun Nov 11 00:46:44 2007 From: stephen@physics.unc.edu (Stephen Joyce) Date: Sat, 10 Nov 2007 19:46:44 -0500 (EST) Subject: [OpenAFS] Useful visualization tool? Message-ID: If anyone thinks they might find a server/partition visualizer useful, please take a look at http://www.physics.unc.edu/~stephen/afsquotamapper/index.php If you think you might use it, please email me (directly, please don't spam the list) and if there's sufficient interest I'll clean up the code--it's pretty rough at the moment--and release it. I just don't want to go to that trouble if no one is interested. Please be kind as the demo url above does query real servers. Oh, and if anyone has a better tool to do the same thing, feel free to share. Cheers, Stephen -- Stephen Joyce Systems Administrator P A N I C Physics & Astronomy Department Physics & Astronomy University of North Carolina at Chapel Hill Network Infrastructure voice: (919) 962-7214 and Computing fax: (919) 962-0480 http://www.panic.unc.edu Don't judge a book by its movie. From sxw@inf.ed.ac.uk Sun Nov 11 09:40:54 2007 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Sun, 11 Nov 2007 09:40:54 +0000 Subject: [OpenAFS] openafs-1.4.5 install In-Reply-To: <47365099.302@depauw.edu> References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> <47365099.302@depauw.edu> Message-ID: On 11 Nov 2007, at 00:45, Ron Croonenberg wrote: > I looked in the Fedora 6 section, but didn't see a 'openafs-kmod- > common' > rpm ? It's provided by openafs-client - you need to install the userspace for the kernel module to be of any use to you. Cheers, Simon. From ronc@depauw.edu Sun Nov 11 10:03:56 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Sun, 11 Nov 2007 05:03:56 -0500 Subject: [OpenAFS] openafs-1.4.5 install In-Reply-To: References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> <47365099.302@depauw.edu> Message-ID: <4736D38C.8@depauw.edu> On a client machine I used to install: - OpenAFS - kernel module - client. So the order changed ? and I should install, OpenAFS, client, kernel mod in this order ? Ron Simon Wilkinson wrote: > > On 11 Nov 2007, at 00:45, Ron Croonenberg wrote: > >> I looked in the Fedora 6 section, but didn't see a 'openafs-kmod-common' >> rpm ? > > It's provided by openafs-client - you need to install the userspace for > the kernel module to be of any use to you. > > Cheers, > > Simon. > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info -- ================================================================= Ron Croonenberg | | Phone: 1 765 658 4761 Lab Instructor & | Fax: 1 765 658 4732 Technology Coordinator | | Department of Computer Science | e-mail: ronc@DePauw.edu DePauw University | 275 Julian Science & Math Center | 602 South College Ave. | Greencastle, IN 46135 | ================================================================= From sxw@inf.ed.ac.uk Sun Nov 11 12:16:55 2007 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Sun, 11 Nov 2007 12:16:55 +0000 Subject: [OpenAFS] openafs-1.4.5 install In-Reply-To: <4736D38C.8@depauw.edu> References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> <47365099.302@depauw.edu> <4736D38C.8@depauw.edu> Message-ID: <9F2576D6-DEDB-4E9F-A52D-7CBB60A129E3@inf.ed.ac.uk> On 11 Nov 2007, at 10:03, Ron Croonenberg wrote: > On a client machine I used to install: > > - OpenAFS > - kernel module > - client. > > So the order changed ? and I should install, > OpenAFS, client, kernel mod in this order ? You should install them all in a single RPM transaction. The argument here is that openafs-client isn't usable without the kernel module, and the kernel module isn't usable without openafs-client. So, the dependencies won't let you get into a situation where you have one installed, but not the other. rpm -i openafs-1.4.5* openafs-client-1.4.5* kmod-openafs-1.4.5- your.kernel.version-* If you use 'yum', just install the appropriate openafs-repository RPM, and "yum install openafs-client" will take care of all of these dependencies. Simon. From knape@njit.edu Mon Nov 12 15:36:55 2007 From: knape@njit.edu (Dean Knape) Date: Mon, 12 Nov 2007 10:36:55 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: References: <4733727F.2040500@njit.edu> Message-ID: <47387317.3000401@njit.edu> Christopher D. Clausen wrote: > Dean Knape wrote: > >> I have a W2K3 R2 server VM with OpenAFS 1.5.26 provisioned from a >> sysprep image. >> >> When I try to run a program out of AFS from explorer I get "Windows >> cannot access the specified device, path, or file. You may not have >> the appropriate permissions to access the item.". Running the same >> programs from a command line works correctly. ACLs are correct. I >> get the same error with and without token. Refreshing the AFScache >> file did not help. >> > > What application? > > Could you copy the application in question to local disk? (E.g. did you > actually have read access to it?) > > fs checks & fs checkv & fs flusha > and try again. > > < Run -> \\afs\uis\platform\local\platform\Perl\bin\perl.exe -V or \\afs\uis\platform\local\@sys\SSH Secure Shell\SshClient.exe Both fail with error described above. If I copy either of these to local disk they do run. If I run either of these from a command line they do run. fs checks & fs checkv & fs flusha did not help. After debugging and preparing for bug submission I uninstalled 1.5.26 and upgraded to 1.5.27. Still cannot run from explorer. Logs and detailed description sent to openafs-bugs. dean From cclausen@acm.org Mon Nov 12 16:22:40 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Mon, 12 Nov 2007 10:22:40 -0600 Subject: [OpenAFS] trouble running programs out of AFS after imaging References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> Message-ID: <943BC6F428A34876862F0EB1E37821A5@CDCHOME> Dean Knape wrote: > Christopher D. Clausen wrote: >> What application? >> >> Could you copy the application in question to local disk? (E.g. did >> you actually have read access to it?) > > For testing and example, I've installed perl and and SSH client in > AFS. > Start -> Run -> \\afs\uis\platform\local\platform\Perl\bin\perl.exe > -V or \\afs\uis\platform\local\@sys\SSH Secure > Shell\SshClient.exe > Both fail with error described above. > > Logs and detailed description sent to openafs-bugs. What are the ACLs on the files? F:\>fs la \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\ Access list for \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\ is Normal rights: winadmin rlidwk acm.admin rlidwka acm.users rlk system:administrators rlidwka system:anyuser rlk F:\>fs -version OpenAFS1.5.2607 I can run \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\SshClient.exe just fine. \\AFS\acm.uiuc.edu\system\sys\local\util\Perl\bin\perl.exe -V seems to work from start -> run as well. Try using the above paths yourself. < References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> Message-ID: <47388346.5040802@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms010607030204070003090605 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Dean Knape wrote: > For testing and example, I've installed perl and and SSH client in AFS. > > Start -> Run -> \\afs\uis\platform\local\platform\Perl\bin\perl.exe -V > or \\afs\uis\platform\local\@sys\SSH Secure > Shell\SshClient.exe > > Both fail with error described above. Assuming that you sent the entire procmon log to ticket 77066, I see absolutely no errors. > If I copy either of these to local disk they do run. If I run either of > these from a command line they do run. If you look at the procmon log you will see that perl.exe is never executed. At least not by anything that is captured in the log. explorer.exe performs a directory listing, reads the attributes of perl.exe and closes the file. At no point does it attempt to open the file, obtain a lock, or read the contents of the file as would be necessary to execute it. > fs checks & fs checkv & fs flusha did not help. I wouldn't suspect they would since you are not experiencing "Path Not Found" or "Network path unavailable" errors. > After debugging and preparing for bug submission I uninstalled 1.5.26 and > upgraded to 1.5.27. Still cannot run from explorer. Something to note. You are attempting to run a 32-bit exe from 64-bit Server 2003. I wonder if that is a variable. According to your reproduction steps, you need to sysprep the vmware image. Is that really a requirement for reproduction? --------------ms010607030204070003090605 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMTIxNjQ1NThaMCMGCSqGSIb3DQEJBDEWBBSLjRcv knrgU2JPkQbfPLYuj/m37DBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEArIVRi2JACA7+PtlkfNRYyL2bBNsCQVCYB2Uk+QPZ4LNf5mxEfAJA WROsF1XGwdjbspFzX6tnEyyNoqwjfwYv7UvT7P/JsAjkeQGIDkZTixw5Nj69RYHQE6yPJHnT k/lLvoP1hmvrDxMws3Q/1ugtX4ZSakLT45XT/7v5mC7lfgz6GRz1BvlbePZ+vBPVvrGoDxgA 8ic5hhDxCO6yGhGfD++pxjxqDRwsFxpHm8I4qL3mj0VgAfQ0SoZovy1ZkTyPe4ojc9AhJUVk Il1JmMymFwpu03JiUz81axCPQ9xXW20DU4lel3O4KpL2xmFkLt5GjWUt9ss7eVwskDTwlyWA gwAAAAAAAA== --------------ms010607030204070003090605-- From knape@njit.edu Mon Nov 12 17:09:52 2007 From: knape@njit.edu (Dean Knape) Date: Mon, 12 Nov 2007 12:09:52 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: <943BC6F428A34876862F0EB1E37821A5@CDCHOME> References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <943BC6F428A34876862F0EB1E37821A5@CDCHOME> Message-ID: <473888E0.7080703@njit.edu> Christopher D. Clausen wrote: > Dean Knape wrote: > >> Christopher D. Clausen wrote: >> >>> What application? >>> >>> Could you copy the application in question to local disk? (E.g. did >>> you actually have read access to it?) >>> >> For testing and example, I've installed perl and and SSH client in >> AFS. >> Start -> Run -> \\afs\uis\platform\local\platform\Perl\bin\perl.exe >> -V or \\afs\uis\platform\local\@sys\SSH Secure >> Shell\SshClient.exe >> Both fail with error described above. >> >> Logs and detailed description sent to openafs-bugs. >> > > What are the ACLs on the files? > > F:\>fs la \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\ > Access list for > \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\ is > Normal rights: > winadmin rlidwk > acm.admin rlidwka > acm.users rlk > system:administrators rlidwka > system:anyuser rlk > > F:\>fs -version > OpenAFS1.5.2607 > > I can run > \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\SshClient.exe > just fine. > > \\AFS\acm.uiuc.edu\system\sys\local\util\Perl\bin\perl.exe -V seems to > work from start -> run as well. > > Try using the above paths yourself. > > < > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > Z:\platform\local\platform\SSH Secure Shell>fs la Access list for . is Normal rights: system:administrators rlidwka system:anyuser rl Z:\platform\local\platform\SSH Secure Shell>fs -version OpenAFS1.5.2700 I'm getting same behaviour using your SSH client - wont run from explorer but will from command prompt. -dean From knape@njit.edu Mon Nov 12 17:18:31 2007 From: knape@njit.edu (Dean Knape) Date: Mon, 12 Nov 2007 12:18:31 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: <47388346.5040802@secure-endpoints.com> References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <47388346.5040802@secure-endpoints.com> Message-ID: <47388AE7.6010109@njit.edu> Jeffrey Altman wrote: > Dean Knape wrote: > >> For testing and example, I've installed perl and and SSH client in AFS. >> >> Start -> Run -> \\afs\uis\platform\local\platform\Perl\bin\perl.exe -V >> or \\afs\uis\platform\local\@sys\SSH Secure >> Shell\SshClient.exe >> >> Both fail with error described above. >> > > Assuming that you sent the entire procmon log to ticket 77066, I see > absolutely no errors. > That was all of it. > >> If I copy either of these to local disk they do run. If I run either of >> these from a command line they do run. >> > > If you look at the procmon log you will see that perl.exe is never > executed. At least not by anything that is captured in the log. > explorer.exe performs a directory listing, reads the attributes of > perl.exe and closes the file. At no point does it attempt to open the > file, obtain a lock, or read the contents of the file as would be > necessary to execute it. > > >> fs checks & fs checkv & fs flusha did not help. >> > > I wouldn't suspect they would since you are not experiencing "Path Not > Found" or "Network path unavailable" errors. > > >> After debugging and preparing for bug submission I uninstalled 1.5.26 and >> upgraded to 1.5.27. Still cannot run from explorer. >> > > Something to note. You are attempting to run a 32-bit exe from 64-bit > Server 2003. I wonder if that is a variable. > I was thinking same but everything runs fine on my base image. It's only after a sysprep that things fall apart. > According to your reproduction steps, you need to sysprep the vmware > image. Is that really a requirement for reproduction? > This avoids the duplicate SID and AFS UUID problem. -dean From cclausen@acm.org Mon Nov 12 17:20:18 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Mon, 12 Nov 2007 11:20:18 -0600 Subject: [OpenAFS] trouble running programs out of AFS after imaging References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <47388346.5040802@secure-endpoints.com> <47388AE7.6010109@njit.edu> Message-ID: Dean Knape wrote: > Jeffrey Altman wrote: >> Something to note. You are attempting to run a 32-bit exe from >> 64-bit Server 2003. I wonder if that is a variable. >> > I was thinking same but everything runs fine on my base image. It's > only after a sysprep that > things fall apart. Hmm... Did you completely DELETE the %TEMP%\AFSCache file before cloning the system? This is specifically mentioned in the release notes and weird things happen if you have multiple systems based off of the same image. The AFS uuid is stored in the cache file and multiple machines can look like the same client if these are not unique. >> According to your reproduction steps, you need to sysprep the vmware >> image. Is that really a requirement for reproduction? > > This avoids the duplicate SID and AFS UUID problem. Sysprep by itself does not take care of duplicate UUIDs. You need to delete the AFSCache file. Or run fs uuid -generate < References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <47388346.5040802@secure-endpoints.com> <47388AE7.6010109@njit.edu> Message-ID: <47388D60.6080103@njit.edu> Christopher D. Clausen wrote:
Dean Knape <knape@njit.edu> wrote:
  
Jeffrey Altman wrote:
    
Something to note.  You are attempting to run a 32-bit exe from
64-bit Server 2003.  I wonder if that is a variable.

      
I was thinking same but everything runs fine on my base image.  It's
only after a sysprep that
things fall apart.
    

Hmm...  Did you completely DELETE the %TEMP%\AFSCache file before 
cloning the system?  This is specifically mentioned in the release notes 
and weird things happen if you have multiple systems based off of the 
same image.  The AFS uuid is stored in the cache file and multiple 
machines can look like the same client if these are not unique.
  

No and yes.  According to section 3.38 of release notes, if SID is regenerated by sysprep then
there is no need to delete the file.  However, I did eventually delete the cache file as I was
troubleshooting.

  
According to your reproduction steps, you need to sysprep the vmware
image.  Is that really a requirement for reproduction?
      
This avoids the duplicate SID and AFS UUID problem.
    

Sysprep by itself does not take care of duplicate UUIDs.  You need to 
delete the AFSCache file.

Or run fs uuid -generate

<<CDC 

  

-dean
From cclausen@acm.org Mon Nov 12 17:30:33 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Mon, 12 Nov 2007 11:30:33 -0600 Subject: [OpenAFS] trouble running programs out of AFS after imaging References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <47388346.5040802@secure-endpoints.com> <47388AE7.6010109@njit.edu> <47388D60.6080103@njit.edu> Message-ID: <7698C8A5748D4F3EB43F3230C506BFF4@CDCHOME> Dean Knape wrote: > No and yes. According to section 3.38 of release notes, if SID is > regenerated by sysprep then there is no need to delete the file. > However, I did eventually delete the cache file as I was > troubleshooting. Well, this is eay to check. Just run fs uuid from each system and compare the UUIDs. < References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <943BC6F428A34876862F0EB1E37821A5@CDCHOME> Message-ID: <47388FE0.4000900@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms070208080200070008010105 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Christopher D. Clausen wrote: > What are the ACLs on the files? > > F:\>fs la \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\ > Access list for > \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\ is > Normal rights: > winadmin rlidwk > acm.admin rlidwka > acm.users rlk > system:administrators rlidwka > system:anyuser rlk Checking ACLs occurred to me as well but I wanted to make sure that there was a lock request failing first. There isn't a lock request being made that could fail according to the logs. > F:\>fs -version > OpenAFS1.5.2607 > > I can run > \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\SshClient.exe > just fine. > > \\AFS\acm.uiuc.edu\system\sys\local\util\Perl\bin\perl.exe -V seems to > work from start -> run as well. Note that these paths won't work on 64-bit OpenAFS because of the sysname: Current sysname list is 'amd64_win64' 'x86_win32' 'i386_w2k' The 32-bit sysname list is: Current sysname list is 'x86_win32' 'i386_w2k' 'i386_nt40' In the above path "sys" is a symlink to "@sys". Jeffrey Altman --------------ms070208080200070008010105 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMTIxNzM5NDRaMCMGCSqGSIb3DQEJBDEWBBTK+yvK dx61q5VsRrOdohkzkz+jpzBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAI2iCXTk9qPRzyiAbLFvzLVXhDC2DR7tQqqZBeG9p2mgEt+u4lXZx BZrTu9JkPM/DhAFI86aCAie56Wc46NC8oP6xPLto4l7sNWvKehHlvfM+4pplbMEmCJBMrzkx 6NfPwHJNA4FGkIq0Pvgl6vt0g/G/i9JWe++cgBnXtXmazSkcCiL8XkK3qnknmQlebessYYxJ OpeL0SIHwD+fdgbykBmwfANK3g+RUE1XYrI8/pRjyTZmyaXAFuvfUn7M9WbTXy5acrYfv3En qkVtU1vhkr09b2PJcQohM70yXVCRzCTvBSBcrmrdE5oE/pLFAZtw23pRxbdYW1kRv9S9KTDN SgAAAAAAAA== --------------ms070208080200070008010105-- From jaltman@secure-endpoints.com Mon Nov 12 17:42:37 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Mon, 12 Nov 2007 12:42:37 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: <47388AE7.6010109@njit.edu> References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <47388346.5040802@secure-endpoints.com> <47388AE7.6010109@njit.edu> Message-ID: <4738908D.1050304@secure-endpoints.com> Dean Knape wrote: > This avoids the duplicate SID and AFS UUID problem. You need to generate a new Windows SID or you are going to have problems with Windows. This has nothing to do with the AFS issue. For AFS, if the Windows SID changes, it will generate a new UUID. A new UUID is also generated if you delete the AFSCache file or execute "fs uuid -generate". Jeffrey Altman From jaltman@secure-endpoints.com Mon Nov 12 17:44:01 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Mon, 12 Nov 2007 12:44:01 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: <7698C8A5748D4F3EB43F3230C506BFF4@CDCHOME> References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <47388346.5040802@secure-endpoints.com> <47388AE7.6010109@njit.edu> <47388D60.6080103@njit.edu> <7698C8A5748D4F3EB43F3230C506BFF4@CDCHOME> Message-ID: <473890E1.7060204@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms080202040008030405020500 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Christopher D. Clausen wrote: > Dean Knape wrote: >> No and yes. According to section 3.38 of release notes, if SID is >> regenerated by sysprep then there is no need to delete the file. >> However, I did eventually delete the cache file as I was >> troubleshooting. > > Well, this is eay to check. Just run fs uuid from each system and > compare the UUIDs. UUIDs are not the issue here. explorer.exe is not opening the file to execute. That has nothing to do with the UUID. --------------ms080202040008030405020500 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMTIxNzQ0MDFaMCMGCSqGSIb3DQEJBDEWBBRQ7f9k Iby4ld1lWIGRo9vEmmd78DBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAIvqBCOy8eq7O1GXhdBLxsKvo9uZ/P+qUFHeVZj9KVveMhuxXPEpq cb/SLOcuoY0ApCAzr2WNSpeqAv5qJJNtbtuXXCuBC2pJeCf9DnYm+Kdk7goZoTJSZoHGDuVp s3VcuIg4YDKZwFqMwzZanaV4cwJ/hMBAwgCU1h8am5x/bTkAbyGAJxdUEGETYwbykQ2911Yk s/Sw2+bkm7NgOn9+z5NeFlP8QYmFfL09DQPEKq4mfqih9YSaH0XoY6Hg/eHw5aFvbIOFgPDE ToD4oCgoT1AUlB9jNEbdkPYVi5p/b/MLzvDA6cagRZitPHjZFayUA6hWGPeVgfOJeT0jQuJE 3wAAAAAAAA== --------------ms080202040008030405020500-- From knape@njit.edu Mon Nov 12 17:43:49 2007 From: knape@njit.edu (Dean Knape) Date: Mon, 12 Nov 2007 12:43:49 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: <47388FE0.4000900@secure-endpoints.com> References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <943BC6F428A34876862F0EB1E37821A5@CDCHOME> <47388FE0.4000900@secure-endpoints.com> Message-ID: <473890D5.2090902@njit.edu> Jeffrey Altman wrote: > Christopher D. Clausen wrote: > > >> What are the ACLs on the files? >> >> F:\>fs la \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\ >> Access list for >> \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\ is >> Normal rights: >> winadmin rlidwk >> acm.admin rlidwka >> acm.users rlk >> system:administrators rlidwka >> system:anyuser rlk >> > > Checking ACLs occurred to me as well but I wanted to make sure that > there was a lock request failing first. There isn't a lock request > being made that could fail according to the logs. > > >> F:\>fs -version >> OpenAFS1.5.2607 >> >> I can run >> \\AFS\acm.uiuc.edu\system\sys\local\util\network\SSH.com\SshClient.exe >> just fine. >> >> \\AFS\acm.uiuc.edu\system\sys\local\util\Perl\bin\perl.exe -V seems to >> work from start -> run as well. >> > > Note that these paths won't work on 64-bit OpenAFS because of the sysname: > > Current sysname list is 'amd64_win64' 'x86_win32' 'i386_w2k' > > The 32-bit sysname list is: > > Current sysname list is 'x86_win32' 'i386_w2k' 'i386_nt40' > > In the above path "sys" is a symlink to "@sys". > > Jeffrey Altman > For me to test the acm.uiuc.edu path I set fs sysname i386_nt40. -dean From jaltman@secure-endpoints.com Mon Nov 12 18:19:14 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Mon, 12 Nov 2007 13:19:14 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: <473890D5.2090902@njit.edu> References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <943BC6F428A34876862F0EB1E37821A5@CDCHOME> <47388FE0.4000900@secure-endpoints.com> <473890D5.2090902@njit.edu> Message-ID: <47389922.3030509@secure-endpoints.com> Dean Knape wrote: > For me to test the acm.uiuc.edu path I set fs sysname i386_nt40. > -dean I rewrote the path as: \\AFS\acm.uiuc.edu\system\i386_nt40\local\util\Perl\bin\perl.exe -V and it works just fine from my 64-bit XP and 64-bit 2003 Server system. Obviously some part of your procedure is altering the behavior of the Explorer shell. I do not believe that this is a bug in OpenAFS as it works with 64-bit 2003 Server prior to performing the cloning. I think you need to compare the systems before and after the sysprep and determine what has been changed. Jeffrey Altman From knape@njit.edu Mon Nov 12 18:58:15 2007 From: knape@njit.edu (Dean Knape) Date: Mon, 12 Nov 2007 13:58:15 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: <473890E1.7060204@secure-endpoints.com> References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <47388346.5040802@secure-endpoints.com> <47388AE7.6010109@njit.edu> <47388D60.6080103@njit.edu> <7698C8A5748D4F3EB43F3230C506BFF4@CDCHOME> <473890E1.7060204@secure-endpoints.com> Message-ID: <4738A247.4040108@njit.edu> Jeffrey Altman wrote: > Christopher D. Clausen wrote: > >> Dean Knape wrote: >> >>> No and yes. According to section 3.38 of release notes, if SID is >>> regenerated by sysprep then there is no need to delete the file. >>> However, I did eventually delete the cache file as I was >>> troubleshooting. >>> >> Well, this is eay to check. Just run fs uuid from each system and >> compare the UUIDs. >> > > UUIDs are not the issue here. explorer.exe is not opening the file to > execute. That has nothing to do with the UUID. > For giggles and future reference, I just created two new VMs from my image and have confirmed that they do get a unique SID and UUID. -dean From knape@njit.edu Mon Nov 12 19:49:19 2007 From: knape@njit.edu (Dean Knape) Date: Mon, 12 Nov 2007 14:49:19 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: <47389922.3030509@secure-endpoints.com> References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <943BC6F428A34876862F0EB1E37821A5@CDCHOME> <47388FE0.4000900@secure-endpoints.com> <473890D5.2090902@njit.edu> <47389922.3030509@secure-endpoints.com> Message-ID: <4738AE3F.6030107@njit.edu> Jeffrey Altman wrote: > Dean Knape wrote: > > >> For me to test the acm.uiuc.edu path I set fs sysname i386_nt40. >> -dean >> > > I rewrote the path as: > > \\AFS\acm.uiuc.edu\system\i386_nt40\local\util\Perl\bin\perl.exe -V > > > and it works just fine from my 64-bit XP and 64-bit 2003 Server system. > > Obviously some part of your procedure is altering the behavior of the > Explorer shell. I do not believe that this is a bug in OpenAFS as it > works with 64-bit 2003 Server prior to performing the cloning. > > I think you need to compare the systems before and after the sysprep and > determine what has been changed. > > Jeffrey Altman > You're right! It's not an issue with OpenAFS. The culprit is... Internet Explorer Enhanced Security. Fixed! Thank you, gentlemen! -dean From jaltman@secure-endpoints.com Mon Nov 12 20:06:16 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Mon, 12 Nov 2007 15:06:16 -0500 Subject: [OpenAFS] trouble running programs out of AFS after imaging In-Reply-To: <4738AE3F.6030107@njit.edu> References: <4733727F.2040500@njit.edu> <47387317.3000401@njit.edu> <943BC6F428A34876862F0EB1E37821A5@CDCHOME> <47388FE0.4000900@secure-endpoints.com> <473890D5.2090902@njit.edu> <47389922.3030509@secure-endpoints.com> <4738AE3F.6030107@njit.edu> Message-ID: <4738B238.2030805@secure-endpoints.com> Dean Knape wrote: > You're right! It's not an issue with OpenAFS. The culprit is... > Internet Explorer Enhanced Security. You should be able to fix the problem by adding "AFS" to the list of trusted servers instead of by disabling the Enhanced Security. Jeffrey Altman From ronc@depauw.edu Tue Nov 13 02:05:08 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Mon, 12 Nov 2007 21:05:08 -0500 Subject: [OpenAFS] openafs-1.4.5 install In-Reply-To: References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> <47365099.302@depauw.edu> Message-ID: <47390654.1050201@depauw.edu> Ok, what rpm is the userspace one ? thanks, Ron Simon Wilkinson wrote: > > On 11 Nov 2007, at 00:45, Ron Croonenberg wrote: > >> I looked in the Fedora 6 section, but didn't see a 'openafs-kmod-common' >> rpm ? > > It's provided by openafs-client - you need to install the userspace for > the kernel module to be of any use to you. > > Cheers, > > Simon. > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info -- ================================================================= Ron Croonenberg | | Phone: 1 765 658 4761 Lab Instructor & | Fax: 1 765 658 4732 Technology Coordinator | | Department of Computer Science | e-mail: ronc@DePauw.edu DePauw University | 275 Julian Science & Math Center | 602 South College Ave. | Greencastle, IN 46135 | ================================================================= From shadow@gmail.com Tue Nov 13 04:36:54 2007 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 12 Nov 2007 23:36:54 -0500 Subject: [OpenAFS] openafs-1.4.5 install In-Reply-To: <47390654.1050201@depauw.edu> References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> <47365099.302@depauw.edu> <47390654.1050201@depauw.edu> Message-ID: On Nov 12, 2007 9:05 PM, Ron Croonenberg wrote: > Ok, > > what rpm is the userspace one ? > openafs-client, just like he said. From ronc@depauw.edu Tue Nov 13 07:16:03 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Tue, 13 Nov 2007 02:16:03 -0500 Subject: [OpenAFS] openafs-1.4.5 install In-Reply-To: References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> <47365099.302@depauw.edu> <47390654.1050201@depauw.edu> Message-ID: <47394F33.6020001@depauw.edu> Ok I tried that, same result. I downloaded: openafs-1.4.5-fc6.1.x86_64.rpm kmod-openafs-1.4.5-1.2.6.22.5_49.fc6.x86_64.rpm openafs-client-1.4.5-fc6.1.x86_64.rpm first installed: openafs-1.4.5-fc6.1.x86_64.rpm (no complaints) than tried: kmod-openafs, didn't work. So I tried to install the client first (after openafs-1.4.5-fc6.1.x86_64.rpm) , same msgs # rpm -ivh kmod-openafs-1.4.5-1.2.6.22.5_49.fc6.x86_64.rpmerror: Failed dependencies: openafs-kmod-common >= 1.4.5 is needed by kmod-openafs-1.4.5-1.2.6.22.5_49.fc6.x86_64 # rpm -ivh openafs-client-1.4.5-fc6.1.x86_64.rpm error: Failed dependencies: openafs-kmod >= 1.4.5 is needed by openafs-client-1.4.5-fc6.1.x86_64 Ron Derrick Brashear wrote: > On Nov 12, 2007 9:05 PM, Ron Croonenberg wrote: >> Ok, >> >> what rpm is the userspace one ? >> > openafs-client, just like he said. > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info -- ================================================================= Ron Croonenberg | | Phone: 1 765 658 4761 Lab Instructor & | Fax: 1 765 658 4732 Technology Coordinator | | Department of Computer Science | e-mail: ronc@DePauw.edu DePauw University | 275 Julian Science & Math Center | 602 South College Ave. | Greencastle, IN 46135 | ================================================================= From rra@stanford.edu Tue Nov 13 08:01:32 2007 From: rra@stanford.edu (Russ Allbery) Date: Tue, 13 Nov 2007 00:01:32 -0800 Subject: [OpenAFS] openafs-1.4.5 install In-Reply-To: <47394F33.6020001@depauw.edu> (Ron Croonenberg's message of "Tue, 13 Nov 2007 02:16:03 -0500") References: <472FE85C.8010206@depauw.edu> <87ir4gvwgc.fsf@windlord.stanford.edu> <47365099.302@depauw.edu> <47390654.1050201@depauw.edu> <47394F33.6020001@depauw.edu> Message-ID: <87640660bn.fsf@windlord.stanford.edu> Ron Croonenberg writes: > Ok I tried that, same result. > I downloaded: > openafs-1.4.5-fc6.1.x86_64.rpm > kmod-openafs-1.4.5-1.2.6.22.5_49.fc6.x86_64.rpm > openafs-client-1.4.5-fc6.1.x86_64.rpm > first installed: > openafs-1.4.5-fc6.1.x86_64.rpm (no complaints) > than tried: kmod-openafs, didn't work. > So I tried to install the client first (after > openafs-1.4.5-fc6.1.x86_64.rpm) , same msgs > # rpm -ivh kmod-openafs-1.4.5-1.2.6.22.5_49.fc6.x86_64.rpmerror: Failed > dependencies: > openafs-kmod-common >= 1.4.5 is needed by > kmod-openafs-1.4.5-1.2.6.22.5_49.fc6.x86_64 > # rpm -ivh openafs-client-1.4.5-fc6.1.x86_64.rpm > error: Failed dependencies: > openafs-kmod >= 1.4.5 is needed by openafs-client-1.4.5-fc6.1.x86_64 As Simon already said, you have to install all the RPMs at the same time with a single invocation of rpm. -- Russ Allbery (rra@stanford.edu) From jason@rampaginggeek.com Tue Nov 13 13:42:15 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Tue, 13 Nov 2007 08:42:15 -0500 Subject: [OpenAFS] Best practice: inode or namei fileserver? Message-ID: <4739A9B7.7030105@rampaginggeek.com> Hi all, We are currently running inode-based fileservers on solaris 9. I stumbled across the fact that solaris 9 -9/05HW makes logging the default on UFS. I know that the AFS finode-based fileserver cannot work with a logging filesystem. Does the namei filesystem play nice with logging filesystems? Going forward, which format is recommended, inode or namei? I'm wondering if I should slowly migrate to namei. Thanks, Jason From cclausen@acm.org Tue Nov 13 14:01:42 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Tue, 13 Nov 2007 08:01:42 -0600 Subject: [OpenAFS] Best practice: inode or namei fileserver? References: <4739A9B7.7030105@rampaginggeek.com> Message-ID: <14E224DC27EE4CF6A3F5CE187B2FC695@CDCHOME> Jason Edgecombe wrote: > We are currently running inode-based fileservers on solaris 9. > > Does the namei filesystem play nice with logging filesystems? It seems to. > Going forward, which format is recommended, inode or namei? I migrated some Solaris systems to namei simply to use ZFS as there is no inode support for ZFS currently. < References: <4739A9B7.7030105@rampaginggeek.com> Message-ID: <4739AEF5.3090605@rzg.mpg.de> Jason Edgecombe wrote: > Hi all, > > We are currently running inode-based fileservers on solaris 9. > > I stumbled across the fact that solaris 9 -9/05HW makes logging the > default on UFS. I know that the AFS finode-based fileserver cannot work > with a logging filesystem. > > Does the namei filesystem play nice with logging filesystems? Yes > > Going forward, which format is recommended, inode or namei? Namei has another advantage: if you salvage a single volume it's not necessary to read all inodes, but only those pseudo-inodes (file names) under the subdirectory belonging to the volume group. This is much faster. An overhead traversing the AFSIDat-tree to open a file certainly exists, but I suppose it is neglectible compared to the advantages. -Hartmut > > I'm wondering if I should slowly migrate to namei. > > Thanks, > Jason > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info -- ----------------------------------------------------------------- Hartmut Reuter e-mail reuter@rzg.mpg.de phone +49-89-3299-1328 RZG (Rechenzentrum Garching) fax +49-89-3299-1301 Computing Center of the Max-Planck-Gesellschaft (MPG) and the Institut fuer Plasmaphysik (IPP) ----------------------------------------------------------------- From jfgreen@msu.edu Tue Nov 13 16:44:58 2007 From: jfgreen@msu.edu (Jim Green) Date: Tue, 13 Nov 2007 11:44:58 -0500 Subject: [OpenAFS] kmod rpm problem Message-ID: <007401c82614$872e2990$a2040a23@JGTP> I get the following failed dependencies errors when trying to install kmod-openafs-1.4.5-1.2.6.22.9_61.fc6.i686.rpm on Fedora Core 6: error: Failed dependencies: kernel-i686 = 2.6.22.9-61.fc6 is needed by kmod-openafs-1.4.5-1.2.6.22.9_61.fc6.i686 openafs-kmod-common >= 1.4.5 is needed by kmod-openafs-1.4.5-1.2.6.22.9_61.fc6.i686 I recently upgraded my kernel to 2.6.22.9-61 -- here's the output from uname -a: Linux localhost.localdomain 2.6.22.9-61.fc6 #1 SMP Thu Sep 27 17:45:57 EDT 2007 i686 i686 i386 GNU/Linux Any suggestions would be appreciated, thanks. Jim Green Michigan State University Academic Comuting & Network Services jfgreen@msu.edu From sxw@inf.ed.ac.uk Tue Nov 13 16:54:05 2007 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Tue, 13 Nov 2007 16:54:05 +0000 Subject: [OpenAFS] kmod rpm problem In-Reply-To: <007401c82614$872e2990$a2040a23@JGTP> References: <007401c82614$872e2990$a2040a23@JGTP> Message-ID: <71968D8D-A1B0-47E4-A78E-604A591377C7@inf.ed.ac.uk> On 13 Nov 2007, at 16:44, Jim Green wrote: > I get the following failed dependencies errors when trying to install > kmod-openafs-1.4.5-1.2.6.22.9_61.fc6.i686.rpm on Fedora Core 6: > > error: Failed dependencies: > kernel-i686 = 2.6.22.9-61.fc6 is needed by > kmod-openafs-1.4.5-1.2.6.22.9_61.fc6.i686 > openafs-kmod-common >= 1.4.5 is needed by > kmod-openafs-1.4.5-1.2.6.22.9_61.fc6.i686 First problem is that if you're using the kmod kernel modules, you need to use an openafs-client RPM that's been built to support them. Do you already have the FC6 1.4.5 openafs-client RPM installed? If not, you need to install it in the same transaction as you install the kmod. Second problem is why it's thinking that the kernel-i686 dependency is unsatisfied. I assume that the kernel you have installed is the 2.6.22.9-61.fc6 one, and that you've installed the i686 variant. Ca n you run: rpm -q --whatprovides kernel-i686 and rpm -q --provides kernel and let me know the output. Thanks, Simon. From jfgreen@msu.edu Tue Nov 13 18:09:38 2007 From: jfgreen@msu.edu (Jim Green) Date: Tue, 13 Nov 2007 13:09:38 -0500 Subject: [OpenAFS] kmod rpm problem Message-ID: <008a01c82620$5ae69910$a2040a23@JGTP> Well, this definitely looks goofed up: rpm -q --whatprovides kernel-i686 no package provides kernel-i686 rpm -q --provides kernel kernel-drm = 4.3.0 kernel-drm-nouveau = 6 kernel-i586 = 2.5.22.7-57.fc6 kernel = 2.6.22.7-57.fc6 kernel-drm = 4.3.0 kernel-drm-nouveau = 6 kernel-i586 = 2.6.22.9-61.fc6 kernel = 2.6.22.9-61.fc6 From warlord@MIT.EDU Tue Nov 13 18:15:52 2007 From: warlord@MIT.EDU (Derek Atkins) Date: Tue, 13 Nov 2007 13:15:52 -0500 Subject: [OpenAFS] kmod rpm problem In-Reply-To: <008a01c82620$5ae69910$a2040a23@JGTP> References: <008a01c82620$5ae69910$a2040a23@JGTP> Message-ID: <20071113131552.jnkom6y8y9s0c8c4@webmail.mit.edu> You have the i586 kernel installed, not the i686 kernel. -derek Quoting Jim Green : > Well, this definitely looks goofed up: > > rpm -q --whatprovides kernel-i686 > no package provides kernel-i686 > > rpm -q --provides kernel > kernel-drm = 4.3.0 > kernel-drm-nouveau = 6 > kernel-i586 = 2.5.22.7-57.fc6 > kernel = 2.6.22.7-57.fc6 > kernel-drm = 4.3.0 > kernel-drm-nouveau = 6 > kernel-i586 = 2.6.22.9-61.fc6 > kernel = 2.6.22.9-61.fc6 > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > -- Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory Member, MIT Student Information Processing Board (SIPB) URL: http://web.mit.edu/warlord/ PP-ASEL-IA N1NWH warlord@MIT.EDU PGP key available From jfgreen@msu.edu Tue Nov 13 18:33:55 2007 From: jfgreen@msu.edu (Jim Green) Date: Tue, 13 Nov 2007 13:33:55 -0500 Subject: [OpenAFS] kmod rpm problem In-Reply-To: <20071113131552.jnkom6y8y9s0c8c4@webmail.mit.edu> References: <008a01c82620$5ae69910$a2040a23@JGTP> <20071113131552.jnkom6y8y9s0c8c4@webmail.mit.edu> Message-ID: <00a401c82623$bf3101f0$a2040a23@JGTP> So it seems, although uname -a led me to believe my kernel matched up with the OpenAFS rpms. I don't recall explicitly specifying i586 anywhere in the Fedora install or upadate process. I wonder what my best option is -- try to change kernels, or try to build OpenAFS from source? Or will there be an i586 rpm at some point? -----Original Message----- From: Derek Atkins [mailto:warlord@MIT.EDU] Sent: Tuesday, November 13, 2007 1:16 PM To: Jim Green Cc: openafs-info@openafs.org; sxw@inf.ed.ac.uk Subject: Re: [OpenAFS] kmod rpm problem You have the i586 kernel installed, not the i686 kernel. -derek Quoting Jim Green : > Well, this definitely looks goofed up: > > rpm -q --whatprovides kernel-i686 > no package provides kernel-i686 > > rpm -q --provides kernel > kernel-drm = 4.3.0 > kernel-drm-nouveau = 6 > kernel-i586 = 2.5.22.7-57.fc6 > kernel = 2.6.22.7-57.fc6 > kernel-drm = 4.3.0 > kernel-drm-nouveau = 6 > kernel-i586 = 2.6.22.9-61.fc6 > kernel = 2.6.22.9-61.fc6 > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > -- Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory Member, MIT Student Information Processing Board (SIPB) URL: http://web.mit.edu/warlord/ PP-ASEL-IA N1NWH warlord@MIT.EDU PGP key available From jason@rampaginggeek.com Tue Nov 13 19:22:45 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Tue, 13 Nov 2007 14:22:45 -0500 Subject: [OpenAFS] Best practice: inode or namei fileserver? In-Reply-To: <4739AEF5.3090605@rzg.mpg.de> References: <4739A9B7.7030105@rampaginggeek.com> <4739AEF5.3090605@rzg.mpg.de> Message-ID: <4739F985.8000001@rampaginggeek.com> Hartmut Reuter wrote: > Namei has another advantage: if you salvage a single volume it's not > necessary to read all inodes, but only those pseudo-inodes (file names) > under the subdirectory belonging to the volume group. This is much > faster. > > An overhead traversing the AFSIDat-tree to open a file certainly > exists, but I suppose it is neglectible compared to the advantages. > Does the inode server slow down as the partition size grows? Am I going to see a noticable slowdown on the inode server on a 250GB or 500GB partition? Thanks, Jason From marc.dionne@technoconseil.com Tue Nov 13 23:31:10 2007 From: marc.dionne@technoconseil.com (Marc Dionne) Date: Tue, 13 Nov 2007 18:31:10 -0500 Subject: [OpenAFS] kmod rpm problem In-Reply-To: <00a401c82623$bf3101f0$a2040a23@JGTP> References: <008a01c82620$5ae69910$a2040a23@JGTP> <20071113131552.jnkom6y8y9s0c8c4@webmail.mit.edu> <00a401c82623$bf3101f0$a2040a23@JGTP> Message-ID: <473A33BE.1070700@technoconseil.com> Jim Green wrote: > So it seems, although uname -a led me to believe my kernel matched up with > the OpenAFS rpms. I don't recall explicitly specifying i586 anywhere in the > Fedora install or upadate process. > > I wonder what my best option is -- try to change kernels, or try to build > OpenAFS from source? Or will there be an i586 rpm at some point? This was a bug in the FC6 installer (anaconda) - I got bit by it at the time. You really want to replace the kernel with the i686 version or you'll have trouble building or installing other kernel modules besides AFS (nvidia, etc.). See: http://fedoraproject.org/wiki/Bugs/FC6Common Marc From hans@enem.nl Wed Nov 14 11:17:44 2007 From: hans@enem.nl (Hans Melgers) Date: Wed, 14 Nov 2007 12:17:44 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client Message-ID: <473AD958.9000502@enem.nl> Hello, We just installed some XP machines with the latest 1.5.27 afs client. On several of them Word crashes badly when trying to open a .doc file on afs. Most of them running office XP but also seen on office 2003. I run 1.5.25 and havent seen the problem. Could this be a bug in the latest win client ? acl's are ok. Server is 1.4.4 on freebsd 6.2 Anyone else seeing this behaviour ? In Filelog i only see some callback issues: Wed Nov 14 10:56:32 2007 CB: ProbeUuid for 213.46.24.222:17980 failed -01 Wed Nov 14 10:57:11 2007 CB: ProbeUuid for 213.46.24.222:17982 failed -01 Wed Nov 14 10:58:08 2007 CB: ProbeUuid for 213.46.24.222:17987 failed -01 Wed Nov 14 10:58:18 2007 CB: ProbeUuid for 213.46.24.222:17976 failed -01 Wed Nov 14 11:01:25 2007 CB: ProbeUuid for 213.46.24.222:17995 failed -01 Any help appreciated. Hans From jfgreen@msu.edu Wed Nov 14 12:45:15 2007 From: jfgreen@msu.edu (Jim Green) Date: Wed, 14 Nov 2007 07:45:15 -0500 Subject: [OpenAFS] kmod rpm problem In-Reply-To: <473A33BE.1070700@technoconseil.com> References: <008a01c82620$5ae69910$a2040a23@JGTP> <20071113131552.jnkom6y8y9s0c8c4@webmail.mit.edu> <00a401c82623$bf3101f0$a2040a23@JGTP> <473A33BE.1070700@technoconseil.com> Message-ID: <003701c826bc$34894e60$f1080d23@JGTP> Many thanks for this information and to all who responded to my posts. Nice to know I wasn't crazy.... -----Original Message----- From: Marc Dionne [mailto:marc.dionne@technoconseil.com] Sent: Tuesday, November 13, 2007 6:31 PM To: Jim Green Cc: openafs-info@openafs.org Subject: Re: [OpenAFS] kmod rpm problem Jim Green wrote: > So it seems, although uname -a led me to believe my kernel matched up > with the OpenAFS rpms. I don't recall explicitly specifying i586 > anywhere in the Fedora install or upadate process. > > I wonder what my best option is -- try to change kernels, or try to > build OpenAFS from source? Or will there be an i586 rpm at some point? This was a bug in the FC6 installer (anaconda) - I got bit by it at the time. You really want to replace the kernel with the i686 version or you'll have trouble building or installing other kernel modules besides AFS (nvidia, etc.). See: http://fedoraproject.org/wiki/Bugs/FC6Common Marc From ipmonger@delamancha.org Wed Nov 14 13:49:45 2007 From: ipmonger@delamancha.org (Jon Boone) Date: Wed, 14 Nov 2007 08:49:45 -0500 Subject: [OpenAFS] Problem upgrading from 1.4.4 to 1.4.5 on Mac OS X Tiger / Intel Message-ID: <159EB692-A529-47B7-A4D1-828F4F2F4BCB@delamancha.org> Folks, I downloaded the 1.4.5 .dmg and tried to install 1.4.5 on my MacBook Pro (running Tiger). The system is already running 1.4.4. I double-clicked on the .pkg and went through all of the installer prompts. The installer then says "You can not continue. There is nothing to install." --jon From jaltman@secure-endpoints.com Wed Nov 14 14:42:56 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Wed, 14 Nov 2007 09:42:56 -0500 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <473AD958.9000502@enem.nl> References: <473AD958.9000502@enem.nl> Message-ID: <473B0970.2080009@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms050506050305060802040905 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Hans: Please follow the debugging section of the OpenAFS for Windows Release Notes. In particular, use the SysInternals' Process Monitor to log the requests from Word so that you can determine which requests are failing with 1.5.27 that succeed with 1.5.25. One change that went into 1.5.27 is support for retrying requests when the file server responds to the client with EWOULDBLOCK / EAGAIN. In prior releases the client would simply fail the request. In 1.5.27 the client retries the request every two seconds up to 45 seconds until the request completes or the timeout is reached. In the trace log it is the cm_Analyze() messages will include the error number. Possible error numbers to look for include: UAEWOULDBLOCK 49733416L UAEAGAIN 49733386L EAGAIN 11L Jeffrey Altman Hans Melgers wrote: > > Hello, > > We just installed some XP machines with the latest 1.5.27 afs client. On > several of them Word crashes badly when trying to open a .doc file on > afs. Most of them running office XP but also seen on office 2003. > > I run 1.5.25 and havent seen the problem. Could this be a bug in the > latest win client ? acl's are ok. Server is 1.4.4 on freebsd 6.2 > Anyone else seeing this behaviour ? > > In Filelog i only see some callback issues: > > Wed Nov 14 10:56:32 2007 CB: ProbeUuid for 213.46.24.222:17980 failed -01 > Wed Nov 14 10:57:11 2007 CB: ProbeUuid for 213.46.24.222:17982 failed -01 > Wed Nov 14 10:58:08 2007 CB: ProbeUuid for 213.46.24.222:17987 failed -01 > Wed Nov 14 10:58:18 2007 CB: ProbeUuid for 213.46.24.222:17976 failed -01 > Wed Nov 14 11:01:25 2007 CB: ProbeUuid for 213.46.24.222:17995 failed -01 > > > Any help appreciated. > > Hans --------------ms050506050305060802040905 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMTQxNDQyNTZaMCMGCSqGSIb3DQEJBDEWBBTLTdM0 dxiURAu37nl6hXwBHuDxDzBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEALj4kcQ/VS0yJab0K8UwN8SHAhYVoaGmVNa+0qFV9qranLPvsJAnT nCSpvqcj4+upYQMGAD4e0djpl3KKu0r2eWBvszZZNVddMXRcb89XxQqVWngIO7opnEzkRoDF E+Ky07fvcDhom+psHJfYPNIBXcwklA/oqAavo3+3iyADPArKyz2kAvPchBYdkelHyUDCut9Y V4I9bkKj+JqoBhIN2UNX5MXd5SuKqG9eWLOfkMPDZJiDEMhwM6GJSKmSxW50PXsBDIAWPiZz 5YlFI8NAlzaXe1i+mjqlhwUaSOmjiRiz3g1H1gMgaguL2wS1zMnxvjEyPXFHnHsS+ByIFLWG vQAAAAAAAA== --------------ms050506050305060802040905-- From jaltman@secure-endpoints.com Wed Nov 14 15:03:11 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Wed, 14 Nov 2007 10:03:11 -0500 Subject: [OpenAFS] Problem upgrading from 1.4.4 to 1.4.5 on Mac OS X Tiger / Intel In-Reply-To: <159EB692-A529-47B7-A4D1-828F4F2F4BCB@delamancha.org> References: <159EB692-A529-47B7-A4D1-828F4F2F4BCB@delamancha.org> Message-ID: <473B0E2F.3030106@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms030305080202080306000406 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Jon Boone wrote: > Folks, > > I downloaded the 1.4.5 .dmg and tried to install 1.4.5 on my MacBook > Pro (running Tiger). The system is already running 1.4.4. I > double-clicked on the .pkg and went through all of the installer > prompts. The installer then says "You can not continue. There is > nothing to install." > > --jon Which 1.4.5 dmg did you download? There are separate installers for Tiger and Leopard. --------------ms030305080202080306000406 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMTQxNTAzMTFaMCMGCSqGSIb3DQEJBDEWBBS5I7+s ZTVRKQnmp6B8CwM1YkXjfDBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAUIG5lJEU2yrXpQIUfqC43BjKsTirVMpwKIZDtQMK8jyIm4AyNB1E P3nHcKz5oSbmDkyWtXjAhE+Q+iL+Ei2IUs2QOHmyIB6HtELV04orod6ybSu8BwIBNVWKiXP0 /pyoun9UiggYBVuFH1rRbbANpJbCYximnzyweOaIzEEgyh4WnS2tBUwIsItNSY93MPLoWApC b72Q4Tytgippj5BwYqBqY+5XP1fEUt3DRYOcGIOJxCJ1OmwbXnh/0/TL4MakJNMbCvXEh9cr M/04KbRx1ZLTk+Fq/iOdqUp3y/mFbJRSj+8D2pn7+wn6g79CpT4NSQwUjXYsIgDTlonH/NXZ IgAAAAAAAA== --------------ms030305080202080306000406-- From ipmonger@delamancha.org Wed Nov 14 15:04:05 2007 From: ipmonger@delamancha.org (Jon Boone) Date: Wed, 14 Nov 2007 10:04:05 -0500 Subject: [OpenAFS] Problem upgrading from 1.4.4 to 1.4.5 on Mac OS X Tiger / Intel In-Reply-To: <473B0E2F.3030106@secure-endpoints.com> References: <159EB692-A529-47B7-A4D1-828F4F2F4BCB@delamancha.org> <473B0E2F.3030106@secure-endpoints.com> Message-ID: <5E62B458-F030-44A1-AE18-8406068AABCD@delamancha.org> On Nov 14, 2007, at 10:03, Jeffrey Altman wrote: > Jon Boone wrote: >> Folks, >> >> I downloaded the 1.4.5 .dmg and tried to install 1.4.5 on my >> MacBook >> Pro (running Tiger). The system is already running 1.4.4. I >> double-clicked on the .pkg and went through all of the installer >> prompts. The installer then says "You can not continue. There is >> nothing to install." >> >> --jon > > Which 1.4.5 dmg did you download? There are separate installers for > Tiger and Leopard. The one for Tiger. From jblaine@kickflop.net Wed Nov 14 15:53:39 2007 From: jblaine@kickflop.net (Jeff Blaine) Date: Wed, 14 Nov 2007 10:53:39 -0500 Subject: [OpenAFS] Re: CVS, GSSAPI, and AFS tokens In-Reply-To: <473202C7.9010900@kickflop.net> References: <473202C7.9010900@kickflop.net> Message-ID: <473B1A03.1020705@kickflop.net> Feeding results back for others -- the following appears to work fine so far. I cleared all creds on the server for user jblaine (krb5 and AFS tokens) and was able to checkout from AFS ACL-protected space lacking system:anyuser privs. Client: CVS_RSH=/usr/bin/ssh CVSROOT=:ext:jblaine@whatever.com:/afs/my/cvsroot Server: sshd configured for PAM auth + pam_krb5.so (Russ Alberry's) + pam_afs_session.so I've yet to try to figure it all out with ticket forwarding. Jeff Blaine wrote: > How are people handling krb5 auth with CVS and also getting > tokens for gserver connections (GSSAPI/krb5)? > From deengert@anl.gov Wed Nov 14 16:59:24 2007 From: deengert@anl.gov (Douglas E. Engert) Date: Wed, 14 Nov 2007 10:59:24 -0600 Subject: [OpenAFS] Re: CVS, GSSAPI, and AFS tokens In-Reply-To: <473B1A03.1020705@kickflop.net> References: <473202C7.9010900@kickflop.net> <473B1A03.1020705@kickflop.net> Message-ID: <473B296C.2030601@anl.gov> Jeff Blaine wrote: > Feeding results back for others -- the following appears to > work fine so far. I cleared all creds on the server for > user jblaine (krb5 and AFS tokens) and was able to checkout > from AFS ACL-protected space lacking system:anyuser privs. > > Client: > > CVS_RSH=/usr/bin/ssh > CVSROOT=:ext:jblaine@whatever.com:/afs/my/cvsroot > > Server: > > sshd configured for PAM auth + > pam_krb5.so (Russ Alberry's) + > pam_afs_session.so > > I've yet to try to figure it all out with ticket forwarding. Sounds like it did forward, and sshd uses pam_afs_session to get the token. You could try ssh by itself, then do a klist; tokens Or start sshd -ddd and look at the debug as you connect to cvs. > > Jeff Blaine wrote: >> How are people handling krb5 auth with CVS and also getting >> tokens for gserver connections (GSSAPI/krb5)? >> > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > > -- Douglas E. Engert Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439 (630) 252-5444 From jblaine@kickflop.net Wed Nov 14 17:55:33 2007 From: jblaine@kickflop.net (Jeff Blaine) Date: Wed, 14 Nov 2007 12:55:33 -0500 Subject: [OpenAFS] Re: CVS, GSSAPI, and AFS tokens In-Reply-To: <473B296C.2030601@anl.gov> References: <473202C7.9010900@kickflop.net> <473B1A03.1020705@kickflop.net> <473B296C.2030601@anl.gov> Message-ID: <473B3695.9000606@kickflop.net> Douglas E. Engert wrote: > Jeff Blaine wrote: >> Feeding results back for others -- the following appears to >> work fine so far. I cleared all creds on the server for >> user jblaine (krb5 and AFS tokens) and was able to checkout >> from AFS ACL-protected space lacking system:anyuser privs. >> >> Client: >> >> CVS_RSH=/usr/bin/ssh >> CVSROOT=:ext:jblaine@whatever.com:/afs/my/cvsroot >> >> Server: >> >> sshd configured for PAM auth + >> pam_krb5.so (Russ Alberry's) + >> pam_afs_session.so >> >> I've yet to try to figure it all out with ticket forwarding. > > Sounds like it did forward, and sshd uses pam_afs_session to get the > token. No, I'm being asked for a password by pam_krb5.so. I haven't determined why yet. I'll be sure to post if I figure it out. From shadow@gmail.com Wed Nov 14 18:27:30 2007 From: shadow@gmail.com (Derrick Brashear) Date: Wed, 14 Nov 2007 13:27:30 -0500 Subject: [OpenAFS] Problem upgrading from 1.4.4 to 1.4.5 on Mac OS X Tiger / Intel In-Reply-To: <159EB692-A529-47B7-A4D1-828F4F2F4BCB@delamancha.org> References: <159EB692-A529-47B7-A4D1-828F4F2F4BCB@delamancha.org> Message-ID: Try uninstalling 1.4.4 and installing 1.4.5? I'm unsure why that would happen, though. On Nov 14, 2007 8:49 AM, Jon Boone wrote: > Folks, > > I downloaded the 1.4.5 .dmg and tried to install 1.4.5 on my > MacBook Pro (running Tiger). The system is already running 1.4.4. I > double-clicked on the .pkg and went through all of the installer > prompts. The installer then says "You can not continue. There is > nothing to install." > > --jon > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > From jbrown94305@gmail.com Wed Nov 14 18:31:39 2007 From: jbrown94305@gmail.com (Jim Brown) Date: Wed, 14 Nov 2007 10:31:39 -0800 Subject: [OpenAFS] Problem upgrading from 1.4.4 to 1.4.5 on Mac OS X Tiger / Intel In-Reply-To: References: <159EB692-A529-47B7-A4D1-828F4F2F4BCB@delamancha.org> Message-ID: <473B3F0B.5050409@gmail.com> Derrick Brashear wrote: > Try uninstalling 1.4.4 and installing 1.4.5? > > I'm unsure why that would happen, though. > > On Nov 14, 2007 8:49 AM, Jon Boone wrote: > >> Folks, >> >> I downloaded the 1.4.5 .dmg and tried to install 1.4.5 on my >> MacBook Pro (running Tiger). The system is already running 1.4.4. I >> double-clicked on the .pkg and went through all of the installer >> prompts. The installer then says "You can not continue. There is >> nothing to install." >> >> --jon >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info >> >> > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > Delete the receipt From shadow@gmail.com Wed Nov 14 18:53:44 2007 From: shadow@gmail.com (Derrick Brashear) Date: Wed, 14 Nov 2007 13:53:44 -0500 Subject: [OpenAFS] Problem upgrading from 1.4.4 to 1.4.5 on Mac OS X Tiger / Intel In-Reply-To: <473B3F0B.5050409@gmail.com> References: <159EB692-A529-47B7-A4D1-828F4F2F4BCB@delamancha.org> <473B3F0B.5050409@gmail.com> Message-ID: That should only apply if the receipt claims to be for the version you're installing, but, it's worth a try. On Nov 14, 2007 1:31 PM, Jim Brown wrote: > > Derrick Brashear wrote: > > Try uninstalling 1.4.4 and installing 1.4.5? > > > > I'm unsure why that would happen, though. > > > > On Nov 14, 2007 8:49 AM, Jon Boone wrote: > > > >> Folks, > >> > >> I downloaded the 1.4.5 .dmg and tried to install 1.4.5 on my > >> MacBook Pro (running Tiger). The system is already running 1.4.4. I > >> double-clicked on the .pkg and went through all of the installer > >> prompts. The installer then says "You can not continue. There is > >> nothing to install." > >> > >> --jon > >> _______________________________________________ > >> OpenAFS-info mailing list > >> OpenAFS-info@openafs.org > >> https://lists.openafs.org/mailman/listinfo/openafs-info > >> > >> > > _______________________________________________ > > OpenAFS-info mailing list > > OpenAFS-info@openafs.org > > https://lists.openafs.org/mailman/listinfo/openafs-info > > > Delete the receipt > From ipmonger@delamancha.org Wed Nov 14 18:58:17 2007 From: ipmonger@delamancha.org (Jon Boone) Date: Wed, 14 Nov 2007 13:58:17 -0500 Subject: [OpenAFS] Problem upgrading from 1.4.4 to 1.4.5 on Mac OS X Tiger / Intel In-Reply-To: References: <159EB692-A529-47B7-A4D1-828F4F2F4BCB@delamancha.org> <473B3F0B.5050409@gmail.com> Message-ID: <6BB553C0-DC9C-4ABD-943A-64479285E2B2@delamancha.org> On Nov 14, 2007, at 13:53, Derrick Brashear wrote: > That should only apply if the receipt claims to be for the version > you're installing, but, it's worth a try. > I deleted the receipt and the install worked. --jon From hans@enem.nl Wed Nov 14 23:42:10 2007 From: hans@enem.nl (Hans Melgers) Date: Thu, 15 Nov 2007 00:42:10 +0100 Subject: [OpenAFS] Anybody using BackupPC4afs ? Message-ID: <473B87D2.9080103@enem.nl> I was wondering if anybody on this list is using this for afs backups ? can you recommend it ? pro's/cons ? http://www.physics.unc.edu/~stephen/BackupPC4AFS From Mason.Christopher@mayo.edu Thu Nov 15 04:44:44 2007 From: Mason.Christopher@mayo.edu (Christopher Mason) Date: Wed, 14 Nov 2007 23:44:44 -0500 Subject: [OpenAFS] keyring status Message-ID: <4E47C9EE7F23597F7B00F506@[192.168.1.101]> Forgive if this is naively worded or repetitive: What's the status of AFS support for the linux kernel keyring? Specifically, is it possible for me to write an application that, in a single process, accepts kerberos credentials from multiple users via network connections and then uses the credentials to allow each user access to AFS (or other file systems: NFS4, lustre?) file on a thread-by-thread basis respecting ACLs etc? Is there an example of such a thing around? How does filedrawers do this? multiple apache MPMs? Thanks, -c -- [ Christopher Mason MPRC Bioinformatics http://proteomics ] From ronc@depauw.edu Thu Nov 15 15:26:37 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Thu, 15 Nov 2007 10:26:37 -0500 Subject: [OpenAFS] E212: Can't open file for writing In-Reply-To: <4739F985.8000001@rampaginggeek.com> References: <4739A9B7.7030105@rampaginggeek.com> <4739AEF5.3090605@rzg.mpg.de> <4739F985.8000001@rampaginggeek.com> Message-ID: <473C652D.2060901@depauw.edu> Hello all, I tried to copy 2 files onto an afs volume, but that didn't work too well for some reason. (I never had this problem before.) projection_3a.tif" E212: Can't open file for writing ls -al shows: ?--------- ? ? ? ? ? projection_3a.tif ?--------- ? ? ? ? ? projection_3.tif Anyway, I tried to remove etc. the fiels but that doesn't go anywhere. Basically most things say the files are not there, but with ls I see them show up. any ideas ? thanks, Ron Jason Edgecombe wrote: > Hartmut Reuter wrote: >> Namei has another advantage: if you salvage a single volume it's not >> necessary to read all inodes, but only those pseudo-inodes (file names) >> under the subdirectory belonging to the volume group. This is much >> faster. >> >> An overhead traversing the AFSIDat-tree to open a file certainly >> exists, but I suppose it is neglectible compared to the advantages. >> > > Does the inode server slow down as the partition size grows? Am I going > to see a noticable slowdown on the inode server on a 250GB or 500GB > partition? > > Thanks, > Jason > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info -- ================================================================= Ron Croonenberg | | Phone: 1 765 658 4761 Lab Instructor & | Fax: 1 765 658 4732 Technology Coordinator | | Department of Computer Science | e-mail: ronc@DePauw.edu DePauw University | 275 Julian Science & Math Center | 602 South College Ave. | Greencastle, IN 46135 | ================================================================= From ksumner@physics.unc.edu Thu Nov 15 16:37:49 2007 From: ksumner@physics.unc.edu (Kevin Scott Sumner) Date: Thu, 15 Nov 2007 11:37:49 -0500 (EST) Subject: [OpenAFS] 1.2.X and 1.4.5 Interoperation Message-ID: Hi all, We're planning on upgrading our critical AFS servers, and I wanted to check and see if there are any interoperability problems. We plan on doing a rolling migration, swapping out one server at a time. Our current setup: (3) Sun Netra T1s running Solaris 9 (SPARC). They have OpenAFS 1.2.X (1.2.13?), and run vlserver, ptserver, buserver, and runntp. We are moving to: (3) Dell PowerEdge 860s running Debian 4 (i386). They will have OpenAFS 1.4.5, and run all the same services (with the possible exception of runntp). Is there ANYTHING that could POSSIBLY break when we have a mixture of these 1.2 and 1.4 servers running together? Version issues, OS issues, Endian issues, etc. Thanks! Kevin ----- Kevin Sumner ksumner@physics.unc.edu (919) 962-6494 Assistant Systems Administrator Physics and Astronomy Networking Infrastructure and Computing University of North Carolina at Chapel Hill From ksumner@physics.unc.edu Thu Nov 15 17:00:36 2007 From: ksumner@physics.unc.edu (Kevin Scott Sumner) Date: Thu, 15 Nov 2007 12:00:36 -0500 (EST) Subject: [OpenAFS] E212: Can't open file for writing In-Reply-To: <473C652D.2060901@depauw.edu> References: <4739A9B7.7030105@rampaginggeek.com> <4739AEF5.3090605@rzg.mpg.de> <4739F985.8000001@rampaginggeek.com> <473C652D.2060901@depauw.edu> Message-ID: This is similar to a problem we have seen in Linux 2.6 and 1.4.2. On one host, I vi a script, and on another host I try to run it, and it gives me a File Not Found, and an ls -l shows the ? like you're getting... I think this was discussed here [1] and is (or might be) fixed in 1.4.5. [1] http://www.openafs.org/pipermail/openafs-info/2007-September/027355.html If this isn't akin to your problem, we may need more details... Cheers, Kevin ----- Kevin Sumner ksumner@physics.unc.edu (919) 962-6494 Assistant Systems Administrator Physics and Astronomy Networking Infrastructure and Computing University of North Carolina at Chapel Hill "Creative brains are a valuable, limited resource. They shouldn't be wasted on re-inventing the wheel when there are so many fascinating new problems waiting out there." -Eric S. Raymond On Thu, 15 Nov 2007, Ron Croonenberg wrote: > Hello all, > > I tried to copy 2 files onto an afs volume, but that didn't work too well for > some reason. (I never had this problem before.) > > projection_3a.tif" E212: Can't open file for writing > > ls -al shows: > > ?--------- ? ? ? ? ? projection_3a.tif > ?--------- ? ? ? ? ? projection_3.tif > > Anyway, I tried to remove etc. the fiels but that doesn't go anywhere. > Basically most things say the files are not there, but with ls I see them > show up. > > > > any ideas ? > > > thanks, > > Ron > > > > > Jason Edgecombe wrote: >> Hartmut Reuter wrote: >>> Namei has another advantage: if you salvage a single volume it's not >>> necessary to read all inodes, but only those pseudo-inodes (file names) >>> under the subdirectory belonging to the volume group. This is much >>> faster. >>> >>> An overhead traversing the AFSIDat-tree to open a file certainly >>> exists, but I suppose it is neglectible compared to the advantages. >>> >> >> Does the inode server slow down as the partition size grows? Am I going >> to see a noticable slowdown on the inode server on a 250GB or 500GB >> partition? >> >> Thanks, >> Jason >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info > > -- > > ================================================================= > Ron Croonenberg | > | Phone: 1 765 658 4761 > Lab Instructor & | Fax: 1 765 658 4732 > Technology Coordinator | > | > Department of Computer Science | e-mail: ronc@DePauw.edu > DePauw University | > 275 Julian Science & Math Center | > 602 South College Ave. | > Greencastle, IN 46135 | > ================================================================= > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info From HraskyT@seznam.cz Thu Nov 15 19:28:07 2007 From: HraskyT@seznam.cz (=?iso-8859-2?Q?Tom=E1=B9=20Hr=E1sk=FD?=) Date: Thu, 15 Nov 2007 20:28:07 +0100 (CET) Subject: [OpenAFS] =?us-ascii?Q?Aklog=20for=20Solaris?= Message-ID: <3084.5602-1828-1176149303-1195154887@seznam.cz> Hello, I have a question about aklog utility for (open)Solaris. Is there any distribution for Solaris on x86 with akolg compiled ? In my university we have MIT Kerberos and OpenAFS authenticates to it. I can install and use OpenAFS on Ubuntu linux and now I am trying to run it on my OpenSolaris install (SXDE build 70). I managed to see directories of university's cell under /afs but neither pam_afs nor klog work. Any tips? Thank you Tomas Hrasky student of SW engineering, University of West Bohemia, Pilsen From ronc@depauw.edu Thu Nov 15 21:02:11 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Thu, 15 Nov 2007 16:02:11 -0500 Subject: [OpenAFS] E212: Can't open file for writing In-Reply-To: <473C652D.2060901@depauw.edu> References: <4739A9B7.7030105@rampaginggeek.com> <4739AEF5.3090605@rzg.mpg.de> <4739F985.8000001@rampaginggeek.com> <473C652D.2060901@depauw.edu> Message-ID: <473CB3D3.7010607@depauw.edu> Uhm... I noticed that after a while (an hour or so)that problem "fixed itself" ? It looks like I copied the files there and it took a long while before it was actually there (even though sftp said it transferred the files)? Ron Ron Croonenberg wrote: > Hello all, > > I tried to copy 2 files onto an afs volume, but that didn't work too > well for some reason. (I never had this problem before.) > > projection_3a.tif" E212: Can't open file for writing > > ls -al shows: > > ?--------- ? ? ? ? ? projection_3a.tif > ?--------- ? ? ? ? ? projection_3.tif > > Anyway, I tried to remove etc. the fiels but that doesn't go anywhere. > Basically most things say the files are not there, but with ls I see > them show up. > > > > any ideas ? > > > thanks, > > Ron > > > > > Jason Edgecombe wrote: >> Hartmut Reuter wrote: >>> Namei has another advantage: if you salvage a single volume it's not >>> necessary to read all inodes, but only those pseudo-inodes (file names) >>> under the subdirectory belonging to the volume group. This is much >>> faster. >>> >>> An overhead traversing the AFSIDat-tree to open a file certainly >>> exists, but I suppose it is neglectible compared to the advantages. >>> >> >> Does the inode server slow down as the partition size grows? Am I going >> to see a noticable slowdown on the inode server on a 250GB or 500GB >> partition? >> >> Thanks, >> Jason >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info > -- ================================================================= Ron Croonenberg | | Phone: 1 765 658 4761 Lab Instructor & | Fax: 1 765 658 4732 Technology Coordinator | | Department of Computer Science | e-mail: ronc@DePauw.edu DePauw University | 275 Julian Science & Math Center | 602 South College Ave. | Greencastle, IN 46135 | ================================================================= From cclausen@acm.org Thu Nov 15 21:12:59 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Thu, 15 Nov 2007 15:12:59 -0600 Subject: [OpenAFS] E212: Can't open file for writing References: <4739A9B7.7030105@rampaginggeek.com> <4739AEF5.3090605@rzg.mpg.de> <4739F985.8000001@rampaginggeek.com> <473C652D.2060901@depauw.edu> <473CB3D3.7010607@depauw.edu> Message-ID: <68C9A52FE1FF4D7C8716734E9E4D6567@CDCHOME> Ron Croonenberg wrote: > Uhm... I noticed that after a while (an hour or so)that problem > "fixed itself" ? It looks like I copied the files there and it took a > long while before it was actually there (even though sftp said it > transferred the files)? Writes go to your AFS cache first, and then to the fileserver. You could see a large hang at the "end" of a transfer as the data is flushed out of the local cache and actually written to the AFS fileserver. < Forgive the slightly off topic post but I think it applies here as well on the kerberos list Several years ago we moved to MIT kerberos 5. At the time I set the master key in the kdc.conf to: master_key_type = des-cbc-crc I did this to allow transfer of principals from our old kaserver to the new kdc. Now we are trying to get Windows 2003 AD to auth against our Kerberos server and it seems that it will not work with our kdc as it is configured. My question is am I screwed here or just missing something easy? I have tried multiple allowed enctypes and still no luck. If I build a kdc without specifying a master key it seems to work. Have any others done this same thing? Thanks /sd -- Steve Devine Network Storage & Printing Academic Computing & Network Services Michigan State University 506 Computer Center East Lansing, MI 48824-1042 1-517-432-7327 Baseball is ninety percent mental; the other half is physical. - Yogi Berra From cclausen@acm.org Thu Nov 15 22:29:52 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Thu, 15 Nov 2007 16:29:52 -0600 Subject: [OpenAFS] Kerberos5 and afs References: <473CBE05.601@msu.edu> Message-ID: <640C85F483224A75B79BC40AB2994FD3@CDCHOME> Steve Devine wrote: > Forgive the slightly off topic post but I think it applies here as > well on the kerberos list > Several years ago we moved to MIT kerberos 5. At the time I set the > master key in the kdc.conf to: > master_key_type = des-cbc-crc > I did this to allow transfer of principals from our old kaserver to > the new kdc. > Now we are trying to get Windows 2003 AD to auth against our Kerberos > server and it seems that it will not work with our kdc as it is > configured. My question is am I screwed here or just missing > something easy? I have tried multiple allowed enctypes and still no > luck. If I build a kdc without specifying a master key it seems to > work. > Have any others done this same thing? Can you be more specific with what you are attempting? Windows AD can trust an MIT realm. (I have multiple MIT realms trusting AD.UIUC.EDU, one using a des3 master key type and one using des as above.) As far as I can tell, the master key type should not actually matter. < (Steve Devine's message of "Thu, 15 Nov 2007 16:45:41 -0500") References: <473CBE05.601@msu.edu> Message-ID: <87fxz7jfr1.fsf@windlord.stanford.edu> Steve Devine writes: > Forgive the slightly off topic post but I think it applies here as well > on the kerberos list Several years ago we moved to MIT kerberos 5. At > the time I set the master key in the kdc.conf to: > master_key_type = des-cbc-crc > I did this to allow transfer of principals from our old kaserver to the > new kdc. > Now we are trying to get Windows 2003 AD to auth against our Kerberos > server and it seems that it will not work with our kdc as it is > configured. My question is am I screwed here or just missing something > easy? I have tried multiple allowed enctypes and still no luck. > If I build a kdc without specifying a master key it seems to work. > Have any others done this same thing? The master key type doesn't matter at all for cross-realm trust. The only thing the master key is used for is encrypting the KDC database on disk. It is never seen on the wire and no clients of Kerberos are even aware that it exists. What matters for cross-realm trust is the enctypes on the cross-realm krbtgt keys, which must match in both environments (along with the key and the kvno) and must be of an enctype supported in both environments. Most sites these days use rc4-hmac as the cross-realm key type since it's stronger than DES and supported by both Windows and MIT Kerberos. If you're running the latest and greatest Windows AD, you can use AES, but that's pretty bleeding edge still and most people haven't upgraded that far yet. Most cross-realm trust problems with Windows end up being problems with getting the key and kvno synchronized between the environment or having extra stray enctypes on the MIT end that Windows doesn't support. -- Russ Allbery (rra@stanford.edu) From sdevine@msu.edu Thu Nov 15 23:00:34 2007 From: sdevine@msu.edu (Steve Devine) Date: Thu, 15 Nov 2007 18:00:34 -0500 Subject: [OpenAFS] Kerberos5 and afs In-Reply-To: <87fxz7jfr1.fsf@windlord.stanford.edu> References: <473CBE05.601@msu.edu> <87fxz7jfr1.fsf@windlord.stanford.edu> Message-ID: <473CCF92.7030301@msu.edu> Russ Allbery wrote: > Steve Devine writes: > > >> Forgive the slightly off topic post but I think it applies here as well >> on the kerberos list Several years ago we moved to MIT kerberos 5. At >> the time I set the master key in the kdc.conf to: >> > > >> master_key_type = des-cbc-crc >> > > >> I did this to allow transfer of principals from our old kaserver to the >> new kdc. >> > > >> Now we are trying to get Windows 2003 AD to auth against our Kerberos >> server and it seems that it will not work with our kdc as it is >> configured. My question is am I screwed here or just missing something >> easy? I have tried multiple allowed enctypes and still no luck. >> > > >> If I build a kdc without specifying a master key it seems to work. >> Have any others done this same thing? >> > > The master key type doesn't matter at all for cross-realm trust. The only > thing the master key is used for is encrypting the KDC database on disk. > It is never seen on the wire and no clients of Kerberos are even aware > that it exists. > Ok thats a huge relief. > What matters for cross-realm trust is the enctypes on the cross-realm > krbtgt keys, which must match in both environments (along with the key and > the kvno) and must be of an enctype supported in both environments. Most > sites these days use rc4-hmac as the cross-realm key type since it's > stronger than DES and supported by both Windows and MIT Kerberos. If > you're running the latest and greatest Windows AD, you can use AES, but > that's pretty bleeding edge still and most people haven't upgraded that > far yet. > > Most cross-realm trust problems with Windows end up being problems with > getting the key and kvno synchronized between the environment or having > extra stray enctypes on the MIT end that Windows doesn't support. > > Does the order of the enctypes listed in the kdc affect this? This is my current kdc.conf entry: supported_enctypes = des3-hmac-sha1:normal des-cbc-crc:normal des-cbc-crc:v4 des-cbc-crc:afs3 I'm not sure how to manipulate the kvno on the AD Thanks /sd -- Steve Devine Storage Systems Academic Computing & Network Services Michigan State University 506 Computer Center East Lansing, MI 48824-1042 1-517-432-7327 Baseball is ninety percent mental; the other half is physical. - Yogi Berra From rra@stanford.edu Thu Nov 15 23:11:09 2007 From: rra@stanford.edu (Russ Allbery) Date: Thu, 15 Nov 2007 15:11:09 -0800 Subject: [OpenAFS] Kerberos5 and afs In-Reply-To: <473CCF92.7030301@msu.edu> (Steve Devine's message of "Thu, 15 Nov 2007 18:00:34 -0500") References: <473CBE05.601@msu.edu> <87fxz7jfr1.fsf@windlord.stanford.edu> <473CCF92.7030301@msu.edu> Message-ID: <87ve83hzoy.fsf@windlord.stanford.edu> Steve Devine writes: > Does the order of the enctypes listed in the kdc affect this? In my experience, the enctype list should match exactly. It doesn't matter what order you list the enctypes in; if you have enctypes on the krbtgt key that aren't present in Windows, you may lose. So, in this case: > This is my current kdc.conf entry: > supported_enctypes = des3-hmac-sha1:normal des-cbc-crc:normal > des-cbc-crc:v4 des-cbc-crc:afs3 you need to explicitly specify -e des-cbc-crc:normal when creating the krbtgt cross-realm keys. Otherwise you'll get a des3 key in your KDC and since Windows doesn't support des3, you'll lose. Also, if you're entering a password to create this key, be very careful of the salting algorithm. I think that you'll need to fix that on the Windows side, since IIRC MIT Kerberos can't do the Windows salt but Windows can do the MIT salt (if configured correctly), but it's been a long time and I'm forgetting the details. > I'm not sure how to manipulate the kvno on the AD It depends on the version of Windows. Sometimes you can't at all. And regardless, since on the MIT side you can just use modprinc -kvno, it's way easier to make the MIT side match Windows than vice versa. -- Russ Allbery (rra@stanford.edu) From jaltman@secure-endpoints.com Thu Nov 15 23:18:35 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Thu, 15 Nov 2007 18:18:35 -0500 Subject: [OpenAFS] Kerberos5 and afs In-Reply-To: <87ve83hzoy.fsf@windlord.stanford.edu> References: <473CBE05.601@msu.edu> <87fxz7jfr1.fsf@windlord.stanford.edu> <473CCF92.7030301@msu.edu> <87ve83hzoy.fsf@windlord.stanford.edu> Message-ID: <473CD3CB.90104@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms030401020205020706090103 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Russ Allbery wrote: > Steve Devine writes: > >> Does the order of the enctypes listed in the kdc affect this? > > In my experience, the enctype list should match exactly. It doesn't > matter what order you list the enctypes in; if you have enctypes on the > krbtgt key that aren't present in Windows, you may lose. So, in this > case: > >> This is my current kdc.conf entry: >> supported_enctypes = des3-hmac-sha1:normal des-cbc-crc:normal >> des-cbc-crc:v4 des-cbc-crc:afs3 > > you need to explicitly specify -e des-cbc-crc:normal when creating the > krbtgt cross-realm keys. Otherwise you'll get a des3 key in your KDC and > since Windows doesn't support des3, you'll lose. Windows 2003 SP1 and later supports RC4-HMAC cross-realm keys. --------------ms030401020205020706090103 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMTUyMzE4MzZaMCMGCSqGSIb3DQEJBDEWBBTRvFdk F+z0mBVHjRmuYjdTCj4FIjBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAGmwSkZoJr9MAjsgJoK2QzDXTzLLzpUZNTMkBxkGiACCQbnTXqdf6 GuNdcyOR3U1023F9aZy694jaNBy+m27fRPG/IemftH+E1tZ/S1IRsDaQ1TkPpZHv8GByS9Ok mY8JU1y55YWCp7M8kkSIXNGO6wSc+jOp3VKb4KetiIY5flHu0hoOp9uk98n2+4vsaeutG75D BuYFOUIo3RT4AZ5ChymFKKTcgFfz/DN4RmBsUroJIS7Wh2q7Ks/G5wGJNWMZfA6/k0Ij6Nkn aj/qDz+ZkjMtwItEX8nBymlTKrZSjZIhu4KKrV1xntCmZKGvLeMc+QNyIH237oAfunP607ls QwAAAAAAAA== --------------ms030401020205020706090103-- From rra@stanford.edu Thu Nov 15 23:20:25 2007 From: rra@stanford.edu (Russ Allbery) Date: Thu, 15 Nov 2007 15:20:25 -0800 Subject: [OpenAFS] Kerberos5 and afs In-Reply-To: <473CD3CB.90104@secure-endpoints.com> (Jeffrey Altman's message of "Thu, 15 Nov 2007 18:18:35 -0500") References: <473CBE05.601@msu.edu> <87fxz7jfr1.fsf@windlord.stanford.edu> <473CCF92.7030301@msu.edu> <87ve83hzoy.fsf@windlord.stanford.edu> <473CD3CB.90104@secure-endpoints.com> Message-ID: <87mytfhz9i.fsf@windlord.stanford.edu> Jeffrey Altman writes: > Russ Allbery wrote: >> Steve Devine writes: >>> This is my current kdc.conf entry: >>> supported_enctypes = des3-hmac-sha1:normal des-cbc-crc:normal >>> des-cbc-crc:v4 des-cbc-crc:afs3 >> you need to explicitly specify -e des-cbc-crc:normal when creating the >> krbtgt cross-realm keys. Otherwise you'll get a des3 key in your KDC >> and since Windows doesn't support des3, you'll lose. > Windows 2003 SP1 and later supports RC4-HMAC cross-realm keys. Yeah, I just didn't mention that because his kdc.conf doesn't. Adding rc4-hmac to your supported_enctypes is another alternative (although you still need to use -e, in this case with rc4-hmac, to limit the enctypes of the created key). -- Russ Allbery (rra@stanford.edu) From cclausen@acm.org Thu Nov 15 23:32:04 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Thu, 15 Nov 2007 17:32:04 -0600 Subject: [OpenAFS] Kerberos5 and afs References: <473CBE05.601@msu.edu> <87fxz7jfr1.fsf@windlord.stanford.edu> <473CCF92.7030301@msu.edu> Message-ID: <5B6F685E3C31463F9A5378D1E9BC4F09@CDCHOME> Steve Devine wrote: > Does the order of the enctypes listed in the kdc affect this? > This is my current kdc.conf entry: > supported_enctypes = des3-hmac-sha1:normal des-cbc-crc:normal > des-cbc-crc:v4 des-cbc-crc:afs3 > I'm not sure how to manipulate the kvno on the AD I currently have the following on a KDC with an AD domain trust: supported_enctypes = aes256-cts:normal aes128-cts:normal rc4-hmac:normal des3-hmac-sha1:normal des-cbc-crc:normal I suspect that you may want at least the rc4-hmac:normal in that list, as that is one of the enc_types that AD supports. I remember that I had no luck getting the trust to work when using specific enc_types in the -e option to ktadd. Completely omiting the "-e" seemed to work though. This could be something odd in my environment though. For instance, my cross-realm TGT has AES enc_types that are not actually supported by Windows: kadmin.local: getprinc krbtgt/ILLIGAL.UIUC.EDU@AD.UIUC.EDU Principal: krbtgt/ILLIGAL.UIUC.EDU@AD.UIUC.EDU Key: vno 1, Triple DES cbc mode with HMAC/sha1, no salt Key: vno 1, DES cbc mode with CRC-32, no salt Key: vno 1, AES-256 CTS mode with 96-bit SHA-1 HMAC, no salt Key: vno 1, AES-128 CTS mode with 96-bit SHA-1 HMAC, no salt Key: vno 1, ArcFour with HMAC/md5, no salt You can turn on RC4 for the realm trust using ktpass.exe. If you join #kerberos on Freenode IRC there are smart people in the channel who can help you with this. < References: <473CBE05.601@msu.edu> <87fxz7jfr1.fsf@windlord.stanford.edu> <473CCF92.7030301@msu.edu> <87ve83hzoy.fsf@windlord.stanford.edu> Message-ID: <473CE8E3.4060103@msu.edu> Russ Allbery wrote: > Steve Devine writes: > > >> Does the order of the enctypes listed in the kdc affect this? >> > > In my experience, the enctype list should match exactly. It doesn't > matter what order you list the enctypes in; if you have enctypes on the > krbtgt key that aren't present in Windows, you may lose. So, in this > case: > > >> This is my current kdc.conf entry: >> supported_enctypes = des3-hmac-sha1:normal des-cbc-crc:normal >> des-cbc-crc:v4 des-cbc-crc:afs3 >> > > you need to explicitly specify -e des-cbc-crc:normal when creating the > krbtgt cross-realm keys. Otherwise you'll get a des3 key in your KDC and > since Windows doesn't support des3, you'll lose. > > Ok that was it .. thanks to all. I hate to say how much time I spent on this. I am going to continue testing on this and I may post my results when I have something more coherent. Thanks again. /sd > Also, if you're entering a password to create this key, be very careful of > the salting algorithm. I think that you'll need to fix that on the > Windows side, since IIRC MIT Kerberos can't do the Windows salt but > Windows can do the MIT salt (if configured correctly), but it's been a > long time and I'm forgetting the details. > > >> I'm not sure how to manipulate the kvno on the AD >> > > It depends on the version of Windows. Sometimes you can't at all. And > regardless, since on the MIT side you can just use modprinc -kvno, it's > way easier to make the MIT side match Windows than vice versa. > > -- Steve Devine Storage Systems Academic Computing & Network Services Michigan State University 506 Computer Center East Lansing, MI 48824-1042 1-517-432-7327 Baseball is ninety percent mental; the other half is physical. - Yogi Berra From rtb@pclella.cern.ch Fri Nov 16 11:19:00 2007 From: rtb@pclella.cern.ch (Rainer Toebbicke) Date: Fri, 16 Nov 2007 12:19:00 +0100 Subject: [OpenAFS] namei fileserver runs into "too many open files" with plenty of threads Message-ID: <473D7CA4.4070802@pclella.cern.ch> This is a multi-part message in MIME format. --------------000508070207030108040609 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit The namei fileserver can run into "too many open files" since ih_open assumes that it can always open a new file before deciding another file should be closed when exceeding a threshold. The EMFILE error is simply returned to the caller. The "ad hoc" reserve of 64 free file descriptors can get depleted when more than 64 threads try to open a file simultaneously (the lock is dropped), just making the reserve bigger does not solve the problem as the number of threads grow. The attached patch now recovers from EMFILE and adjusts the threshold as to not make recovery the norm. It does not alter the "reserve", although setting it to 0 now works fine as well. Bcc'ed to openafs-bugs. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Rainer Toebbicke European Laboratory for Particle Physics(CERN) - Geneva, Switzerland Phone: +41 22 767 8985 Fax: +41 22 767 7155 --------------000508070207030108040609 Content-Type: text/plain; name="p_namei_emfile" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="p_namei_emfile" --- openafs/src/vol/ihandle.c.o144 2007-03-20 14:36:37.000000000 +0100 +++ openafs/src/vol/ihandle.c 2007-10-16 14:51:25.000000000 +0200 @@ -325,9 +325,10 @@ */ fdInUseCount += 1; IH_UNLOCK; +ih_open_retry: fd = OS_IOPEN(ihP); IH_LOCK; - if (fd == INVALID_FD) { + if (fd == INVALID_FD && (errno != EMFILE || fdLruHead == NULL) ) { fdInUseCount -= 1; IH_UNLOCK; return NULL; @@ -337,13 +338,23 @@ * we permit the number of open files to exceed fdCacheSize. * We only recycle open file descriptors when the number * of open files reaches the size of the cache */ - if (fdInUseCount > fdCacheSize && fdLruHead != NULL) { + if ((fdInUseCount > fdCacheSize || fd == INVALID_FD) && fdLruHead != NULL) { fdP = fdLruHead; assert(fdP->fd_status == FD_HANDLE_OPEN); DLL_DELETE(fdP, fdLruHead, fdLruTail, fd_next, fd_prev); DLL_DELETE(fdP, fdP->fd_ih->ih_fdhead, fdP->fd_ih->ih_fdtail, fd_ihnext, fd_ihprev); closeFd = fdP->fd_fd; + if (fd == INVALID_FD) { + fdCacheSize--; /* reduce in order to not run into here too often */ + DLL_INSERT_TAIL(fdP, fdAvailHead, fdAvailTail, fd_next, fd_prev); + fdP->fd_status = FD_HANDLE_AVAIL; + fdP->fd_ih = NULL; + fdP->fd_fd = INVALID_FD; + IH_UNLOCK; + OS_CLOSE(closeFd); + goto ih_open_retry; + } } else { if (fdAvailHead == NULL) { fdHandleAllocateChunk(); --------------000508070207030108040609-- From ronc@depauw.edu Fri Nov 16 14:32:55 2007 From: ronc@depauw.edu (Ron Croonenberg) Date: Fri, 16 Nov 2007 09:32:55 -0500 Subject: [OpenAFS] E212: Can't open file for writing In-Reply-To: <68C9A52FE1FF4D7C8716734E9E4D6567@CDCHOME> References: <4739A9B7.7030105@rampaginggeek.com> <4739AEF5.3090605@rzg.mpg.de> <4739F985.8000001@rampaginggeek.com> <473C652D.2060901@depauw.edu> <473CB3D3.7010607@depauw.edu> <68C9A52FE1FF4D7C8716734E9E4D6567@CDCHOME> Message-ID: <473DAA17.9010008@depauw.edu> Especially with large files ? Ron Christopher D. Clausen wrote: > Ron Croonenberg wrote: >> Uhm... I noticed that after a while (an hour or so)that problem >> "fixed itself" ? It looks like I copied the files there and it took a >> long while before it was actually there (even though sftp said it >> transferred the files)? > > Writes go to your AFS cache first, and then to the fileserver. You > could see a large hang at the "end" of a transfer as the data is flushed > out of the local cache and actually written to the AFS fileserver. > > < > -- ================================================================= Ron Croonenberg | | Phone: 1 765 658 4761 Lab Instructor & | Fax: 1 765 658 4732 Technology Coordinator | | Department of Computer Science | e-mail: ronc@DePauw.edu DePauw University | 275 Julian Science & Math Center | 602 South College Ave. | Greencastle, IN 46135 | ================================================================= From cclausen@acm.org Fri Nov 16 16:02:46 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Fri, 16 Nov 2007 10:02:46 -0600 Subject: [OpenAFS] E212: Can't open file for writing References: <4739A9B7.7030105@rampaginggeek.com> <4739AEF5.3090605@rzg.mpg.de> <4739F985.8000001@rampaginggeek.com> <473C652D.2060901@depauw.edu> <473CB3D3.7010607@depauw.edu> <68C9A52FE1FF4D7C8716734E9E4D6567@CDCHOME> <473DAA17.9010008@depauw.edu> Message-ID: <96427102FC7D419DA15D0A71FD9613B2@CDCHOME> Ron Croonenberg wrote: > Christopher D. Clausen wrote: >> Ron Croonenberg wrote: >>> Uhm... I noticed that after a while (an hour or so)that problem >>> "fixed itself" ? It looks like I copied the files there and it took >>> a long while before it was actually there (even though sftp said it >>> transferred the files)? >> >> Writes go to your AFS cache first, and then to the fileserver. You >> could see a large hang at the "end" of a transfer as the data is >> flushed out of the local cache and actually written to the AFS >> fileserver. > > Especially with large files ? Probably, yes. Larger files will take longer to write on the fileserver side. I've found that using a smaller cache size and using -memcache allows the cache to be flushed periodically and lessen these hangs on file close. < I have installed OpenAFS-1.4.5-Leopard on an intel machine and a G5 PPC. No problems on intel, but on the G5 I get the following error message on afs startup: Nov 15 09:15:08 maccmskel1 com.apple.SystemStarter[58]: /Library/ StartupItems/OpenAFS/OpenAFS: line 117: /var/db/openafs/etc/config/ afssettings: Bad CPU type in executable line 117 is $CONFIG/afssettings Despite this error, /afs is mounted at root level, and I have no problem with afs access from the terminal. Even "open afs_file" from the command line manages to open the file in the appropriate application. The problem is that the afs volume is not visible in the Finder. Have others had better experience with OpenAFS-1.4.5-Leopard on a PPC machine? Is there a work-around to make /afs visible in the Finder? Is there a fix for what appears to be the non-universal afssettings? Yours, Richard Kellogg From shadow@gmail.com Fri Nov 16 19:20:04 2007 From: shadow@gmail.com (Derrick Brashear) Date: Fri, 16 Nov 2007 14:20:04 -0500 Subject: [OpenAFS] OpenAFS-1.4.5-Leopard on G5 PPC - no afs volume in Finder In-Reply-To: <8EE784D7-EAD7-4FA5-BA0C-A2A34E99B350@mac.com> References: <8EE784D7-EAD7-4FA5-BA0C-A2A34E99B350@mac.com> Message-ID: On Nov 16, 2007 2:09 PM, Richard Kellogg wrote: > I have installed OpenAFS-1.4.5-Leopard on an intel machine and a G5 > PPC. No problems on intel, but on the G5 I get the following error > message on afs startup: > > Nov 15 09:15:08 maccmskel1 com.apple.SystemStarter[58]: /Library/ > StartupItems/OpenAFS/OpenAFS: line 117: /var/db/openafs/etc/config/ > afssettings: Bad CPU type in executable > > line 117 is $CONFIG/afssettings reget the installer? > Despite this error, /afs is mounted at root level, and I have no > problem with afs access from the terminal. Even "open afs_file" from > the command line manages to open the file in the appropriate > application. > Finder->Preferences->General->Connected Servers. This is not a bug. Apple changed how it worked. It's also documented at http://www.openafs.org/macos.html (next to the known issue line for it, follow the bug link) From richard_kellogg@mac.com Sat Nov 17 16:53:07 2007 From: richard_kellogg@mac.com (Richard Kellogg) Date: Sat, 17 Nov 2007 17:53:07 +0100 Subject: [OpenAFS] OpenAFS-1.4.5-Leopard on G5 PPC - no afs volume in Finder In-Reply-To: References: <8EE784D7-EAD7-4FA5-BA0C-A2A34E99B350@mac.com> Message-ID: <704C3603-8CFE-4957-B129-9CF862C9EC3B@mac.com> Thanks for the hint. This gives me a good work-around, but the issue is actually a bit more complex. I find that on an intel machine, the afs icon always appears in the Finder Sidebar, irrespective of any visibility setting in the Finder/Sidebar Preferences, and on PPC machines it never does. On both architectures the Connected Servers option in Finder Preferences/General/"Show these items on the Desktop" controls whether the afs icon appears on the Desktop. Two other solutions which work on either architecture: - enter "open /afs" from the command line - set "Computer" visible in Finder Sidebar Preferences & click on its icon Yours, Dick On Nov 16, 2007, at 8:20 PM, Derrick Brashear wrote: > On Nov 16, 2007 2:09 PM, Richard Kellogg > wrote: >> I have installed OpenAFS-1.4.5-Leopard on an intel machine and a G5 >> PPC. No problems on intel, but on the G5 I get the following error >> message on afs startup: >> >> Nov 15 09:15:08 maccmskel1 com.apple.SystemStarter[58]: /Library/ >> StartupItems/OpenAFS/OpenAFS: line 117: /var/db/openafs/etc/config/ >> afssettings: Bad CPU type in executable >> >> line 117 is $CONFIG/afssettings > > reget the installer? > >> Despite this error, /afs is mounted at root level, and I have no >> problem with afs access from the terminal. Even "open afs_file" from >> the command line manages to open the file in the appropriate >> application. >> > > Finder->Preferences->General->Connected Servers. > > This is not a bug. Apple changed how it worked. It's also documented > at http://www.openafs.org/macos.html (next to the known issue line for > it, follow the bug link) From stephen@physics.unc.edu Sat Nov 17 18:42:14 2007 From: stephen@physics.unc.edu (Stephen Joyce) Date: Sat, 17 Nov 2007 13:42:14 -0500 (EST) Subject: [OpenAFS] Anybody using BackupPC4afs ? Message-ID: On Thu, 15 Nov 2007, Hans Melgers wrote: > I was wondering if anybody on this list is using this for afs backups ? can > you recommend it ? pro's/cons ? > > http://www.physics.unc.edu/~stephen/BackupPC4AFS > > Well... I am. But I'm a bit biased. :-) I've been using it for over a year for backups and restores without problems. If your site is using BackupPC4AFS, please drop me a note and let me know. It gets quite a few hits and downloads on sourceforge, but I'm only aware of one other site using it. Feedback, ideas, bug reports, and especially bug fixes, are most welcome. I won't discuss the pros/cons of METHODS of backing up AFS (dumping volumes vs. backing up files). That discussion has been rehashed on the list several times and is available via google. BackupPC4AFS backs up volumes. Here are some pros: BackupPC itself is a fairly mature application and the logic for when and what to backup is damn-near bulletproof. BackupPC4AFS just uses BackupPC's logic and much of its code, but instead of connecting to PCs, it backs up and restores AFS volumes. In a worst-case scenario, restores can be performed outside BackupPC4AFS. They're in standard AFS vos dump format. The CGI makes it really easy to add new volumesets, see how much space backups are consuming, view the logs to make sure that everything went well, and perform restores. It's easy to configure email alerts if there are problems. It's much faster than dumping to tape. With modern linux servers and raids, 15-25 MB/s is normal over GigE. My fastest recorded dump is 43.28GB in 17.0 mins (42.5MB/s). It has some neat features like exponential expiry that make it very attractive. It's free (GPLed) and of course you get the source. If it doesn't perform as you want, you're free to make changes, and hopefully contribute them back. Here are some cons: There is presently only 1 developer (you can consider this another plea for testers, feedback, and, if you have skills, code). I'm only aware of one other site that's using it. There are definitely places where the code isn't pretty. The install is straightforward, but the configuration--while easier than the native AFS backups--can take a bit of time. The initial "breakin period" can be cumbersome. For example, I do a full of only 1 volumeset most nights (all other volumesets get incrementals that night). During transition it can become monotonous to add 1 new volumeset each day. If you have a very small cell, you don't need to do this, but for any cell with more than ~500GB, it's almost mandatory to spread your full dumps out over several days/nights in order to keep your backup window small. It's assumed that you're doing disk-to-disk backups. If you want a copy on tape, you have to use another product to grab the dumpfiles, use BackupPC's archive function to interface to another product, or have the other product dump its own copy. Currently the AFS restoration mods are only present in the English translation. Future plans: Upgrade to BackupPC 3.1 codebase once it's released. Most of its changes aren't applicable to AFS, but I want to stay up to date. Add the AFS restoration mods to the other language files; preferably with human help, but since I'm not a diplomat or journalist, with google as a last resort. Add an option to compress the dumpfiles to save disk space. Others? Hope this helps! Cheers, Stephen -- Stephen Joyce Systems Administrator P A N I C Physics & Astronomy Department Physics & Astronomy University of North Carolina at Chapel Hill Network Infrastructure voice: (919) 962-7214 and Computing fax: (919) 962-0480 http://www.panic.unc.edu Don't judge a book by its movie. From bunse@physik.uni-dortmund.de Mon Nov 19 15:53:03 2007 From: bunse@physik.uni-dortmund.de (Moritz Bunse) Date: Mon, 19 Nov 2007 16:53:03 +0100 (CET) Subject: [OpenAFS] Mac OS 10.5 Leopard's TimeMachine with AFS Message-ID: Hello, I recently updated my Mac OS to Leopard and installed OpenAFS. I can browse my AFS-cell. Now I want to use an AFS-directory as backup directory for time machine. I entered defaults write com.apple.systempreferences TMShowUnsupportedNetworkVolumes 1 in the command line and now /afs is shown in the backup volume selection in Time Machine. I can select it, but the problem I think is, that I do not have write access to /afs, only to /afs/cellname/user/username. How can I mount this directory as a network device? There is software at Stanford ( http://www.stanford.edu/services/openafs/mac/index.html ) , but it is not downloadable! Thanks in advance, Moritz From stephen@physics.unc.edu Mon Nov 19 16:10:52 2007 From: stephen@physics.unc.edu (Stephen Joyce) Date: Mon, 19 Nov 2007 11:10:52 -0500 (EST) Subject: [OpenAFS] Mac OS 10.5 Leopard's TimeMachine with AFS In-Reply-To: References: Message-ID: On Mon, 19 Nov 2007, Moritz Bunse wrote: > Hello, > > I recently updated my Mac OS to Leopard and installed OpenAFS. I can > browse my AFS-cell. Now I want to use an AFS-directory as backup directory > for time machine. I entered > > defaults write com.apple.systempreferences TMShowUnsupportedNetworkVolumes > 1 > > in the command line and now /afs is shown in the backup volume selection > in Time Machine. I can select it, but the problem I think is, that I do > not have write access to /afs, only to /afs/cellname/user/username. How > can I mount this directory as a network device? > There is software at Stanford ( > http://www.stanford.edu/services/openafs/mac/index.html ) , but it is not > downloadable! Will Time Machine even use a location in AFS for its backend storage? Forgive me, as I haven't played with 10.5 yet, but only read about it... doesn't Time Machine use hardlinks between directories to build its backup directories? I seem to recall that AFS doesn't support hard links between files in different directories...? (This has disappointed me once or twice). Cheers, Stephen -- Stephen Joyce Systems Administrator P A N I C Physics & Astronomy Department Physics & Astronomy University of North Carolina at Chapel Hill Network Infrastructure voice: (919) 962-7214 and Computing fax: (919) 962-0480 http://www.panic.unc.edu Don't judge a book by its movie. From shadow@gmail.com Mon Nov 19 16:11:41 2007 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 19 Nov 2007 11:11:41 -0500 Subject: [OpenAFS] Mac OS 10.5 Leopard's TimeMachine with AFS In-Reply-To: References: Message-ID: As far as I know Time Machine wants hfs+ semantics, which AFS doesn't have, so it's not going to work, period. On Nov 19, 2007 10:53 AM, Moritz Bunse wrote: > Hello, > > I recently updated my Mac OS to Leopard and installed OpenAFS. I can > browse my AFS-cell. Now I want to use an AFS-directory as backup directory > for time machine. I entered > > defaults write com.apple.systempreferences TMShowUnsupportedNetworkVolumes > 1 > > in the command line and now /afs is shown in the backup volume selection > in Time Machine. I can select it, but the problem I think is, that I do > not have write access to /afs, only to /afs/cellname/user/username. How > can I mount this directory as a network device? > There is software at Stanford ( > http://www.stanford.edu/services/openafs/mac/index.html ) , but it is not > downloadable! > > Thanks in advance, > Moritz > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > From rob@nofocus.org Mon Nov 19 16:20:00 2007 From: rob@nofocus.org (Rob Banz) Date: Mon, 19 Nov 2007 11:20:00 -0500 Subject: [OpenAFS] Mac OS 10.5 Leopard's TimeMachine with AFS In-Reply-To: References: Message-ID: > > > Forgive me, as I haven't played with 10.5 yet, but only read about > it... doesn't Time Machine use hardlinks between directories to > build its backup directories? I seem to recall that AFS doesn't > support hard links between files in different directories...? (This > has disappointed me once or twice). Hard links are generally disappointing, and cause more trouble than they're worth. 10 years of running AFS have trained me to avoid them at all costs, to my benefit ;) -rob From Hagedorn@uni-koeln.de Mon Nov 19 16:24:37 2007 From: Hagedorn@uni-koeln.de (Sebastian Hagedorn) Date: Mon, 19 Nov 2007 17:24:37 +0100 Subject: [OpenAFS] Mac OS 10.5 Leopard's TimeMachine with AFS In-Reply-To: References: Message-ID: <28140C6FA8B66D8123F146A9@tyrion.rrz.uni-koeln.de> --==========276A1283B784A6ABBE7F========== Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: quoted-printable Content-Disposition: inline --On 19. November 2007 11:11:41 -0500 Derrick Brashear =20 wrote: > As far as I know Time Machine wants hfs+ semantics, which AFS doesn't > have, so it's not going to work, period. What you don't seem to know is that Time Machine uses disk images on=20 non-HFS filespace. So it *should* work! This feature is initially disabled=20 by Apple and unsupported. --=20 .:.Sebastian Hagedorn - RZKR-R1 (Geb=C3=A4ude 52), Zimmer 18.:. Zentrum f=C3=BCr angewandte Informatik - Universit=C3=A4tsweiter Service = RRZK .:.Universit=C3=A4t zu K=C3=B6ln / Cologne University - =E2=9C=86 = +49-221-478-5587.:. .:.:.:.Skype: shagedorn.:.:.:. --==========276A1283B784A6ABBE7F========== Content-Type: application/pgp-signature Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (Darwin) iD8DBQFHQbjFGXsGmU0QW0URAnB5AJ9F/X82E89Qcq6LAdHPcQvLHMZQ0gCdE0Fn ZshvdjkCrmRGtMnLcS3D4sE= =MwVf -----END PGP SIGNATURE----- --==========276A1283B784A6ABBE7F==========-- From shadow@gmail.com Mon Nov 19 16:36:07 2007 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 19 Nov 2007 11:36:07 -0500 Subject: [OpenAFS] Mac OS 10.5 Leopard's TimeMachine with AFS In-Reply-To: <28140C6FA8B66D8123F146A9@tyrion.rrz.uni-koeln.de> References: <28140C6FA8B66D8123F146A9@tyrion.rrz.uni-koeln.de> Message-ID: On Nov 19, 2007 11:24 AM, Sebastian Hagedorn wrote: > --On 19. November 2007 11:11:41 -0500 Derrick Brashear > wrote: > > > As far as I know Time Machine wants hfs+ semantics, which AFS doesn't > > have, so it's not going to work, period. > > What you don't seem to know is that Time Machine uses disk images on > non-HFS filespace. So it *should* work! This feature is initially disabled > by Apple and unsupported. > I know only that all the people that used the unsupported functionality told me they had data loss across disconnecting the storage, so I am treating it as "doesn't exist" From scs@umich.edu Mon Nov 19 20:41:54 2007 From: scs@umich.edu (Steve Simmons) Date: Mon, 19 Nov 2007 15:41:54 -0500 Subject: [OpenAFS] 1.2.X and 1.4.5 Interoperation In-Reply-To: References: Message-ID: <8F75E636-EF53-4D20-A916-8B3EBB8EEA8F@umich.edu> On Nov 15, 2007, at 11:37 AM, Kevin Scott Sumner wrote: > Is there ANYTHING that could POSSIBLY break when we have a mixture > of these 1.2 and 1.4 servers running together? Version issues, OS > issues, Endian issues, etc. I did pretty much what you describe going from old 1.2 to early 1.4 and don't recall any significant problems. On the other hand, our DB server was a separate host, so we didn't quite represent your environment. From stephen@physics.unc.edu Mon Nov 19 21:53:37 2007 From: stephen@physics.unc.edu (Stephen Joyce) Date: Mon, 19 Nov 2007 16:53:37 -0500 (EST) Subject: [OpenAFS] 1.2.X and 1.4.5 Interoperation In-Reply-To: <8F75E636-EF53-4D20-A916-8B3EBB8EEA8F@umich.edu> References: <8F75E636-EF53-4D20-A916-8B3EBB8EEA8F@umich.edu> Message-ID: On Mon, 19 Nov 2007, Steve Simmons wrote: > On Nov 15, 2007, at 11:37 AM, Kevin Scott Sumner wrote: > >> Is there ANYTHING that could POSSIBLY break when we have a mixture of these >> 1.2 and 1.4 servers running together? Version issues, OS issues, Endian >> issues, etc. > > I did pretty much what you describe going from old 1.2 to early 1.4 and don't > recall any significant problems. On the other hand, our DB server was a > separate host, so we didn't quite represent your environment. Our DB servers are separate from our fileservers. I think what Kevin was asking was in the transition from 1.2.10 on solaris/sparc to 1.4.5 on debian/x86, there will be times when there is a mixed environment where all the DB servers are not running the same OS or even on the same architecture. Is this OK or should all the DB servers be upgraded at once (which is logistically more difficult). Cheers, Stephen -- Stephen Joyce Systems Administrator P A N I C Physics & Astronomy Department Physics & Astronomy University of North Carolina at Chapel Hill Network Infrastructure voice: (919) 962-7214 and Computing fax: (919) 962-0480 http://www.panic.unc.edu Don't judge a book by its movie. From rra@stanford.edu Mon Nov 19 21:55:17 2007 From: rra@stanford.edu (Russ Allbery) Date: Mon, 19 Nov 2007 13:55:17 -0800 Subject: [OpenAFS] 1.2.X and 1.4.5 Interoperation In-Reply-To: (Stephen Joyce's message of "Mon\, 19 Nov 2007 16\:53\:37 -0500 \(EST\)") References: <8F75E636-EF53-4D20-A916-8B3EBB8EEA8F@umich.edu> Message-ID: <87r6ilyk6y.fsf@windlord.stanford.edu> Stephen Joyce writes: > I think what Kevin was asking was in the transition from 1.2.10 on > solaris/sparc to 1.4.5 on debian/x86, there will be times when there is > a mixed environment where all the DB servers are not running the same OS > or even on the same architecture. Is this OK or should all the DB > servers be upgraded at once (which is logistically more difficult). This is okay. The database format hasn't changed in years. (The one caveat is that if you plan to enable supergroups as part of the process, that's fine, but don't actually create or use them until you've upgraded all of your DB servers.) -- Russ Allbery (rra@stanford.edu) From ghc4+@pitt.edu Tue Nov 20 15:14:21 2007 From: ghc4+@pitt.edu (George Cebulka) Date: Tue, 20 Nov 2007 10:14:21 -0500 Subject: [OpenAFS] Solaris 10, openssh, pam.conf and afs Message-ID: <4742F9CD.2080203@pitt.edu> Hello All, I am trying to get Openafs 1.4.5, Openssh 4.7 to work with pam authentication on a Sunfire 4200 running Solaris 10 x86 8/07. I am able to compile, install and run afs and ssh. However, I am unsure of what changes I need to make to /etc/pam.conf to allow a user to log into the machine via ssh and get an afs token without having to do a separate klog. Can any of you Solaris 10 types provide some pam.conf examples? Thanks in advance, George Cebulka From deengert@anl.gov Tue Nov 20 15:36:22 2007 From: deengert@anl.gov (Douglas E. Engert) Date: Tue, 20 Nov 2007 09:36:22 -0600 Subject: [OpenAFS] Solaris 10, openssh, pam.conf and afs In-Reply-To: <4742F9CD.2080203@pitt.edu> References: <4742F9CD.2080203@pitt.edu> Message-ID: <4742FEF6.8070604@anl.gov> George Cebulka wrote: > Hello All, > I am trying to get Openafs 1.4.5, Openssh 4.7 to work with pam > authentication on a Sunfire 4200 running Solaris 10 x86 8/07. I am able > to compile, install and run afs and ssh. However, I am unsure of what > changes I need to make to /etc/pam.conf to allow a user to log into the > machine via ssh and get an afs token without having to do a separate klog. > Can any of you Solaris 10 types provide some pam.conf examples? First of all, is there some reason you are using the OpenSSH rather then Solaris 10 ssh and sshd? (It works well on sparc, I assume it will on x86.) The Sun ssh has kerberos, gssapi, and even gssapi-keyex. It can call pam, and can call pam_afs_session to get pag and tokens. (We are using pam_afs2.) Google for pam_afs_session. > Thanks in advance, > George Cebulka > > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > > -- Douglas E. Engert Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439 (630) 252-5444 From phalenor@gmail.com Sat Nov 24 15:02:47 2007 From: phalenor@gmail.com (Andrew Cobaugh) Date: Sat, 24 Nov 2007 10:02:47 -0500 Subject: [OpenAFS] issue with Fedora 8 and retaining tokens after graphical login Message-ID: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> In the past (up until Fedora 8), afs has always Just Worked. The supplied pam_krb5 was able to obtain a tgt and tokens, both with sshd and when logging in through things like gdm. I recently upgraded from 7 to 8. Logging in through sshd works perfectly, but gdm stopped working. I get a tgt, but not tokens. Odd thing is if I set pam_krb5 to debug in /etc/pam.d/system-auth, I get all kinds of debug messages in /var/log/secure as expected when going in through ssh, but I get none of this with gdm, so it's almost as if gdm is skipping over pam, which doesn't make sense. So, has anyone had trouble with Fedora 8 and afs in this respect? I've downgraded to the pam_krb5 from Fedora 7 (pam_krb5-2.2.11-1), and that hasn't fixed it. I suspect either something with gdm, or something weird they're doing because of all of this ConsoleKit stuff... Ideas? -- Andy Cobaugh phalenor@gmail.com From W.J.Murray@rl.ac.uk Sat Nov 24 16:27:24 2007 From: W.J.Murray@rl.ac.uk (William Murray) Date: Sat, 24 Nov 2007 16:27:24 +0000 Subject: [OpenAFS] issue with Fedora 8 and retaining tokens after graphical login In-Reply-To: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> References: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> Message-ID: <1195921644.26707.6.camel@BillMurray> Hi Andrew, 'afs has just worked?' with previous fedora? Well...I'd say Fedora has been on latest kernels and AFS is struggling to keep up. However, using openafs-client from atrpms it seems to be pretty good with f8. At least at work.. unfortunately, at home, via a tunnel and NAT etc, it doesn't seem happy... Bill On Sat, 2007-11-24 at 10:02 -0500, Andrew Cobaugh wrote: > In the past (up until Fedora 8), afs has always Just Worked. The > supplied pam_krb5 was able to obtain a tgt and tokens, both with sshd > and when logging in through things like gdm. > > I recently upgraded from 7 to 8. Logging in through sshd works > perfectly, but gdm stopped working. I get a tgt, but not tokens. Odd > thing is if I set pam_krb5 to debug in /etc/pam.d/system-auth, I get > all kinds of debug messages in /var/log/secure as expected when going > in through ssh, but I get none of this with gdm, so it's almost as if > gdm is skipping over pam, which doesn't make sense. > > So, has anyone had trouble with Fedora 8 and afs in this respect? I've > downgraded to the pam_krb5 from Fedora 7 (pam_krb5-2.2.11-1), and that > hasn't fixed it. I suspect either something with gdm, or something > weird they're doing because of all of this ConsoleKit stuff... > > Ideas? > From sxw@inf.ed.ac.uk Sat Nov 24 19:29:23 2007 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Sat, 24 Nov 2007 19:29:23 +0000 Subject: [OpenAFS] issue with Fedora 8 and retaining tokens after graphical login In-Reply-To: <1195921644.26707.6.camel@BillMurray> References: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> <1195921644.26707.6.camel@BillMurray> Message-ID: On 24 Nov 2007, at 16:27, William Murray wrote: > 'afs has just worked?' with previous fedora? It certainly "Just works" for us at Edinburgh. We roll our own kernels for Fedora, though. > Well...I'd say > Fedora has been on latest kernels and AFS is struggling to keep up. There's a couple of problems. Some new upstream kernel releases contain changes which break OpenAFS - these changes are some times backported into "earlier" Fedora releases, so a lack of change in kernel version doesn't always indicate a lack of API changes. When this happens, new deltas tend to make into into OpenAFS CVS pretty rapidly. Support for these as RPMs (for both the official OpenAFS RPMS, and the atrpms ones), then requires either a new OpenAFS release, or someone pulling out the delta, adding it to the RPM, and testing it. This all takes time, and so can cause these changes to lag behind the availability of new kernel RPMs. The second problem is simpler RPM rebuilds - at the moment there are a number of manual steps in this process - whilst the availability of new Fedora kernels is now automatically detected, doing the rebuild requires someone (often me) to load up the appropriate VM, kick of the builds, upload the RPMs to the appropriate location, and ask someone else to push them out to the download site. This all depends on time and availability, which is often in scarce supply. There are plans afoot to improve this so that the build process will be entirely automated. > However, using openafs-client from atrpms it seems to be pretty > good > with f8. At least at work.. The drawback to the atrpms rpms is that they're using the FHS style paths. > unfortunately, at home, via a tunnel and NAT etc, it doesn't seem > happy... I'm running OpenAFS 1.4.5 on Fedora 8 from home, behind a NAT, and all seems fine. What problems are you seeing? Simon. From sxw@inf.ed.ac.uk Sat Nov 24 19:46:52 2007 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Sat, 24 Nov 2007 19:46:52 +0000 Subject: [OpenAFS] issue with Fedora 8 and retaining tokens after graphical login In-Reply-To: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> References: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> Message-ID: On 24 Nov 2007, at 15:02, Andrew Cobaugh wrote: > In the past (up until Fedora 8), afs has always Just Worked. The > supplied pam_krb5 was able to obtain a tgt and tokens, both with sshd > and when logging in through things like gdm. We've always used either pam_afs2 or pam_afs_session to handle AFS tokens, so I can't comment directly on the RedHat pam_krb5 module. One common problem, however, is if you are calling pam_keyinit in the session layer. This resets the default keyring, losing any tokens that an auth stack module has inserted into the keyring during the authenticate operation. I don't know enough about how the RedHat module works to say if it can work around this - but I'd strongly suggest that you look at Russ's pam_krb5 and pam_afs_session modules (available from http://www.eyrie.org/) which will do the right thing in this, and many other, cases. Cheers, Simon. From openafs-info Sat Nov 24 20:41:32 2007 From: openafs-info (Axel Thimm) Date: Sat, 24 Nov 2007 22:41:32 +0200 Subject: [OpenAFS] Re: issue with Fedora 8 and retaining tokens after graphical login In-Reply-To: References: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> <1195921644.26707.6.camel@BillMurray> Message-ID: <20071124204132.GB28686@puariko.nirvana> --DBIVS5p969aUjpLe Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Hi, On Sat, Nov 24, 2007 at 07:29:23PM +0000, Simon Wilkinson wrote: > There's a couple of problems. Some new upstream kernel releases > contain changes which break OpenAFS - these changes are some times > backported into "earlier" Fedora releases, so a lack of change in > kernel version doesn't always indicate a lack of API changes. When > this happens, new deltas tend to make into into OpenAFS CVS pretty > rapidly. Support for these as RPMs (for both the official OpenAFS > RPMS, and the atrpms ones), then requires either a new OpenAFS > release, or someone pulling out the delta, adding it to the RPM, and > testing it. This all takes time, and so can cause these changes to > lag behind the availability of new kernel RPMs. Sounds like a good angle to find some common momentum and share some work :) > > However, using openafs-client from atrpms it seems to be pretty > > good with f8. At least at work.. >=20 > The drawback to the atrpms rpms is that they're using the FHS style =20 > paths. Why is that a drawback (the question is sincere). While perhaps long time users will need to rethink the paths they have to do so anyway on a Linux system compared to AIX, Solaris etc. And any openafs acceptance into any Linux distribution has better cards to be FHS than not. If openafs every makes it into RHEL/Fedora proper it will have to be FHS style. But let's make everybody happy: How about an rpmbuild time switch to allow anyone to rebuild with their favourite path style. I would still want to see FHS being the default, but transarc style is just a one-line change in the rpmbuild specfile. Just looking for synergy and a common openafs rpm design. Which could very well be more than Fedora/RHEL (but I admit not being a SuSE or Mandriva expert, so someone would have to step in to aid in this part). :) --=20 Axel.Thimm at ATrpms.net --DBIVS5p969aUjpLe Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) iD8DBQFHSIx7QBVS1GOamfERAsqJAJ0VwkWv6s13rqCTbjpnirOBXvMAqQCfQkzY rePD2DB/HJVb72k0vPIRXwI= =EHDN -----END PGP SIGNATURE----- --DBIVS5p969aUjpLe-- From shadow@gmail.com Sun Nov 25 01:19:38 2007 From: shadow@gmail.com (Derrick Brashear) Date: Sat, 24 Nov 2007 20:19:38 -0500 Subject: [OpenAFS] Re: issue with Fedora 8 and retaining tokens after graphical login In-Reply-To: <20071124204132.GB28686@puariko.nirvana> References: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> <1195921644.26707.6.camel@BillMurray> <20071124204132.GB28686@puariko.nirvana> Message-ID: > > The drawback to the atrpms rpms is that they're using the FHS style > > paths. > > Why is that a drawback (the question is sincere). While perhaps long > time users will need to rethink the paths they have to do so anyway on > a Linux system compared to AIX, Solaris etc. And any openafs > acceptance into any Linux distribution has better cards to be FHS than > not. If openafs every makes it into RHEL/Fedora proper it will have to > be FHS style. If there's a chance of that, fine, but otherwise, change for the sake of change is dumb. > But let's make everybody happy: How about an rpmbuild time switch to > allow anyone to rebuild with their favourite path style. I would still > want to see FHS being the default, but transarc style is just a > one-line change in the rpmbuild specfile. Likewise, the people who are using Transarc style would want that to be default :) > Just looking for synergy and a common openafs rpm design. Which could > very well be more than Fedora/RHEL (but I admit not being a SuSE or > Mandriva expert, so someone would have to step in to aid in this > part). :) In theory we already have it. I've built the release SRPM on SLES. From Axel.Thimm@ATrpms.net Sun Nov 25 07:14:23 2007 From: Axel.Thimm@ATrpms.net (Axel Thimm) Date: Sun, 25 Nov 2007 09:14:23 +0200 Subject: [OpenAFS] Re: issue with Fedora 8 and retaining tokens after graphical login In-Reply-To: References: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> <1195921644.26707.6.camel@BillMurray> <20071124204132.GB28686@puariko.nirvana> Message-ID: <20071125071423.GA4543@puariko.nirvana> --cWoXeonUoKmBZSoM Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sat, Nov 24, 2007 at 08:19:38PM -0500, Derrick Brashear wrote: > > And any openafs acceptance into any Linux distribution has better > > cards to be FHS than not. If openafs ever makes it into > > RHEL/Fedora proper it will have to be FHS style. >=20 > If there's a chance of that, fine, but otherwise, change for the > sake of change is dumb. I can't speak for RH, but I know there are some customers that want it and push towards this happening. Having something that is ready to be integrated makes it easier. But if/when RH wants to do that you'll know before anyone else - typically RH contacts lead deveopers of a project in private for integration purposes. (Although if it comes that far the kernel module bits will be merged into RH' kernel rpm and not shipped as a separate package, there is a kernel-module-outside-the-kernel-rpm phobia among the kernel packagers at RH) --=20 Axel.Thimm at ATrpms.net --cWoXeonUoKmBZSoM Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) iD8DBQFHSSDPQBVS1GOamfERApvUAJ91TkzWbhme7kqjI3TUu+VrswnAhgCdFR57 UuyjJuyJzgH/bfwrMy2Uybg= =najh -----END PGP SIGNATURE----- --cWoXeonUoKmBZSoM-- From hans@enem.nl Mon Nov 26 15:18:24 2007 From: hans@enem.nl (ENEM | Hans Melgers) Date: Mon, 26 Nov 2007 16:18:24 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <473B0970.2080009@secure-endpoints.com> Message-ID: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> I dont thing the problem is in the afs client. We found this problem only o= ccurs when opening a file that already has been opened by another user. We = upgraded to office 2003 and it takes alot of time before word gives any res= ponse. This only seems to happen on the "original" pc's, when trying from our own = pc everything works fine, word opens in read-only mode but that doesnt happ= en on these pc's ? Maybe it has something to do with the fact that they are also running the n= etware client. Are there any known issues with this setup ? Should we run a= fs before netware ? Maybe check some settings ? -----Original Message----- From: Jeffrey Altman [mailto:jaltman@secure-endpoints.com] Sent: woensdag 14 november 2007 15:43 To: Hans Melgers Cc: openafs-info@openafs.org Subject: Re: [OpenAFS] MS Word crashing when opening files, 1.5.27 client Hans: Please follow the debugging section of the OpenAFS for Windows Release Note= s. In particular, use the SysInternals' Process Monitor to log the request= s from Word so that you can determine which requests are failing with 1.5.2= 7 that succeed with 1.5.25. One change that went into 1.5.27 is support for retrying requests when the = file server responds to the client with EWOULDBLOCK / EAGAIN. In prior rel= eases the client would simply fail the request. In 1.5.27 the client retri= es the request every two seconds up to 45 seconds until the request complet= es or the timeout is reached. In the trace log it is the cm_Analyze() messages will include the error number. Possible error number= s to look for include: UAEWOULDBLOCK 49733416L UAEAGAIN 49733386L EAGAIN 11L Jeffrey Altman Hans Melgers wrote: > > Hello, > > We just installed some XP machines with the latest 1.5.27 afs client. > On several of them Word crashes badly when trying to open a .doc file > on afs. Most of them running office XP but also seen on office 2003. > > I run 1.5.25 and havent seen the problem. Could this be a bug in the > latest win client ? acl's are ok. Server is 1.4.4 on freebsd 6.2 > Anyone else seeing this behaviour ? > > In Filelog i only see some callback issues: > > Wed Nov 14 10:56:32 2007 CB: ProbeUuid for 213.46.24.222:17980 failed > -01 Wed Nov 14 10:57:11 2007 CB: ProbeUuid for 213.46.24.222:17982 > failed -01 Wed Nov 14 10:58:08 2007 CB: ProbeUuid for > 213.46.24.222:17987 failed -01 Wed Nov 14 10:58:18 2007 CB: ProbeUuid > for 213.46.24.222:17976 failed -01 Wed Nov 14 11:01:25 2007 CB: > ProbeUuid for 213.46.24.222:17995 failed -01 > > > Any help appreciated. > > Hans From jaltman@secure-endpoints.com Mon Nov 26 15:35:39 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Mon, 26 Nov 2007 10:35:39 -0500 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> Message-ID: <474AE7CB.3040504@secure-endpoints.com> Hans: Again I must point you at the debugging section of the release notes. If you would gather the necessary data the problem could be examined based upon the actual requests that are being issued by the software and the responses that are provided by the AFS client instead of attempting to guess at the behavior by looking at the entire system (Office XP, Windows, AFS client, AFS server) as a black box. While it is possible the Netware client might be involved in some way I would find that unlikely. The application is requesting a lock and the file server is unable to grant it because another client already has it. In this situation, the file server will report EAGAIN to the client and the AFS client will retry in the hopes of being able to obtain the lock. Perhaps the application is timing out the request before the AFS client returns. I don't know for sure because you have not obtained the necessary data. Jeffrey Altman ENEM | Hans Melgers wrote: > > > I dont thing the problem is in the afs client. We found this problem only occurs when opening a file that already has been opened by another user. We upgraded to office 2003 and it takes alot of time before word gives any response. > This only seems to happen on the "original" pc's, when trying from our own pc everything works fine, word opens in read-only mode but that doesnt happen on these pc's ? > > Maybe it has something to do with the fact that they are also running the netware client. Are there any known issues with this setup ? Should we run afs before netware ? Maybe check some settings ? > > > -----Original Message----- > From: Jeffrey Altman [mailto:jaltman@secure-endpoints.com] > Sent: woensdag 14 november 2007 15:43 > To: Hans Melgers > Cc: openafs-info@openafs.org > Subject: Re: [OpenAFS] MS Word crashing when opening files, 1.5.27 client > > Hans: > > Please follow the debugging section of the OpenAFS for Windows Release Notes. In particular, use the SysInternals' Process Monitor to log the requests from Word so that you can determine which requests are failing with 1.5.27 that succeed with 1.5.25. > > One change that went into 1.5.27 is support for retrying requests when the file server responds to the client with EWOULDBLOCK / EAGAIN. In prior releases the client would simply fail the request. In 1.5.27 the client retries the request every two seconds up to 45 seconds until the request completes or the timeout is reached. In the trace log it is the > cm_Analyze() messages will include the error number. Possible error numbers to look for include: > > UAEWOULDBLOCK 49733416L > UAEAGAIN 49733386L > EAGAIN 11L > > > Jeffrey Altman > > > > > Hans Melgers wrote: >> Hello, >> >> We just installed some XP machines with the latest 1.5.27 afs client. >> On several of them Word crashes badly when trying to open a .doc file >> on afs. Most of them running office XP but also seen on office 2003. >> >> I run 1.5.25 and havent seen the problem. Could this be a bug in the >> latest win client ? acl's are ok. Server is 1.4.4 on freebsd 6.2 >> Anyone else seeing this behaviour ? >> >> In Filelog i only see some callback issues: >> >> Wed Nov 14 10:56:32 2007 CB: ProbeUuid for 213.46.24.222:17980 failed >> -01 Wed Nov 14 10:57:11 2007 CB: ProbeUuid for 213.46.24.222:17982 >> failed -01 Wed Nov 14 10:58:08 2007 CB: ProbeUuid for >> 213.46.24.222:17987 failed -01 Wed Nov 14 10:58:18 2007 CB: ProbeUuid >> for 213.46.24.222:17976 failed -01 Wed Nov 14 11:01:25 2007 CB: >> ProbeUuid for 213.46.24.222:17995 failed -01 >> >> >> Any help appreciated. >> >> Hans > > > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info From lara.lloret.iglesias@cern.ch Fri Nov 16 08:48:07 2007 From: lara.lloret.iglesias@cern.ch (lara_maktub) Date: Fri, 16 Nov 2007 00:48:07 -0800 (PST) Subject: [OpenAFS] AFS-modified login utility Message-ID: <13788944.post@talk.nabble.com> Hello everybody! I would like to ask you something, I don't know a lot about AFS, i've just installed it in several new clients following some instructions that I found in Internet, and it works fine. The problem is that I would like to log in and authenticate in one step, but I don't know how. Each time, when I log in I have to write "klog" . What is strange, is that in one of ours computers that was already installed before I came here, it makes it automatically, but in the new ones it doesn't and I can't find any different configuration.. Can anyone help me??? Thank you very much! Lara -- View this message in context: http://www.nabble.com/AFS-modified-login-utility-tf4819767.html#a13788944 Sent from the OpenAFS - General mailing list archive at Nabble.com. From sjp46@cornell.edu Fri Nov 16 22:17:30 2007 From: sjp46@cornell.edu (Steven James Pelley) Date: Fri, 16 Nov 2007 17:17:30 -0500 (EST) Subject: [OpenAFS] changing server ip address problems. Message-ID: <39466.128.253.86.133.1195251450.squirrel@webmail.cornell.edu> I recently changed the ip address of my afs server. This server is the only component of the afs server- it runs the database servers, file server, and kerberos 5. I had tested this before, so that after I changed the ip address, config files, and dns info I ran into the problem that a 'vos listvldb' still listed the old ip address. After a 'bos restart fs' everything worked fine and I got on with life. This time, however, restarting the fs server does not fix my problem. The entries of 'vos listvldb' still show the old ip address. I am rather certain that it is contacting the server, as a bos stop fs followed by a bos status shows that the fs server stops (and I can start it up again). Is there any other way to update the file server to reflect the change in ip-address? I'm using the openafs build of openafs-1.4.4 on fedora 7. Thanks, Steven Pelley From dys@mac.com Sat Nov 17 01:03:41 2007 From: dys@mac.com (Darren Patterson) Date: Fri, 16 Nov 2007 17:03:41 -0800 Subject: [OpenAFS] rhel3 openafs 1.4.5 spec patch Message-ID: <473E3DED.9020006@mac.com> It seems that the libafsauthent and libafsrpc libraries aren't being removed correctly when rebuilding openafs 1.4.5 from SRPM. This throws an error before the RPMs are generated. Below is a patch for the spec file. -darren --- openafs.spec.orig 2007-11-16 16:36:01.000000000 -0800 +++ openafs.spec 2007-11-16 16:36:28.000000000 -0800 @@ -1163,8 +1163,8 @@ done %if !%{build_authlibs} -rm -f $RPM_BUILD_ROOT%{_libdir}/libafsauthent.so -rm -f $RPM_BUILD_ROOT%{_libdir}/libafsrpc.so +rm -f $RPM_BUILD_ROOT%{_libdir}/libafsauthent.so* +rm -f $RPM_BUILD_ROOT%{_libdir}/libafsrpc.so* %endif %endif From deengert@anl.gov Mon Nov 26 22:26:38 2007 From: deengert@anl.gov (Douglas E. Engert) Date: Mon, 26 Nov 2007 16:26:38 -0600 Subject: [OpenAFS] AFS-modified login utility In-Reply-To: <13788944.post@talk.nabble.com> References: <13788944.post@talk.nabble.com> Message-ID: <474B481E.7070502@anl.gov> lara_maktub wrote: > Hello everybody! > > I would like to ask you something, I don't know a lot about AFS, i've just > installed it in several new clients following some instructions that I found > in Internet, and it works fine. The problem is that I would like to log in > and authenticate in one step, but I don't know how. Each time, when I log in > I have to write "klog" . > What is strange, is that in one of ours computers that was already installed > before I came here, it makes it automatically, but in the new ones it > doesn't and I can't find any different configuration.. > Can anyone help me??? Have you tried you site's help desk? http://consult.cern.ch/service/afs/ > > Thank you very much! > > Lara -- Douglas E. Engert Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439 (630) 252-5444 From rns@unimelb.edu.au Tue Nov 27 06:17:34 2007 From: rns@unimelb.edu.au (Robert Sturrock) Date: Tue, 27 Nov 2007 17:17:34 +1100 Subject: [OpenAFS] OpenAFS 1.4.5 on Solaris 10/x86 - spurious pid(s) in ctstat and ocassional kernel panics In-Reply-To: <20071127011436.GC7430@unimelb.edu.au> References: <20071127011436.GC7430@unimelb.edu.au> Message-ID: <20071127061734.GB8139@unimelb.edu.au> Hi All. We're noticing some problems on some new Solaris 10U4 (x86, kernel 120012-14) machines we've deployed as OpenAFS 1.4.5 clients. There seem to be spurious process-ids showing up when we run "ctstat -v": $ ctstat -vi 48 CTID ZONEID TYPE STATE HOLDER EVENTS QTIME NTIME 48 0 process owned 7 0 - - cookie: 0x20 informative event set: none critical event set: core signal hwerr empty fatal event set: none parameter set: inherit regent member processes: 335 336 338 343 344 345 346 347 1936749667 inherited contracts: none pids 335-347 are afsd pides, but 1936749667 is obviously not valid! We run AFS via SMF. Shutting down AFS seems to result in an infinite loop and we see these messages in the SMF log: # tail /var/svc/log/site-openafs-client:default.log [ Nov 27 12:06:57 Method or service exit timed out. Killing contract 47 ] [ Nov 27 12:06:58 Method or service exit timed out. Killing contract 47 ] [ Nov 27 12:06:59 Method or service exit timed out. Killing contract 47 ] [ Nov 27 12:07:01 Method or service exit timed out. Killing contract 47 ] [ Nov 27 12:07:02 Method or service exit timed out. Killing contract 47 ] [ Nov 27 12:07:03 Method or service exit timed out. Killing contract 47 ] [ Nov 27 12:07:04 Method or service exit timed out. Killing contract 47 ] [ Nov 27 12:07:05 Method or service exit timed out. Killing contract 47 ] [...] (We can of course run AFS outside of SMF, but the underlying problem is still there) When we try to reboot the machine, it does not reboot cleanly, and if I run a "ctstat -v" while it's shutting down it frequently kernel panics. (The kernel panics can happen even *after* I shutdown AFS manually first!). panic[cpu0]/thread=ffffffff820cdc80: BAD TRAP: type=e (#pf Page fault) rp=fffffe80005ccb80 addr=fffffffffffff4e8 ctstat: #pf Page fault Bad kernel fault at addr=0xfffffffffffff4e8 pid=497, pc=0xfffffffffb9d26f5, sp=0xfffffe80005ccc70, eflags=0x10282 cr0: 8005003b cr4: 6f0 cr2: fffffffffffff4e8 cr3: f7fd000 cr8: c rdi: ffffffff874823c0 rsi: ffffffff81d6da60 rdx: 0 rcx: fffffffffffff438 r8: fffffe80005ccd30 r9: 100000 rax: fffffffffffff438 rbx: 6 rbp: fffffe80005cccf0 r10: fffffffffbc64860 r11: 0 r12: ffffffff87482200 r13: ffffffff874823b0 r14: ffffffff874822b8 r15: d fsb: ffffffff80000000 gsb: fffffffffbc25460 ds: 43 es: 43 fs: 0 gs: 1c3 trp: e err: 0 rip: fffffffffb9d26f5 cs: 28 rfl: 10282 rsp: fffffe80005ccc70 ss: 30 fffffe80005cca90 unix:real_mode_end+7051 () fffffe80005ccb70 unix:trap+d86 () fffffe80005ccb80 unix:cmntrap+13f () fffffe80005cccf0 genunix:contract_process_status+155 () fffffe80005ccdb0 ctfs:ctfs_stat_ioctl+10e () fffffe80005ccde0 genunix:fop_ioctl+25 () fffffe80005ccec0 genunix:ioctl+ac () fffffe80005ccf10 unix:brand_sys_syscall32+1a3 () syncing file systems... done dumping to /dev/md/dsk/d10, offset 1719074816, content: kernel This problem seems similar to the one that's described here: http://www.openafs.org/pipermail/openafs-info/2005-October/019765.html We're running the prebuilt 1.4.5 Solaris10/x86 binaries from openafs.org, although we put a locally built one in as a test and it displayed the same symptoms. Any help appreciated. Regards, Robert. From hanke@rzg.mpg.de Tue Nov 27 06:44:40 2007 From: hanke@rzg.mpg.de (Christof Hanke) Date: Tue, 27 Nov 2007 08:44:40 +0200 Subject: [OpenAFS] Re: issue with Fedora 8 and retaining tokens after graphical login In-Reply-To: References: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> <1195921644.26707.6.camel@BillMurray> <20071124204132.GB28686@puariko.nirvana> Message-ID: <474BBCD8.5090705@rzg.mpg.de> Derrick Brashear wrote: > >> But let's make everybody happy: How about an rpmbuild time switch to >> allow anyone to rebuild with their favourite path style. I would still >> want to see FHS being the default, but transarc style is just a >> one-line change in the rpmbuild specfile. > > Likewise, the people who are using Transarc style would want that to > be default :) > That's why on the SUSE build-service, there are two package-families : "openafs" is FHS (hopefully, mostly, fully ? I haven't checked that.) "openafs-transarc" uses the transarc-pathes. rpms from one of those family conflicts with those of the other one. >> Just looking for synergy and a common openafs rpm design. Which could >> very well be more than Fedora/RHEL (but I admit not being a SuSE or >> Mandriva expert, so someone would have to step in to aid in this >> part). :) > I've had some communication with the SUSE-packagers and I think they would be very open to that. And I could volunteer to do some of the work on the SuSE side. > In theory we already have it. I've built the release SRPM on SLES. From l.schimmer@cgv.tugraz.at Tue Nov 27 09:52:59 2007 From: l.schimmer@cgv.tugraz.at (Lars Schimmer) Date: Tue, 27 Nov 2007 10:52:59 +0100 Subject: [OpenAFS] Win 1.5.27 seems to be fine Message-ID: <474BE8FB.7040009@cgv.tugraz.at> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi! Just a quick note (after some weeks testing): 1.5.27 seems to be far better than .24-.26 for my users who got their profile in AFS space. Til yet no afsd error and shutdown. MfG, Lars Schimmer - -- - ------------------------------------------------------------- TU Graz, Institut f=FCr ComputerGraphik & WissensVisualisierung Tel: +43 316 873-5405 E-Mail: l.schimmer@cgv.tugraz.at Fax: +43 316 873-5402 PGP-Key-ID: 0x4A9B1723 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFHS+j7mWhuE0qbFyMRAkZLAJ9t8IrxKhTj099a0VWxF8C2FUy1SQCdGgWN P3lAx1Ib2hTQ17D6ZE2hZ2M=3D =3DjEDX -----END PGP SIGNATURE----- From chas@cmf.nrl.navy.mil Tue Nov 27 14:12:22 2007 From: chas@cmf.nrl.navy.mil (chas williams - CONTRACTOR) Date: Tue, 27 Nov 2007 09:12:22 -0500 Subject: [OpenAFS] OpenAFS 1.4.5 on Solaris 10/x86 - spurious pid(s) in ctstat and ocassional kernel panics In-Reply-To: <20071127061734.GB8139@unimelb.edu.au> Message-ID: <200711271412.lARECMkC011130@cmf.nrl.navy.mil> In message <20071127061734.GB8139@unimelb.edu.au>,Robert Sturrock writes: > $ ctstat -vi 48 > CTID ZONEID TYPE STATE HOLDER EVENTS QTIME NTIME > 48 0 process owned 7 0 - - > cookie: 0x20 > informative event set: none > critical event set: core signal hwerr empty > fatal event set: none > parameter set: inherit regent > member processes: 335 336 338 343 344 345 346 347 1936749667 > inherited contracts: none > >pids 335-347 are afsd pides, but 1936749667 is obviously not valid! i remember having problem before when afsd made its syscalls to create the kernel threads. i dont remember what i was doing to try to fix this issue. From jaltman@secure-endpoints.com Tue Nov 27 14:39:29 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Tue, 27 Nov 2007 15:39:29 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <474C0D5B.2060802@enem.nl> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> Message-ID: <474C2C21.7010507@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms080505060206080702090008 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Your log shows that at 12:56.57 winword.exe attempted the following request that was denied due to the file being in use. Sequence: 669571 Date & Time: 11/27/2007 12:56:57 PM Event Class: File System Operation: IRP_MJ_CREATE Result: NOT GRANTED Path: P:\2006\06226-06250\06249\Rapportages\rapportage Friesland 15 november 2000-02.doc TID: 336 Duration: 92.1418212 Desired Access: Generic Read/Write Disposition: Open Options: Synchronous IO Non-Alert, Non-Directory File Attributes: N ShareMode: Read, Write AllocationSize: n/a Winword.exe then attempts to repeat the request and you terminated the logging before a result could be determined. Sequence: 709074 Date & Time: 11/27/2007 12:58:29 PM Event Class: File System Operation: IRP_MJ_CREATE Result: Path: P:\2006\06226-06250\06249\Rapportages\rapportage Friesland 15 november 2000-02.doc TID: 336 Duration: Desired Access: Generic Read Disposition: Open Options: Synchronous IO Non-Alert, Non-Directory File Attributes: N ShareMode: Read AllocationSize: n/a I do not see in your log any "DISCONNECTED" errors which is what would be listed if the CIFS client timed out and terminated the connection to the AFS Client Service. As far as I can tell WinWord is simply trying to open a file it can't obtain a lock on fails, and tries again indefinitely. You said previously that earlier versions of OpenAFS worked better for you. What are the differences in the requests performed between the current and previous versions of OpenAFS? You should be able to perform this analysis yourself. Use the Process Monitor Filter option to restrict the displayed events to winword.exe, and paths beginning with P:\ and \\AFS Jeffrey Altman Hans Melgers wrote: > > Hi Jeffrey, > > We just made the log, its quite big (38MB) so please download it from > the site mentioned at the bottom of the mail. > I hope you can find something. > > Kind regards, > Hans Melgers > > > > Jeffrey Altman wrote: >> Hans: >> >> Again I must point you at the debugging section of the release notes. >> If you would gather the necessary data the problem could be examined >> based upon the actual requests that are being issued by the software and >> the responses that are provided by the AFS client instead of attempting >> to guess at the behavior by looking at the entire system (Office XP, >> Windows, AFS client, AFS server) as a black box. >> >> While it is possible the Netware client might be involved in some way I >> would find that unlikely. The application is requesting a lock and the >> file server is unable to grant it because another client already has it. >> In this situation, the file server will report EAGAIN to the client >> and the AFS client will retry in the hopes of being able to obtain >> the lock. >> Perhaps the application is timing out the request before the AFS client >> returns. I don't know for sure because you have not obtained the >> necessary data. >> >> Jeffrey Altman >> >> >> ENEM | Hans Melgers wrote: >> >>> I dont thing the problem is in the afs client. We found this problem >>> only occurs when opening a file that already has been opened by >>> another user. We upgraded to office 2003 and it takes alot of time >>> before word gives any response. >>> This only seems to happen on the "original" pc's, when trying from >>> our own pc everything works fine, word opens in read-only mode but >>> that doesnt happen on these pc's ? >>> >>> Maybe it has something to do with the fact that they are also >>> running the netware client. Are there any known issues with this >>> setup ? Should we run afs before netware ? Maybe check some settings ? >>> >>> >>> -----Original Message----- >>> From: Jeffrey Altman [mailto:jaltman@secure-endpoints.com] Sent: >>> woensdag 14 november 2007 15:43 >>> To: Hans Melgers >>> Cc: openafs-info@openafs.org >>> Subject: Re: [OpenAFS] MS Word crashing when opening files, 1.5.27 >>> client >>> >>> Hans: >>> >>> Please follow the debugging section of the OpenAFS for Windows >>> Release Notes. In particular, use the SysInternals' Process Monitor >>> to log the requests from Word so that you can determine which >>> requests are failing with 1.5.27 that succeed with 1.5.25. >>> >>> One change that went into 1.5.27 is support for retrying requests >>> when the file server responds to the client with EWOULDBLOCK / >>> EAGAIN. In prior releases the client would simply fail the >>> request. In 1.5.27 the client retries the request every two seconds >>> up to 45 seconds until the request completes or the timeout is >>> reached. In the trace log it is the >>> cm_Analyze() messages will include the error number. Possible error >>> numbers to look for include: >>> >>> UAEWOULDBLOCK 49733416L >>> UAEAGAIN 49733386L >>> EAGAIN 11L >>> >>> >>> Jeffrey Altman >>> >>> >>> >>> >>> Hans Melgers wrote: >>> >>>> Hello, >>>> >>>> We just installed some XP machines with the latest 1.5.27 afs >>>> client. On several of them Word crashes badly when trying to open a >>>> .doc file on afs. Most of them running office XP but also seen on >>>> office 2003. >>>> >>>> I run 1.5.25 and havent seen the problem. Could this be a bug in >>>> the latest win client ? acl's are ok. Server is 1.4.4 on freebsd >>>> 6.2 Anyone else seeing this behaviour ? >>>> >>>> In Filelog i only see some callback issues: >>>> >>>> Wed Nov 14 10:56:32 2007 CB: ProbeUuid for 213.46.24.222:17980 >>>> failed -01 Wed Nov 14 10:57:11 2007 CB: ProbeUuid for >>>> 213.46.24.222:17982 failed -01 Wed Nov 14 10:58:08 2007 CB: >>>> ProbeUuid for 213.46.24.222:17987 failed -01 Wed Nov 14 10:58:18 >>>> 2007 CB: ProbeUuid for 213.46.24.222:17976 failed -01 Wed Nov 14 >>>> 11:01:25 2007 CB: ProbeUuid for 213.46.24.222:17995 failed -01 >>>> >>>> >>>> Any help appreciated. >>>> >>>> Hans >>>> >>> >>> _______________________________________________ >>> OpenAFS-info mailing list >>> OpenAFS-info@openafs.org >>> https://lists.openafs.org/mailman/listinfo/openafs-info >>> --------------ms080505060206080702090008 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMjcxNDM5MjlaMCMGCSqGSIb3DQEJBDEWBBQMKPrX eFjpiCck3yCl7cJrdsJi3TBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAV4CwazEx3otIekq6MxqXKwzMdC4tqr6FpsHUzFiIHk2tL3haAiGN up/C6/vizSVYaPdMISItFg7Z2BKz2KUwdpXBG2dckObxld3Kf+8+hXF4gfKdHZKWam9czlkA 5xIPIpTAHiKYt/9p8Ko13rJtmEFA+QhGGmkuQe2zkVLXyYuGCRltrFJUnUhgjFLTLX2zemCw TI32K+KzYmx9O5SCnqfEr9UZmRTMy8ZmDc5E/3iehd0oyT0OU/LgLe3h1RuQ0npt+lKOHlMG x0CfqiaH+crOP7r3WIlVWDOn43gP375dG11BGiU+QljfV+IPaIxh3QQKwRXfbUqlBrVY2BVw ygAAAAAAAA== --------------ms080505060206080702090008-- From shadow@gmail.com Tue Nov 27 14:49:16 2007 From: shadow@gmail.com (Derrick Brashear) Date: Tue, 27 Nov 2007 09:49:16 -0500 Subject: [OpenAFS] Re: issue with Fedora 8 and retaining tokens after graphical login In-Reply-To: <474BBCD8.5090705@rzg.mpg.de> References: <1b8d56200711240702l743de3br594640f105a9b824@mail.gmail.com> <1195921644.26707.6.camel@BillMurray> <20071124204132.GB28686@puariko.nirvana> <474BBCD8.5090705@rzg.mpg.de> Message-ID: On Nov 27, 2007 1:44 AM, Christof Hanke wrote: > Derrick Brashear wrote: > > > >> But let's make everybody happy: How about an rpmbuild time switch to > >> allow anyone to rebuild with their favourite path style. I would still > >> want to see FHS being the default, but transarc style is just a > >> one-line change in the rpmbuild specfile. > > > > Likewise, the people who are using Transarc style would want that to > > be default :) > > > That's why on the SUSE build-service, there are two package-families : > "openafs" is FHS (hopefully, mostly, fully ? I haven't checked that.) > "openafs-transarc" uses the transarc-pathes. > rpms from one of those family conflicts with those of the other one. If we could modify our specfile to allow this from one specfile, I'd ship it. Likewise, we could generate both kinds of rpms from one (set of) kernel modules, which is 99% of the time we spend on builds. From jason@rampaginggeek.com Tue Nov 27 18:51:03 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Tue, 27 Nov 2007 13:51:03 -0500 Subject: [OpenAFS] OpenSuSE rpms? Message-ID: <474C6717.4050500@rampaginggeek.com> Would someone please point me to some openafs client rpms for opensuse? Barring that, Are there any gotcha about installing from source on OpenSuSE vs Redhat? Thanks, Jason From kaaker@brocade.com Tue Nov 27 19:08:23 2007 From: kaaker@brocade.com (Ken Aaker) Date: Tue, 27 Nov 2007 13:08:23 -0600 Subject: [OpenAFS] OpenSuSE rpms? In-Reply-To: <474C6717.4050500@rampaginggeek.com> References: <474C6717.4050500@rampaginggeek.com> Message-ID: <474C6B27.8070907@brocade.com> Jason Edgecombe wrote: > Would someone please point me to some openafs client rpms for opensuse? > > Barring that, Are there any gotcha about installing from source on > OpenSuSE vs Redhat? > > Thanks, > Jason > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > I've used Christof's OpenSuSE rpm's from http://software.opensuse.org/search on a few different machines and they've worked well. Just search for openafs. The 1-click install works nicely on 10.3, and using the repository for 10.2 also has worked well for me. Ken Aaker From hans@enem.nl Tue Nov 27 20:43:11 2007 From: hans@enem.nl (Hans Melgers) Date: Tue, 27 Nov 2007 21:43:11 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <474C2C21.7010507@secure-endpoints.com> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> <474C2C21.7010507@secure-endpoints.com> Message-ID: <474C815F.2010308@enem.nl> Thanks Jeffrey. The older version was in the early stages, we found it also happens with 1.5.25. On some pc's this happens, on others it doesnt. So the problem is in winword trying to obtain a lock indefinitely. What do you recommend ? is there any (registry) setting that does this ? Would this be a solution ? I mean, would winword accept and act on changing this setting to 0 ? : If the CIFS-AFS server is unable to renew the AFS file server locks, then it will invalidate the associated file handles. This is the same behavior that an application will experience if it was using a Windows File Share and the connection was broken. Invalidating the file handles prevents subsequent data corruption from taking place. If you wish to disable the acquisition of locks from the file server, this can be performed using the EnableServerLocks registry value. Value: EnableServerLocks Type: DWORD {0, 1, 2} Default: 1 Determines whether or not the AFS file server is contacted for 0: never obtain server locks 1: obtain server locks unless the file server says not to 2: always obtain server locks Jeffrey Altman wrote: > Your log shows that at 12:56.57 winword.exe attempted the following > request that was denied due to the file being in use. > > Sequence: 669571 > Date & Time: 11/27/2007 12:56:57 PM > Event Class: File System > Operation: IRP_MJ_CREATE > Result: NOT GRANTED > Path: P:\2006\06226-06250\06249\Rapportages\rapportage Friesland 15 > november 2000-02.doc > TID: 336 > Duration: 92.1418212 > Desired Access: Generic Read/Write > Disposition: Open > Options: Synchronous IO Non-Alert, Non-Directory File > Attributes: N > ShareMode: Read, Write > AllocationSize: n/a > > Winword.exe then attempts to repeat the request and you terminated the > logging before a result could be determined. > > Sequence: 709074 > Date & Time: 11/27/2007 12:58:29 PM > Event Class: File System > Operation: IRP_MJ_CREATE > Result: > Path: P:\2006\06226-06250\06249\Rapportages\rapportage Friesland 15 > november 2000-02.doc > TID: 336 > Duration: > Desired Access: Generic Read > Disposition: Open > Options: Synchronous IO Non-Alert, Non-Directory File > Attributes: N > ShareMode: Read > AllocationSize: n/a > > I do not see in your log any "DISCONNECTED" errors which is what would > be listed if the CIFS client timed out and terminated the connection to > the AFS Client Service. As far as I can tell WinWord is simply trying > to open a file it can't obtain a lock on fails, and tries again > indefinitely. > > You said previously that earlier versions of OpenAFS worked better for > you. What are the differences in the requests performed between the > current and previous versions of OpenAFS? You should be able to > perform this analysis yourself. Use the Process Monitor Filter option > to restrict the displayed events to winword.exe, and paths beginning > with P:\ and \\AFS > > Jeffrey Altman > > > > Hans Melgers wrote: > >> Hi Jeffrey, >> >> We just made the log, its quite big (38MB) so please download it from >> the site mentioned at the bottom of the mail. >> I hope you can find something. >> >> Kind regards, >> Hans Melgers >> >> >> >> Jeffrey Altman wrote: >> >>> Hans: >>> >>> Again I must point you at the debugging section of the release notes. >>> If you would gather the necessary data the problem could be examined >>> based upon the actual requests that are being issued by the software and >>> the responses that are provided by the AFS client instead of attempting >>> to guess at the behavior by looking at the entire system (Office XP, >>> Windows, AFS client, AFS server) as a black box. >>> >>> While it is possible the Netware client might be involved in some way I >>> would find that unlikely. The application is requesting a lock and the >>> file server is unable to grant it because another client already has it. >>> In this situation, the file server will report EAGAIN to the client >>> and the AFS client will retry in the hopes of being able to obtain >>> the lock. >>> Perhaps the application is timing out the request before the AFS client >>> returns. I don't know for sure because you have not obtained the >>> necessary data. >>> >>> Jeffrey Altman >>> >>> >>> ENEM | Hans Melgers wrote: >>> >>> >>>> I dont thing the problem is in the afs client. We found this problem >>>> only occurs when opening a file that already has been opened by >>>> another user. We upgraded to office 2003 and it takes alot of time >>>> before word gives any response. >>>> This only seems to happen on the "original" pc's, when trying from >>>> our own pc everything works fine, word opens in read-only mode but >>>> that doesnt happen on these pc's ? >>>> >>>> Maybe it has something to do with the fact that they are also >>>> running the netware client. Are there any known issues with this >>>> setup ? Should we run afs before netware ? Maybe check some settings ? >>>> >>>> >>>> -----Original Message----- >>>> From: Jeffrey Altman [mailto:jaltman@secure-endpoints.com] Sent: >>>> woensdag 14 november 2007 15:43 >>>> To: Hans Melgers >>>> Cc: openafs-info@openafs.org >>>> Subject: Re: [OpenAFS] MS Word crashing when opening files, 1.5.27 >>>> client >>>> >>>> Hans: >>>> >>>> Please follow the debugging section of the OpenAFS for Windows >>>> Release Notes. In particular, use the SysInternals' Process Monitor >>>> to log the requests from Word so that you can determine which >>>> requests are failing with 1.5.27 that succeed with 1.5.25. >>>> >>>> One change that went into 1.5.27 is support for retrying requests >>>> when the file server responds to the client with EWOULDBLOCK / >>>> EAGAIN. In prior releases the client would simply fail the >>>> request. In 1.5.27 the client retries the request every two seconds >>>> up to 45 seconds until the request completes or the timeout is >>>> reached. In the trace log it is the >>>> cm_Analyze() messages will include the error number. Possible error >>>> numbers to look for include: >>>> >>>> UAEWOULDBLOCK 49733416L >>>> UAEAGAIN 49733386L >>>> EAGAIN 11L >>>> >>>> >>>> Jeffrey Altman >>>> >>>> >>>> >>>> >>>> Hans Melgers wrote: >>>> >>>> >>>>> Hello, >>>>> >>>>> We just installed some XP machines with the latest 1.5.27 afs >>>>> client. On several of them Word crashes badly when trying to open a >>>>> .doc file on afs. Most of them running office XP but also seen on >>>>> office 2003. >>>>> >>>>> I run 1.5.25 and havent seen the problem. Could this be a bug in >>>>> the latest win client ? acl's are ok. Server is 1.4.4 on freebsd >>>>> 6.2 Anyone else seeing this behaviour ? >>>>> >>>>> In Filelog i only see some callback issues: >>>>> >>>>> Wed Nov 14 10:56:32 2007 CB: ProbeUuid for 213.46.24.222:17980 >>>>> failed -01 Wed Nov 14 10:57:11 2007 CB: ProbeUuid for >>>>> 213.46.24.222:17982 failed -01 Wed Nov 14 10:58:08 2007 CB: >>>>> ProbeUuid for 213.46.24.222:17987 failed -01 Wed Nov 14 10:58:18 >>>>> 2007 CB: ProbeUuid for 213.46.24.222:17976 failed -01 Wed Nov 14 >>>>> 11:01:25 2007 CB: ProbeUuid for 213.46.24.222:17995 failed -01 >>>>> >>>>> >>>>> Any help appreciated. >>>>> >>>>> Hans >>>>> >>>>> >>>> _______________________________________________ >>>> OpenAFS-info mailing list >>>> OpenAFS-info@openafs.org >>>> https://lists.openafs.org/mailman/listinfo/openafs-info >>>> >>>> From jaltman@secure-endpoints.com Tue Nov 27 21:33:00 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Tue, 27 Nov 2007 22:33:00 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <474C815F.2010308@enem.nl> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> <474C2C21.7010507@secure-endpoints.com> <474C815F.2010308@enem.nl> Message-ID: <474C8D0C.50500@secure-endpoints.com> If you set that value to 0, then winword will start but two users on different machines will be able to open the file at the same time. Doing so will corrupt the file if both decide to write to it. Jeffrey Altman Hans Melgers wrote: > > > Thanks Jeffrey. The older version was in the early stages, we found it > also happens with 1.5.25. On some pc's this happens, on others it doesnt. > > So the problem is in winword trying to obtain a lock indefinitely. > What do you recommend ? is there any (registry) setting that does this ? > > Would this be a solution ? I mean, would winword accept and act on > changing this setting to 0 ? : > > If the CIFS-AFS server is unable to renew the AFS file server locks, > then it will invalidate the associated file handles. This is the same > behavior that an application will experience if it was using a Windows > File Share and the connection was broken. Invalidating the file > handles prevents subsequent data corruption from taking place. > If you wish to disable the acquisition of locks from the file server, > this can be performed using the EnableServerLocks registry value. > > Value: EnableServerLocks > Type: DWORD {0, 1, 2} > Default: 1 > > Determines whether or not the AFS file server is contacted for > > 0: never obtain server locks > > 1: obtain server locks unless the file server says not to > > 2: always obtain server locks From hans@enem.nl Tue Nov 27 21:44:40 2007 From: hans@enem.nl (Hans Melgers) Date: Tue, 27 Nov 2007 22:44:40 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <474C8D0C.50500@secure-endpoints.com> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> <474C2C21.7010507@secure-endpoints.com> <474C815F.2010308@enem.nl> <474C8D0C.50500@secure-endpoints.com> Message-ID: <474C8FC8.5000405@enem.nl> This is a multi-part message in MIME format. --------------060701030100060703050305 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit I understand, but probably most files are used as templates so that wouldnt be a problem in this case. What still puzzles me is why this doesnt work on those pc's with such a standard application as MS Word. (is there a more standard application in the pc world ?) Could it be a legacy from the novell client ? Never any problems when they used that. I also noticed that some are running XP home edition, does that make any difference ? There must be something causing this. Maybe somebody else seen the same problem ? Jeffrey Altman wrote: > If you set that value to 0, then winword will start but two users on > different machines will be able to open the file at the same time. > Doing so will corrupt the file if both decide to write to it. > > Jeffrey Altman > > > Hans Melgers wrote: > >> Thanks Jeffrey. The older version was in the early stages, we found it >> also happens with 1.5.25. On some pc's this happens, on others it doesnt. >> >> So the problem is in winword trying to obtain a lock indefinitely. >> What do you recommend ? is there any (registry) setting that does this ? >> >> Would this be a solution ? I mean, would winword accept and act on >> changing this setting to 0 ? : >> >> If the CIFS-AFS server is unable to renew the AFS file server locks, >> then it will invalidate the associated file handles. This is the same >> behavior that an application will experience if it was using a Windows >> File Share and the connection was broken. Invalidating the file >> handles prevents subsequent data corruption from taking place. >> If you wish to disable the acquisition of locks from the file server, >> this can be performed using the EnableServerLocks registry value. >> >> Value: EnableServerLocks >> Type: DWORD {0, 1, 2} >> Default: 1 >> >> Determines whether or not the AFS file server is contacted for >> >> 0: never obtain server locks >> >> 1: obtain server locks unless the file server says not to >> >> 2: always obtain server locks >> > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > --------------060701030100060703050305 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit I understand, but probably most files are used as templates so that wouldnt be a problem in this case.

What still puzzles me is why this doesnt work on those pc's with such a standard application as MS Word. (is there a more standard application in the pc world ?)
Could it be a legacy from the novell client ? Never any problems when they used that. I also noticed that some are running  XP home edition, does that make any difference ?
There must be something causing this. Maybe somebody else seen the same problem ?

Jeffrey Altman wrote:
If you set that value to 0, then winword will start but two users on
different machines will be able to open the file at the same time.
Doing so will corrupt the file if both decide to write to it.

Jeffrey Altman


Hans Melgers wrote:
  
Thanks Jeffrey.  The older version was in the early stages, we found it
also happens with 1.5.25.  On some pc's this happens, on others it doesnt.

So the problem is in winword trying to obtain a lock indefinitely.
What do you recommend ? is there any (registry) setting that does this ?

Would this be a solution ? I mean, would winword accept and act on
changing this setting to 0 ? :

If the CIFS-AFS server is unable to renew the AFS file server locks,
then it will invalidate the associated file handles.  This is the same
behavior that an application will experience if it was using a Windows
File Share and the connection was broken.   Invalidating the file
handles prevents subsequent data corruption from taking place.
If you wish to disable the acquisition of locks from the file server,
this can be performed using the EnableServerLocks registry value.

Value: EnableServerLocks
Type: DWORD {0, 1, 2}
Default: 1

Determines whether or not the AFS file server is contacted for

0: never obtain server locks

1: obtain server locks unless the file server says not to

2: always obtain server locks
    

_______________________________________________
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info
  
--------------060701030100060703050305-- From jaltman@secure-endpoints.com Wed Nov 28 02:14:04 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Wed, 28 Nov 2007 03:14:04 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <474C8FC8.5000405@enem.nl> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> <474C2C21.7010507@secure-endpoints.com> <474C815F.2010308@enem.nl> <474C8D0C.50500@secure-endpoints.com> <474C8FC8.5000405@enem.nl> Message-ID: <474CCEEC.5040103@secure-endpoints.com> If the files are truly intended for read-only use, store them in a directory that provides only 'rl' access to the end users or store them in a .readonly volume. In both of those cases the AFS Cache Manager knows that the user cannot obtain a lock on the file and will issue one locally. Jeffrey Altman Hans Melgers wrote: > I understand, but probably most files are used as templates so that > wouldnt be a problem in this case. > > What still puzzles me is why this doesnt work on those pc's with such a > standard application as MS Word. (is there a more standard application > in the pc world ?) > Could it be a legacy from the novell client ? Never any problems when > they used that. I also noticed that some are running XP home edition, > does that make any difference ? > There must be something causing this. Maybe somebody else seen the same > problem ? > > Jeffrey Altman wrote: >> If you set that value to 0, then winword will start but two users on >> different machines will be able to open the file at the same time. >> Doing so will corrupt the file if both decide to write to it. >> >> Jeffrey Altman >> >> >> Hans Melgers wrote: >> >>> Thanks Jeffrey. The older version was in the early stages, we found it >>> also happens with 1.5.25. On some pc's this happens, on others it doesnt. >>> >>> So the problem is in winword trying to obtain a lock indefinitely. >>> What do you recommend ? is there any (registry) setting that does this ? >>> >>> Would this be a solution ? I mean, would winword accept and act on >>> changing this setting to 0 ? : >>> >>> If the CIFS-AFS server is unable to renew the AFS file server locks, >>> then it will invalidate the associated file handles. This is the same >>> behavior that an application will experience if it was using a Windows >>> File Share and the connection was broken. Invalidating the file >>> handles prevents subsequent data corruption from taking place. >>> If you wish to disable the acquisition of locks from the file server, >>> this can be performed using the EnableServerLocks registry value. >>> >>> Value: EnableServerLocks >>> Type: DWORD {0, 1, 2} >>> Default: 1 >>> >>> Determines whether or not the AFS file server is contacted for >>> >>> 0: never obtain server locks >>> >>> 1: obtain server locks unless the file server says not to >>> >>> 2: always obtain server locks >>> >> >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info >> From lara.lloret.iglesias@cern.ch Wed Nov 28 11:25:18 2007 From: lara.lloret.iglesias@cern.ch (Lara Lloret Iglesias) Date: Wed, 28 Nov 2007 12:25:18 +0100 Subject: [OpenAFS] AFS-client Message-ID: <9A6A62B6B84859469F3EBB5F09D818CA994168@cernxchg50.cern.ch> Hello everybody, I'm new here... I want to ask you a question: I'm trying to install an = afs-client using Scientific Linux Cern 4.5. The problem is that I think = i've done everything right, but when I reboot my computer, the = CellServDB changes alone, and doesn't contain the new servers that I = added. I've tried with fs newcells but it stills changes on reboot... = and it doesn't work... any idea??? Lara From mikkel@linet.dk Wed Nov 28 11:32:45 2007 From: mikkel@linet.dk (Mikkel Kruse Johnsen) Date: Wed, 28 Nov 2007 12:32:45 +0100 Subject: [OpenAFS] AFS-client In-Reply-To: <9A6A62B6B84859469F3EBB5F09D818CA994168@cernxchg50.cern.ch> References: <9A6A62B6B84859469F3EBB5F09D818CA994168@cernxchg50.cern.ch> Message-ID: <1196249565.2653.6.camel@tux.lib.cbs.dk> --=-GB+o40g4mwskOUogxw3+ Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Hi Lara You have to make your changes in '/usr/vice/etc/CellServDB.local' That file and '/usr/vice/etc/CellServDB.dist' is merged to form the final CellServDB. /Mikkel On Wed, 2007-11-28 at 12:25 +0100, Lara Lloret Iglesias wrote: > Hello everybody, > > I'm new here... I want to ask you a question: I'm trying to install an afs-client using Scientific Linux Cern 4.5. The problem is that I think i've done everything right, but when I reboot my computer, the CellServDB changes alone, and doesn't contain the new servers that I added. I've tried with fs newcells but it stills changes on reboot... and it doesn't work... > any idea??? > > Lara > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info Med Venlig Hilsen / Kind Regards Mikkel Kruse Johnsen Adm.Dir. Linet Ørholmgade 6 st tv Copenhagen N 2200 Denmark Work: +45 21287793 Mobile: +45 21287793 Email: mikkel@linet.dk IM: mikkel@linet.dk (MSN) Professional Profile Healthcare Network Consultant --=-GB+o40g4mwskOUogxw3+ Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: 7bit Hi Lara

You have to make your changes in '/usr/vice/etc/CellServDB.local'

That file and '/usr/vice/etc/CellServDB.dist' is merged to form the final CellServDB.

/Mikkel

On Wed, 2007-11-28 at 12:25 +0100, Lara Lloret Iglesias wrote:
Hello everybody,

I'm new here... I want to ask you a question: I'm trying to install an afs-client using Scientific Linux Cern 4.5. The problem is that I think i've done everything right, but when I reboot my computer, the CellServDB changes alone, and doesn't contain the new servers that I added. I've tried with fs newcells but it stills changes on reboot... and it doesn't work...
any idea???

Lara
_______________________________________________
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info
Med Venlig Hilsen / Kind Regards


Mikkel Kruse Johnsen
Adm.Dir.

Linet
Ørholmgade 6 st tv
Copenhagen N 2200 Denmark
Work:    +45 21287793
Mobile: +45 21287793
Email: mikkel@linet.dk
IM: mikkel@linet.dk (MSN)
Professional Profile
Healthcare


Network Consultant
--=-GB+o40g4mwskOUogxw3+-- From lara.lloret.iglesias@cern.ch Wed Nov 28 12:38:47 2007 From: lara.lloret.iglesias@cern.ch (Lara Lloret Iglesias) Date: Wed, 28 Nov 2007 13:38:47 +0100 Subject: [OpenAFS] (no subject) Message-ID: <9A6A62B6B84859469F3EBB5F09D818CA99416B@cernxchg50.cern.ch> Hi again, Thank you very much for your answer, it doesn't change now :) But now i've another problem in another AFS-client, even if my = CellServDB file is the same, with all the servers that i wanted, when i = write=20 ls /afs the only server that i found is cern.ch. None of all the other servers = that are in my CellServDB file... very strange... Lara From haba@kth.se Wed Nov 28 12:46:40 2007 From: haba@kth.se (Harald Barth) Date: Wed, 28 Nov 2007 13:46:40 +0100 (CET) Subject: [OpenAFS] (no subject) In-Reply-To: <9A6A62B6B84859469F3EBB5F09D818CA99416B@cernxchg50.cern.ch> References: <9A6A62B6B84859469F3EBB5F09D818CA99416B@cernxchg50.cern.ch> Message-ID: <20071128.134640.181297910.haba@habanero.pdc.kth.se> > the only server that i found is cern.ch I guess the only _cell_ you found. Depending on if you use the dynamic /afs root feature or not, the content of /afs is generated from CellServDB or the content that the sysadmin did put into the volume root.afs. Harald. From lara.lloret.iglesias@cern.ch Wed Nov 28 13:04:48 2007 From: lara.lloret.iglesias@cern.ch (Lara Lloret Iglesias) Date: Wed, 28 Nov 2007 14:04:48 +0100 Subject: [OpenAFS] (no subject) Message-ID: <9A6A62B6B84859469F3EBB5F09D818CA99416D@cernxchg50.cern.ch> Yes, I meant "cell" :) Sorry, I don't understand. When the content of AFS is not generated by CellServDB what should I do? Lara From haba@kth.se Wed Nov 28 13:44:47 2007 From: haba@kth.se (Harald Barth) Date: Wed, 28 Nov 2007 14:44:47 +0100 (CET) Subject: [OpenAFS] (no subject) In-Reply-To: <9A6A62B6B84859469F3EBB5F09D818CA99416D@cernxchg50.cern.ch> References: <9A6A62B6B84859469F3EBB5F09D818CA99416D@cernxchg50.cern.ch> Message-ID: <20071128.144447.76029877.haba@habanero.pdc.kth.se> $ /usr/openafs/sbin/afsd -help Usage: /usr/openafs/sbin/afsd [-blocks <1024 byte blocks in cache>] [-files ] [-rootvol ] [-stat ] [-memcache] [-cachedir ] [-mountdir ] [-daemons ] [-nosettime] [-verbose] [-rmtsys] [-debug] [-chunksize ] [-dcache ] [-volumes ] [-biods ] [-prealloc ] [-confdir ] [-logfile ] [-waitclose] [-shutdown] [-afsdb] [-files_per_subdir ] [-dynroot] [-fakestat] [-fakestat-all] [-nomount] [-backuptree] [-rxbind] [-settime] [-rxpck ] [-vicepaccess] [-check-partitions] [-help] Where: -memcache run diskless -nosettime don't set the time -verbose display lots of information -rmtsys start NFS rmtsysd program -debug display debug info -waitclose make close calls synchronous -shutdown Shutdown all afs state -afsdb Enable AFSDB support -dynroot Enable dynroot support -fakestat Enable fakestat support for cross-cell mounts -fakestat-all Enable fakestat support for all mounts -nomount Do not mount AFS -backuptree Prefer backup volumes for mointpoints in backup volumes -rxbind Bind the Rx socket (one interface only) -settime set the time -vicepaccess Enable direct I/O to visible vicep-partitions -check-partitions Check fileserver partitions and exit -dynroot is for /afs from CellServDB. -afsdb is a good thing if you want to access cells that have their info in DNS instead of CellServDB Harald. From jason@rampaginggeek.com Wed Nov 28 14:19:30 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Wed, 28 Nov 2007 09:19:30 -0500 Subject: [OpenAFS] Slow AFS performance Message-ID: <474D78F2.40808@rampaginggeek.com> I'm experiencing poor AFS performance on under Sparc solaris 9 09/05HW running Openafs server 1.4.1 on a Sun StorageTeck 3511 Fibre channel to SATA array At first, I thought that having UFS logging disabled was the culprit, but I have enabled UFS logging and I am using the namei server, but performance still stinks. It took 1.5 hours to move a 3.2GB volume to the server. Things seem fine except on the Fibre channel disks. Here is a snippet from "iostat 2": tty sd1 ssd0 ssd1 ssd2 cpu tin tout kps tps serv kps tps serv kps tps serv kps tps serv us sy wt id 3 41 0 0 0 575 82 60 0 0 0 0 0 0 0 0 26 73 21 125 0 0 0 544 78 78 0 0 0 0 0 0 0 1 26 72 0 40 0 0 0 663 96 67 0 0 0 0 0 0 0 1 23 76 0 40 0 0 0 442 50 23 0 0 0 0 0 0 0 0 25 75 21 47 0 0 0 398 50 23 0 0 0 0 0 0 0 0 25 75 20 102 0 0 0 388 51 24 1558 6 41 0 0 0 0 2 28 69 0 40 0 0 0 425 54 24 45657 56 140 0 0 0 1 13 62 25 0 41 3 1 9 735 57 22 46988 60 136 0 0 0 0 14 57 28 0 41 0 0 0 826 66 23 44894 63 138 0 0 0 0 14 59 27 0 41 0 0 0 951 73 24 46346 61 137 0 0 0 0 14 59 26 0 66 1 1 7 812 73 35 10235 48 116 0 0 0 0 6 36 58 0 41 0 0 0 1026 89 31 228 2 21 0 0 0 0 2 24 74 kps on ssd0 (/vicepa) stays around 300-400. The 44000kps on ssd1 is when I ran "dd if=/dev/zero of=/vicepc/dummy" to a different disk on the same array My bonnie++ performance numbers for vicepa are here: http://www.coe.uncc.edu/~jwedgeco/bonnie.html What could AFS be doing that causes the performance to stink? Thanks, Jason From chas@cmf.nrl.navy.mil Wed Nov 28 14:42:40 2007 From: chas@cmf.nrl.navy.mil (chas williams - CONTRACTOR) Date: Wed, 28 Nov 2007 09:42:40 -0500 Subject: [OpenAFS] Slow AFS performance In-Reply-To: <474D78F2.40808@rampaginggeek.com> Message-ID: <200711281442.lASEge8L028063@cmf.nrl.navy.mil> In message <474D78F2.40808@rampaginggeek.com>,Jason Edgecombe writes: >array > >My bonnie++ performance numbers for vicepa are here: >http://www.coe.uncc.edu/~jwedgeco/bonnie.html > > >What could AFS be doing that causes the performance to stink? run strace on fileserver/volserver and see what its doing as far as i/o. as i recall, the fileserver doesnt always use the best size for getting performance but it could be another issue. From warlord@MIT.EDU Wed Nov 28 14:45:26 2007 From: warlord@MIT.EDU (Derek Atkins) Date: Wed, 28 Nov 2007 09:45:26 -0500 Subject: [OpenAFS] (no subject) In-Reply-To: <9A6A62B6B84859469F3EBB5F09D818CA99416D@cernxchg50.cern.ch> (Lara Lloret Iglesias's message of "Wed\, 28 Nov 2007 14\:04\:48 +0100") References: <9A6A62B6B84859469F3EBB5F09D818CA99416D@cernxchg50.cern.ch> Message-ID: "Lara Lloret Iglesias" writes: > Yes, I meant "cell" :) > > Sorry, I don't understand. > When the content of AFS is not generated by CellServDB what should I do? cd /afs/athena.mit.edu/ This will automagically create "athena.mit.edu" if you're using dynroot. I'm PRETTY sure that it wont create the entry until it needs to. > Lara -derek -- Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory Member, MIT Student Information Processing Board (SIPB) URL: http://web.mit.edu/warlord/ PP-ASEL-IA N1NWH warlord@MIT.EDU PGP key available From botsch@cnf.cornell.edu Wed Nov 28 16:20:16 2007 From: botsch@cnf.cornell.edu (Dave Botsch) Date: Wed, 28 Nov 2007 11:20:16 -0500 Subject: [OpenAFS] rhel3 openafs 1.4.5 spec patch In-Reply-To: <473E3DED.9020006@mac.com> References: <473E3DED.9020006@mac.com> Message-ID: <20071128162016.GC14425@puff.cnf.cornell.edu> Why are these files being removed, anyway? On Fri, Nov 16, 2007 at 05:03:41PM -0800, Darren Patterson wrote: > It seems that the libafsauthent and libafsrpc libraries aren't being > removed correctly when rebuilding openafs 1.4.5 from SRPM. This throws > an error before the RPMs are generated. Below is a patch for the spec file. > > -darren > > --- openafs.spec.orig 2007-11-16 16:36:01.000000000 -0800 > +++ openafs.spec 2007-11-16 16:36:28.000000000 -0800 > @@ -1163,8 +1163,8 @@ > done > > %if !%{build_authlibs} > -rm -f $RPM_BUILD_ROOT%{_libdir}/libafsauthent.so > -rm -f $RPM_BUILD_ROOT%{_libdir}/libafsrpc.so > +rm -f $RPM_BUILD_ROOT%{_libdir}/libafsauthent.so* > +rm -f $RPM_BUILD_ROOT%{_libdir}/libafsrpc.so* > %endif > > %endif > > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > -- ******************************** David William Botsch Programmer/Analyst CNF Computing botsch@cnf.cornell.edu ******************************** From shadow@gmail.com Wed Nov 28 16:25:10 2007 From: shadow@gmail.com (Derrick Brashear) Date: Wed, 28 Nov 2007 11:25:10 -0500 Subject: [OpenAFS] rhel3 openafs 1.4.5 spec patch In-Reply-To: <20071128162016.GC14425@puff.cnf.cornell.edu> References: <473E3DED.9020006@mac.com> <20071128162016.GC14425@puff.cnf.cornell.edu> Message-ID: ------=_Part_3341_3900583.1196267111026 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline If you build without the authlibs rpm, you are asking to not have these. Look at the change in the context below. On Nov 28, 2007 11:20 AM, Dave Botsch wrote: > Why are these files being removed, anyway? > > On Fri, Nov 16, 2007 at 05:03:41PM -0800, Darren Patterson wrote: > > It seems that the libafsauthent and libafsrpc libraries aren't being > > removed correctly when rebuilding openafs 1.4.5 from SRPM. This throws > > an error before the RPMs are generated. Below is a patch for the spec > file. > > > > -darren > > > > --- openafs.spec.orig 2007-11-16 16:36:01.000000000 -0800 > > +++ openafs.spec 2007-11-16 16:36:28.000000000 -0800 > > @@ -1163,8 +1163,8 @@ > > done > > > > %if !%{build_authlibs} > > -rm -f $RPM_BUILD_ROOT%{_libdir}/libafsauthent.so > > -rm -f $RPM_BUILD_ROOT%{_libdir}/libafsrpc.so > > +rm -f $RPM_BUILD_ROOT%{_libdir}/libafsauthent.so* > > +rm -f $RPM_BUILD_ROOT%{_libdir}/libafsrpc.so* > > %endif > > > > %endif > > > > > > _______________________________________________ > > OpenAFS-info mailing list > > OpenAFS-info@openafs.org > > https://lists.openafs.org/mailman/listinfo/openafs-info > > > > -- > ******************************** > David William Botsch > Programmer/Analyst > CNF Computing > botsch@cnf.cornell.edu > ******************************** > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > ------=_Part_3341_3900583.1196267111026 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline If you build without the authlibs rpm, you are asking to not have these. Look at the change in the context below.


On Nov 28, 2007 11:20 AM, Dave Botsch < botsch@cnf.cornell.edu> wrote:
Why are these files being removed, anyway?

On Fri, Nov 16, 2007 at 05:03:41PM -0800, Darren Patterson wrote:
> It seems that the libafsauthent and libafsrpc libraries aren't being
> removed correctly when rebuilding openafs 1.4.5 from SRPM.  This throws
> an error before the RPMs are generated.  Below is a patch for the spec file.
>
> -darren
>
> --- openafs.spec.orig   2007-11-16 16:36:01.000000000 -0800
> +++ openafs.spec        2007-11-16 16:36:28.000000000 -0800
> @@ -1163,8 +1163,8 @@
> done
>
> %if !%{build_authlibs}
> -rm -f $RPM_BUILD_ROOT%{_libdir}/libafsauthent.so
> -rm -f $RPM_BUILD_ROOT%{_libdir}/libafsrpc.so
> +rm -f $RPM_BUILD_ROOT%{_libdir}/libafsauthent.so*
> +rm -f $RPM_BUILD_ROOT%{_libdir}/libafsrpc.so*
> %endif
>
> %endif
>
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
>

--
********************************
David William Botsch
Programmer/Analyst
CNF Computing
botsch@cnf.cornell.edu
********************************
_______________________________________________
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

------=_Part_3341_3900583.1196267111026-- From jason@rampaginggeek.com Wed Nov 28 16:40:28 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Wed, 28 Nov 2007 11:40:28 -0500 Subject: [OpenAFS] Slow AFS performance In-Reply-To: <200711281442.lASEge8L028063@cmf.nrl.navy.mil> References: <200711281442.lASEge8L028063@cmf.nrl.navy.mil> Message-ID: <474D99FC.3060209@rampaginggeek.com> chas williams - CONTRACTOR wrote: > In message <474D78F2.40808@rampaginggeek.com>,Jason Edgecombe writes: > >> array >> >> My bonnie++ performance numbers for vicepa are here: >> http://www.coe.uncc.edu/~jwedgeco/bonnie.html >> >> >> What could AFS be doing that causes the performance to stink? >> > > run strace on fileserver/volserver and see what its doing as far > as i/o. as i recall, the fileserver doesnt always use the best > size for getting performance but it could be another issue. > > # truss -c -p 624 shows: syscall seconds calls errors read .054 1261 write .069 1258 close .016 1235 time .047 7468 fdsync .048 310 fcntl .009 622 lwp_park .017 1411 127 lwp_unpark .009 1190 mkdir .000 5 4 fchmod .002 311 fchown .004 311 yield .000 62 llseek .011 1555 fstat64 .006 604 open64 .060 1236 1 recvmsg .040 2133 sendmsg .065 2133 -------- ------ ---- sys totals: .465 23105 132 usr time: .157 elapsed: 16.030 A snippet of the long truss: Base time stamp: 1196267406.3976 [ Wed Nov 28 11:30:06 EST 2007 ] /5: 0.0520 fcntl(14, F_SETLK, 0xFED7B22C) = 0 /3: 0.0521 lwp_park(0xFEF7BE50, 0) Err#62 ETIME /5: 0.0522 close(14) = 0 /5: 0.0525 open64("/vicepc/AFSIDat/g/g6==U/7/o2/8Ua++Eeq1", O_RDWR|O_CREAT|O_TRUN C|O_EXCL, 0) = 14 /5: 0.0527 fchown(14, 1, 0) = 0 /5: 0.0527 fchmod(14, 0) = 0 /5: 0.0528 close(14) = 0 /5: 0.0528 open64("/vicepc/AFSIDat/g/g6==U/7/o2/8Ua++Eeq1", O_RDWR) = 14 /5: 0.0529 time() = 1196267406 /3: 0.0530 lwp_park(0xFEF7BE50, 0) = 0 /5: 0.0530 lwp_unpark(3, 1) = 0 /5: 0.0531 time() = 1196267406 /5: 0.0531 time() = 1196267406 /5: 0.0532 sendmsg(4, 0xFED7B258, 0) = 94 /2: 0.0532 recvmsg(4, 0xFF07BEAC, 0) = 94 /5: 0.0533 time() = 1196267406 /2: 0.0534 time() = 1196267406 /5: 0.0534 time() = 1196267406 /2: 0.0535 lwp_unpark(9, 1) = 0 /5: 0.0535 time() = 1196267406 /9: 0.0535 lwp_park(0x00000000, 0) = 0 /5: 0.0536 time() = 1196267406 /9: 0.0536 close(6) = 0 /2: 0.0537 recvmsg(4, 0xFF07BEAC, 0) = 91 /5: 0.0537 sendmsg(4, 0xFED7B258, 0) = 91 /2: 0.0538 time() = 1196267406 /5: 0.0539 time() = 1196267406 /9: 0.0538 open64("/vicepa/AFSIDat/g/g6==U/7/o2/CUa++Meq1", O_RDWR) = 6 /5: 0.0540 write(14, " G I F 8 9 a "01 &01F7\0".., 8192) = 8192 /9: 0.0541 fstat64(6, 0xFE97B8D8) = 0 /5: 0.0542 time() = 1196267406 /9: 0.0542 fstat64(6, 0xFE97B740) = 0 /5: 0.0542 time() = 1196267406 /5: 0.0543 time() = 1196267406 /5: 0.0543 sendmsg(4, 0xFED7B258, 0) = 88 /2: 0.0544 recvmsg(4, 0xFF07BEAC, 0) = 88 /5: 0.0544 time() = 1196267406 ...snip... /5: 0.1116 time() = 1196267406 /5: 0.1117 write(14, " G I F 8 9 aBD01 w01F7\0".., 8192) = 8192 /9: 0.1118 read(6, " G I F 8 9 a X\014\0F7\0".., 1351) = 1351 /5: 0.1118 time() = 1196267406 /9: 0.1119 close(6) = 0 /5: 0.1119 time() = 1196267406 /5: 0.1120 time() = 1196267406 /9: 0.1120 open64("/vicepa/AFSIDat/g/g6==U/7/o2/WUa+++fq1", O_RDWR) = 6 /5: 0.1121 sendmsg(4, 0xFED7B258, 0) = 88 /9: 0.1121 fstat64(6, 0xFE97B8D8) = 0 /2: 0.1121 recvmsg(4, 0xFF07BEAC, 0) = 88 /9: 0.1122 fstat64(6, 0xFE97B740) = 0 /5: 0.1122 time() = 1196267406 /2: 0.1123 time() = 1196267406 /5: 0.1124 time() = 1196267406 /5: 0.1125 time() = 1196267406 /9: 0.1125 read(6, " G I F 8 9 a d\0 <\0F7\0".., 1935) = 1935 /5: 0.1125 time() = 1196267406 /9: 0.1126 close(6) = 0 /2: 0.1127 recvmsg(4, 0xFF07BEAC, 0) = 85 /2: 0.1127 time() = 1196267406 /5: 0.1127 sendmsg(4, 0xFED7B258, 0) = 85 /9: 0.1127 open64("/vicepa/AFSIDat/g/g6==U/7/o2/YUa++2fq1", O_RDWR) = 6 /5: 0.1129 time() = 1196267406 /9: 0.1129 fstat64(6, 0xFE97B8D8) = 0 /9: 0.1129 fstat64(6, 0xFE97B740) = 0 /5: 0.1130 write(14, "06EA1BA8 oA0 >7FFA 4 jD4".., 8192) = 8192 /5: 0.1131 time() = 1196267406 /5: 0.1131 time() = 1196267406 /5: 0.1132 time() = 1196267406 /9: 0.1132 read(6, " G I F 8 9 a ~\0 <\0F7\0".., 2161) = 2161 /2: 0.1133 recvmsg(4, 0xFF07BEAC, 0) = 82 /5: 0.1132 sendmsg(4, 0xFED7B258, 0) = 82 /9: 0.1134 sendmsg(4, 0xFE97B448, 0) = 5688 /2: 0.1134 time() = 1196267406 /2: 0.1135 lwp_unpark(9, 1) = 0 /5: 0.1135 time() = 1196267406 ...snip... Any insights? Thanks, Jason From Jerry.Normandin@dafca.com Wed Nov 28 20:20:57 2007 From: Jerry.Normandin@dafca.com (Jerry Normandin) Date: Wed, 28 Nov 2007 12:20:57 -0800 Subject: [OpenAFS] openafs upgrade from 1.4.1 to 1.5.7 Message-ID: This is a multi-part message in MIME format. ------_=_NextPart_001_01C831FC.07E8ACF4 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hello I'm working on solving an AFS issue that has been a problem at my employer before I arrived here. Performance in general has been poor, much slower than NFS. I ran bonnie++ to check this out and apparently read/write performance was not bad, it was file creation, deletion and renaming that took forever. Looking further this was do to the fact when transactions like this are done, it invalidates the cache, so the filesystem must be reread. When you are in a software development environment this is a problem. Engineers checking out an entire trunk to their AFS home would take FOREVER. So digging further it appears that code Has been added to improve performance. We are running an old version of AFS.. 1.4.1. Are there any configuration differences between 1.4.1 and 1.5.7? Is there a FAQ posted regarding the upgrade? Can I have a mixed environment of versions? I'd like to update my replication servers first and run some tests Before I upgrade all the servers in the cell. =20 Thanks. =20 =20 ------_=_NextPart_001_01C831FC.07E8ACF4 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hello

   I’m working on solving an AFS = issue that has been a problem at my employer before I arrived here.  Performance in = general has been poor, much slower than NFS. I ran bonnie++ to check this out and apparently read/write performance was not bad, it was file creation, deletion and = renaming that took forever.  Looking further this was do to the fact when = transactions like this are done, it invalidates the cache, so the filesystem must be reread.  When you are in a software development environment this is = a problem.  Engineers checking out an entire trunk to their AFS home would take = FOREVER.  So digging further it appears that code Has been added to improve = performance.  We are running an old version of AFS.. 1.4.1.   Are there any = configuration differences between 1.4.1 and 1.5.7?

Is there a FAQ posted regarding the = upgrade?    Can I have a mixed environment of versions?  I’d like to update my = replication servers first and run some tests

Before I upgrade all the servers in the = cell.

 

Thanks.

 

 

------_=_NextPart_001_01C831FC.07E8ACF4-- From shadow@gmail.com Wed Nov 28 20:24:35 2007 From: shadow@gmail.com (Derrick Brashear) Date: Wed, 28 Nov 2007 15:24:35 -0500 Subject: [OpenAFS] openafs upgrade from 1.4.1 to 1.5.7 In-Reply-To: References: Message-ID: ------=_Part_4369_12615474.1196281475091 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Nov 28, 2007 3:20 PM, Jerry Normandin wrote: > Hello > > I'm working on solving an AFS issue that has been a problem at my > employer before I arrived here. Performance in general has been poor, much > slower than NFS. I ran bonnie++ to check this out and apparently read/write > performance was not bad, it was file creation, deletion and renaming that > took forever. > What's the underlying filesystem? AFS passes through the semantics of metadata operations of the underlying filesystem, and ext* for instance is poor at it. > Looking further this was do to the fact when transactions like this are > done, it invalidates the cache, so the filesystem must be reread. When you > are in a software development environment this is a problem. Engineers > checking out an entire trunk to their AFS home would take FOREVER. So > digging further it appears that code Has been added to improve performance. > > Yes. > We are running an old version of AFS.. 1.4.1. Are there any > configuration differences between 1.4.1 and 1.5.7? > Lots. Of course, we recommend 1.4.5, and not some random 1.5, especially not an old one. 1.5.7 is similarly old to 1.4.1. > Is there a FAQ posted regarding the upgrade? > "Replace the binaries and restart" > Can I have a mixed environment of versions? > Unless you have pts supergroups enabled, yes, though there is a pending bug regarding moving volumes between current 1.5.x and 1.4.x. > ------=_Part_4369_12615474.1196281475091 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On Nov 28, 2007 3:20 PM, Jerry Normandin <Jerry.Normandin@dafca.com> wrote:

Hello

   I'm working on solving an AFS issue that has been a problem at my employer before I arrived here.  Performance in general has been poor, much slower than NFS. I ran bonnie++ to check this out and apparently read/write performance was not bad, it was file creation, deletion and renaming that took forever. 


What's the underlying filesystem? AFS passes through the semantics of metadata operations of the underlying filesystem, and ext* for instance is poor at it.

Looking further this was do to the fact when transactions like this are done, it invalidates the cache, so the filesystem must be reread.  When you are in a software development environment this is a problem.  Engineers checking out an entire trunk to their AFS home would take FOREVER.  So digging further it appears that code Has been added to improve performance. 


Yes.

We are running an old version of AFS.. 1.4.1.   Are there any configuration differences between 1.4.1 and 1.5.7?


Lots. Of course, we recommend 1.4.5, and not some random 1.5, especially not an old one. 1.5.7 is similarly old to 1.4.1.
 

Is there a FAQ posted regarding the upgrade?   


"Replace the binaries and restart"

Can I have a mixed environment of versions? 

Unless you have pts supergroups enabled, yes, though there is a pending bug regarding moving volumes between current 1.5.x and 1.4.x.

 


------=_Part_4369_12615474.1196281475091-- From daleg@umbc.edu Thu Nov 29 03:00:22 2007 From: daleg@umbc.edu (Dale Ghent) Date: Wed, 28 Nov 2007 22:00:22 -0500 Subject: [OpenAFS] Slow AFS performance In-Reply-To: <474D78F2.40808@rampaginggeek.com> References: <474D78F2.40808@rampaginggeek.com> Message-ID: On Nov 28, 2007, at 9:19 AM, Jason Edgecombe wrote: > I'm experiencing poor AFS performance on under Sparc solaris 9 09/05HW > running Openafs server 1.4.1 on a Sun StorageTeck 3511 Fibre > channel to > SATA array > > At first, I thought that having UFS logging disabled was the culprit, > but I have enabled UFS logging and I am using the namei server, but > performance still stinks. It took 1.5 hours to move a 3.2GB volume to > the server. Things seem fine except on the Fibre channel disks. ... > My bonnie++ performance numbers for vicepa are here: > http://www.coe.uncc.edu/~jwedgeco/bonnie.html > > What could AFS be doing that causes the performance to stink? Well first, your bonnie run tested sequential IO, so of course you're going to get stellar numbers with sequential large writes and reads. Arrays and disks in general eat that stuff right on up. AFS namei does writes in small batches, up to 64k, and all of it random (think: lots of users accessing lots of small files) It would be advantageous to tun to this environment. Check out your 3511s first. 1) Make sure write-back caching is on. This is a global as well as a per-Logical Drive setting on the 3510/3511. 2) Optimized for Random I/O (be sure the latest 3511 fw is applied, it betters the performance of this setting) 3) You didn't mention the type of RAID config you have on the 3511s. RAID5? If so, is the stripe size low (64k?) Some 'iostat -nx 1' output would be helpful, too. The wsvc_t and %b columns would be telling. /dale From rra@stanford.edu Thu Nov 29 04:03:09 2007 From: rra@stanford.edu (Russ Allbery) Date: Wed, 28 Nov 2007 20:03:09 -0800 Subject: [OpenAFS-Doc] Re: [OpenAFS] Quick Start Guide updated for Kerberos v5 In-Reply-To: <471A2B44.1010900@rampaginggeek.com> (Jason Edgecombe's message of "Sat\, 20 Oct 2007 12\:22\:28 -0400") References: <471A2B44.1010900@rampaginggeek.com> Message-ID: <8763zlpuki.fsf@windlord.stanford.edu> Jason Edgecombe writes: > chapter1: > * about upgrading OS: > **Should the namei fileserver be mentioned? Is namei the > recommended way? inode is still recommended for Solaris. namei is recommended in all other cases, and generally is the only possible method. > chapter 2: > * getting started on solaris: > ** it still mentions copying files from cd-rom (grep for "CD-ROM") Yeah, this needs to get fixed throughout the manual, replaced with instructions about how to start from the downloaded binary build or to build from source. > **only mentions solaris 7, it should mention 8, 9 & > 10/opensolaris or just say 7 & later versions Yup. > ** about fsck: does solaris use inode, namei or both? Is > clarification needed? Solaris can use either, so yes, clarification is needed. I'm fairly sure you don't need the custom fsck if you use namei. > "The entry for AFS server processes, called either *afs* or > *afs//cell/*. No user logs in under this identity, but it is used to > encrypt the server tickets that granted to AFS clients for presentation > to server processes during mutual authentication." > > should that be "...that are granted to AFS clients..."? Yup. The source for this is in DocBook in the OpenAFS CVS head, and patches are certainly welcome. -- Russ Allbery (rra@stanford.edu) From shadow@gmail.com Thu Nov 29 04:53:59 2007 From: shadow@gmail.com (Derrick Brashear) Date: Wed, 28 Nov 2007 23:53:59 -0500 Subject: [OpenAFS-Doc] Re: [OpenAFS] Quick Start Guide updated for Kerberos v5 In-Reply-To: <8763zlpuki.fsf@windlord.stanford.edu> References: <471A2B44.1010900@rampaginggeek.com> <8763zlpuki.fsf@windlord.stanford.edu> Message-ID: ------=_Part_5977_18700542.1196312039171 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Nov 28, 2007 11:03 PM, Russ Allbery wrote: > Jason Edgecombe writes: > > > chapter1: > > * about upgrading OS: > > **Should the namei fileserver be mentioned? Is namei the > > recommended way? > > inode is still recommended for Solaris. namei is recommended in all other > cases, and generally is the only possible method. > Actually, inode also still works on HPUX 11 and IRIX 6.5, and we only do inode builds for IRIX. > > > chapter 2: > > * getting started on solaris: > > ** it still mentions copying files from cd-rom (grep for > "CD-ROM") > > Yeah, this needs to get fixed throughout the manual, replaced with > instructions about how to start from the downloaded binary build or to > build from source. > > > **only mentions solaris 7, it should mention 8, 9 & > > 10/opensolaris or just say 7 & later versions > > Yup. > > > ** about fsck: does solaris use inode, namei or both? Is > > clarification needed? > > Solaris can use either, so yes, clarification is needed. I'm fairly sure > you don't need the custom fsck if you use namei. > Correct. > > > "The entry for AFS server processes, called either *afs* or > > *afs//cell/*. No user logs in under this identity, but it is used to > > encrypt the server tickets that granted to AFS clients for presentation > > to server processes during mutual authentication." > > > > should that be "...that are granted to AFS clients..."? > > Yup. > > The source for this is in DocBook in the OpenAFS CVS head, and patches are > certainly welcome. > > ------=_Part_5977_18700542.1196312039171 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On Nov 28, 2007 11:03 PM, Russ Allbery <rra@stanford.edu> wrote:
Jason Edgecombe <jason@rampaginggeek.com> writes:

> chapter1:
>     * about upgrading OS:
>         **Should the namei fileserver be mentioned? Is namei the
> recommended way?

inode is still recommended for Solaris.  namei is recommended in all other
cases, and generally is the only possible method.

Actually, inode also still works on HPUX 11 and IRIX 6.5, and we only do inode builds for IRIX.

> chapter 2:
>   * getting started on solaris:
>         ** it still mentions copying files from cd-rom (grep for "CD-ROM")

Yeah, this needs to get fixed throughout the manual, replaced with
instructions about how to start from the downloaded binary build or to
build from source.

>         **only mentions solaris 7, it should mention 8, 9 &
> 10/opensolaris or just say 7 & later versions

Yup.

>        ** about fsck: does solaris use inode, namei or both? Is
> clarification needed?

Solaris can use either, so yes, clarification is needed.  I'm fairly sure
you don't need the custom fsck if you use namei.

Correct.

> "The entry for AFS server processes, called either *afs* or
> *afs//cell/*. No user logs in under this identity, but it is used to
> encrypt the server tickets that granted to AFS clients for presentation
> to server processes during mutual authentication."
>
> should that be "...that are granted to AFS clients..."?

Yup.

The source for this is in DocBook in the OpenAFS CVS head, and patches are
certainly welcome.


------=_Part_5977_18700542.1196312039171-- From haba@kth.se Thu Nov 29 09:29:15 2007 From: haba@kth.se (Harald Barth) Date: Thu, 29 Nov 2007 10:29:15 +0100 (CET) Subject: [OpenAFS] openafs upgrade from 1.4.1 to 1.5.7 In-Reply-To: References: Message-ID: <20071129.102915.181841539.haba@habanero.pdc.kth.se> > What's the underlying filesystem? AFS passes through the semantics of > metadata operations of the underlying filesystem, and ext* for instance is > poor at it. For example xfs on Linux has worked well for us. > > We are running an old version of AFS.. 1.4.1. Are there any > > configuration differences between 1.4.1 and 1.5.7? > > > > Lots. Of course, we recommend 1.4.5, and not some random 1.5, especially not > an old one. 1.5.7 is similarly old to 1.4.1. As Derrick said, use 1.4.x for servers. > > Can I have a mixed environment of versions? > Unless you have pts supergroups enabled, yes, though there is a pending bug > regarding moving volumes between current 1.5.x and 1.4.x. Better than that. You can mix even if you have supergroups enabled, but don't create any supergroups until all your servers are upgraded. There is no need for a complete cell shutdown to perform an upgrade. There has not been such a need for years. Harald. From haba@kth.se Thu Nov 29 09:37:59 2007 From: haba@kth.se (Harald Barth) Date: Thu, 29 Nov 2007 10:37:59 +0100 (CET) Subject: [OpenAFS-Doc] Re: [OpenAFS] Quick Start Guide updated for Kerberos v5 In-Reply-To: <8763zlpuki.fsf@windlord.stanford.edu> References: <471A2B44.1010900@rampaginggeek.com> <8763zlpuki.fsf@windlord.stanford.edu> Message-ID: <20071129.103759.212267739.haba@habanero.pdc.kth.se> > inode is still recommended for Solaris. namei is recommended in all other > cases, and generally is the only possible method. I would recommend namei for all new installations. Is there any reason against that which I'm not aware of? Harald. From mginkel@mpi-magdeburg.mpg.de Thu Nov 29 10:07:29 2007 From: mginkel@mpi-magdeburg.mpg.de (Martin Ginkel) Date: Thu, 29 Nov 2007 11:07:29 +0100 Subject: [OpenAFS] What's the problem with reiser Message-ID: <474E8F61.6030509@mpi-magdeburg.mpg.de> Hi, We use openafs clients on a lot of machines. The local Filesystems are usually reiser. But for the DiskCache we have to install one partition with ext2. Why is that? What's the problem with reiser as cacheFS ? Just curious Martin From l.schimmer@cgv.tugraz.at Thu Nov 29 10:58:49 2007 From: l.schimmer@cgv.tugraz.at (Lars Schimmer) Date: Thu, 29 Nov 2007 11:58:49 +0100 Subject: [OpenAFS] MacOS X 1.4 (tiger) PPC built of 1.5.27? Message-ID: <474E9B69.4080602@cgv.tugraz.at> Hi! I' ve updated our PowerPC Mac with latest available OpenAFS version for=20 PPC Mac OS X Tiger, which was 1.5.26. For bad sake, MacOS X just went bad every shutdown and OpenAFS seems to=20 be the reason because of which MacOS X won't shutdown clear and ends in=20 a kernel panic. Is there any 1.5.27 for MacOS Tiger PPC available? Maybe that bu is fixed. Btw: I clicked send that bug to apple, does OpenAFS get these error=20 reports just like on win ? MfG, Lars Schimmer --=20 ------------------------------------------------------------- TU Graz, Institut f=FCr ComputerGraphik & WissensVisualisierung Tel: +43 316 873-5405 E-Mail: l.schimmer@cgv.tugraz.at Fax: +43 316 873-5402 PGP-Key-ID: 0x4A9B1723 From haba@kth.se Thu Nov 29 12:24:50 2007 From: haba@kth.se (Harald Barth) Date: Thu, 29 Nov 2007 13:24:50 +0100 (CET) Subject: [OpenAFS] What's the problem with reiser In-Reply-To: <474E8F61.6030509@mpi-magdeburg.mpg.de> References: <474E8F61.6030509@mpi-magdeburg.mpg.de> Message-ID: <20071129.132450.222201899.haba@habanero.pdc.kth.se> > > We use openafs clients on a lot of machines. The local Filesystems are > usually reiser. But for the DiskCache we have to install one partition > with ext2. To all my experience, reiserfs is broken. I recommend NOT to use that file system. At all. As a cache file system ext2 is fine, because it is fast, most kernels have the drivers and if it breaks because of a system crash, so what, it was just a cache. Harald. From dirk.heinrichs.ext@nsn.com Thu Nov 29 12:41:16 2007 From: dirk.heinrichs.ext@nsn.com (Dirk Heinrichs) Date: Thu, 29 Nov 2007 13:41:16 +0100 Subject: [OpenAFS] What's the problem with reiser In-Reply-To: <20071129.132450.222201899.haba@habanero.pdc.kth.se> References: <474E8F61.6030509@mpi-magdeburg.mpg.de> <20071129.132450.222201899.haba@habanero.pdc.kth.se> Message-ID: <200711291341.21115.dirk.heinrichs.ext@nsn.com> --nextPart3268345.tESlmKZYcz Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Am Donnerstag, 29. November 2007 schrieb ext Harald Barth: > > We use openafs clients on a lot of machines. The local Filesystems are > > usually reiser. But for the DiskCache we have to install one partition > > with ext2. > > To all my experience, reiserfs is broken. I recommend NOT to use that > file system. At all. OK. replase reiser with xfs, jfs, whatever. I guess the real question was:= =20 What's the reason why one should not use other filesystems than ext2 for=20 the cache partition on a Linux client? Bye... Dirk =2D-=20 Dirk Heinrichs | Tel: +49 (0)162 234 3408 Configuration Manager | Fax: +49 (0)211 47068 111 Capgemini Deutschland | Mail: dirk.heinrichs@capgemini.com Wanheimerstra=C3=9Fe 68 | Web: http://www.capgemini.com D-40468 D=C3=BCsseldorf | ICQ#: 110037733 GPG Public Key C2E467BB | Keyserver: www.keyserver.net --nextPart3268345.tESlmKZYcz Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.7 (GNU/Linux) iD8DBQBHTrNx8NVtnsLkZ7sRAkNxAJwLDsWRcJvAwMbW4Ige5AWj/l35ygCglaiN Y4clXA6Qd5SvHTBCgoBWmmw= =Wu6m -----END PGP SIGNATURE----- --nextPart3268345.tESlmKZYcz-- From deengert@anl.gov Thu Nov 29 12:51:53 2007 From: deengert@anl.gov (Douglas E. Engert) Date: Thu, 29 Nov 2007 06:51:53 -0600 Subject: [OpenAFS-Doc] Re: [OpenAFS] Quick Start Guide updated for Kerberos v5 In-Reply-To: <8763zlpuki.fsf@windlord.stanford.edu> References: <471A2B44.1010900@rampaginggeek.com> <8763zlpuki.fsf@windlord.stanford.edu> Message-ID: <474EB5E9.2070803@anl.gov> Russ Allbery wrote: > Jason Edgecombe writes: > >> chapter1: >> * about upgrading OS: >> **Should the namei fileserver be mentioned? Is namei the >> recommended way? > > inode is still recommended for Solaris. namei is recommended in all other > cases, and generally is the only possible method. You should consider recommending namei on Solaris too. inode only works on ufs and you must have logging turned off. If you want to use zfs then you must use namei. > >> chapter 2: >> * getting started on solaris: >> ** it still mentions copying files from cd-rom (grep for "CD-ROM") > > Yeah, this needs to get fixed throughout the manual, replaced with > instructions about how to start from the downloaded binary build or to > build from source. > >> **only mentions solaris 7, it should mention 8, 9 & >> 10/opensolaris or just say 7 & later versions > > Yup. > >> ** about fsck: does solaris use inode, namei or both? Is >> clarification needed? > > Solaris can use either, so yes, clarification is needed. I'm fairly sure > you don't need the custom fsck if you use namei. Correct. It makes me feel more comfortable using the vendor's fsck rather then a modified fsck. > >> "The entry for AFS server processes, called either *afs* or >> *afs//cell/*. No user logs in under this identity, but it is used to >> encrypt the server tickets that granted to AFS clients for presentation >> to server processes during mutual authentication." >> >> should that be "...that are granted to AFS clients..."? > > Yup. > > The source for this is in DocBook in the OpenAFS CVS head, and patches are > certainly welcome. > -- Douglas E. Engert Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439 (630) 252-5444 From rob@nofocus.org Thu Nov 29 13:24:30 2007 From: rob@nofocus.org (Rob Banz) Date: Thu, 29 Nov 2007 08:24:30 -0500 Subject: [OpenAFS] What's the problem with reiser In-Reply-To: <200711291341.21115.dirk.heinrichs.ext@nsn.com> References: <474E8F61.6030509@mpi-magdeburg.mpg.de> <20071129.132450.222201899.haba@habanero.pdc.kth.se> <200711291341.21115.dirk.heinrichs.ext@nsn.com> Message-ID: On Nov 29, 2007, at 07:41, Dirk Heinrichs wrote: > Am Donnerstag, 29. November 2007 schrieb ext Harald Barth: >>> We use openafs clients on a lot of machines. The local Filesystems >>> are >>> usually reiser. But for the DiskCache we have to install one >>> partition >>> with ext2. >> >> To all my experience, reiserfs is broken. I recommend NOT to use that >> file system. At all. > > OK. replase reiser with xfs, jfs, whatever. I guess the real > question was: > What's the reason why one should not use other filesystems than ext2 > for > the cache partition on a Linux client? For a cache partition, at least on other *ixes, the cache partition has always needed special attention because of the way its used by the AFS kernel module. Certain care has to be taken as to do operations in such a way that kernel deadlocks and such are avoided. For example, on Solaris you use ufs, however, you can't use logging ufs because of known deadlock problems. I'd assume that the use of ext2 on Linux is for a similar reason. -rob From chas@cmf.nrl.navy.mil Thu Nov 29 13:36:31 2007 From: chas@cmf.nrl.navy.mil (chas williams - CONTRACTOR) Date: Thu, 29 Nov 2007 08:36:31 -0500 Subject: [OpenAFS] What's the problem with reiser In-Reply-To: <20071129.132450.222201899.haba@habanero.pdc.kth.se> Message-ID: <200711291336.lATDaV38009074@cmf.nrl.navy.mil> In message <20071129.132450.222201899.haba@habanero.pdc.kth.se>,Harald Barth wr ites: >> We use openafs clients on a lot of machines. The local Filesystems are >> usually reiser. But for the DiskCache we have to install one partition >> with ext2. > >To all my experience, reiserfs is broken. I recommend NOT to use that >file system. At all. As a cache file system ext2 is fine, because it as i recall resiferfs doesnt work because it doesnt keep a fixed mapping between file objects and what afs would consider the inode. i believe people have been lucky with using a reisferfs cache filesystem but it had to be on a seperate partition. normally, its journaling that creates trouble for caching filesystems. personally, unless you have a need for massive amounts of cache, use memcache. search through the list archives for a better answers about this. this info doesnt appear to be in the wiki, so perhaps it needs one. (and one that is more correct than my vague ramblings). From chas@cmf.nrl.navy.mil Thu Nov 29 13:39:06 2007 From: chas@cmf.nrl.navy.mil (chas williams - CONTRACTOR) Date: Thu, 29 Nov 2007 08:39:06 -0500 Subject: [OpenAFS-Doc] Re: [OpenAFS] Quick Start Guide updated for Kerberos v5 In-Reply-To: <8763zlpuki.fsf@windlord.stanford.edu> Message-ID: <200711291339.lATDd6Pj009101@cmf.nrl.navy.mil> In message <8763zlpuki.fsf@windlord.stanford.edu>,Russ Allbery writes: >> ** about fsck: does solaris use inode, namei or both? Is >> clarification needed? > >Solaris can use either, so yes, clarification is needed. I'm fairly sure >you don't need the custom fsck if you use namei. you do not need the custom fsck for namei. further, namei only works for ufs nonlogging filesystems. if you have say zfs perhaps, namei is your only choice. given this, i think its reasonable to suggest that people just use namei only. From rob@nofocus.org Thu Nov 29 13:44:03 2007 From: rob@nofocus.org (Rob Banz) Date: Thu, 29 Nov 2007 08:44:03 -0500 Subject: [OpenAFS-Doc] Re: [OpenAFS] Quick Start Guide updated for Kerberos v5 In-Reply-To: <200711291339.lATDd6Pj009101@cmf.nrl.navy.mil> References: <200711291339.lATDd6Pj009101@cmf.nrl.navy.mil> Message-ID: On Nov 29, 2007, at 08:39, chas williams - CONTRACTOR wrote: > In message <8763zlpuki.fsf@windlord.stanford.edu>,Russ Allbery writes: >>> ** about fsck: does solaris use inode, namei or both? Is >>> clarification needed? >> >> Solaris can use either, so yes, clarification is needed. I'm >> fairly sure >> you don't need the custom fsck if you use namei. > > you do not need the custom fsck for namei. further, namei only works > for ufs nonlogging filesystems. if you have say zfs perhaps, namei > is your only choice. given this, i think its reasonable to suggest > that people just use namei only Is inode that only works on unlogging ufs, namei works on logging, unlogging, zfs, etc. ;) -rob From jason@rampaginggeek.com Thu Nov 29 13:46:57 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Thu, 29 Nov 2007 08:46:57 -0500 Subject: [OpenAFS] What's the problem with reiser In-Reply-To: References: <474E8F61.6030509@mpi-magdeburg.mpg.de> <20071129.132450.222201899.haba@habanero.pdc.kth.se> <200711291341.21115.dirk.heinrichs.ext@nsn.com> Message-ID: <474EC2D1.1060407@rampaginggeek.com> Rob Banz wrote: > > On Nov 29, 2007, at 07:41, Dirk Heinrichs wrote: > >> Am Donnerstag, 29. November 2007 schrieb ext Harald Barth: >>>> We use openafs clients on a lot of machines. The local Filesystems are >>>> usually reiser. But for the DiskCache we have to install one partition >>>> with ext2. >>> >>> To all my experience, reiserfs is broken. I recommend NOT to use that >>> file system. At all. >> >> OK. replase reiser with xfs, jfs, whatever. I guess the real question >> was: >> What's the reason why one should not use other filesystems than ext2 for >> the cache partition on a Linux client? > > For a cache partition, at least on other *ixes, the cache partition > has always needed special attention because of the way its used by the > AFS kernel module. Certain care has to be taken as to do operations > in such a way that kernel deadlocks and such are avoided. For > example, on Solaris you use ufs, however, you can't use logging ufs > because of known deadlock problems. > > I'd assume that the use of ext2 on Linux is for a similar reason. > > -rob Fascinating. I did not know of UFS logging issue on the cache partition. Strangely, I haven't heard of any issues. does ext3 have this issue as well? Thanks, Jason From shadow@gmail.com Thu Nov 29 13:56:47 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 29 Nov 2007 08:56:47 -0500 Subject: [OpenAFS] MacOS X 1.4 (tiger) PPC built of 1.5.27? In-Reply-To: <474E9B69.4080602@cgv.tugraz.at> References: <474E9B69.4080602@cgv.tugraz.at> Message-ID: ------=_Part_7597_16405545.1196344607282 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Nov 29, 2007 5:58 AM, Lars Schimmer wrote: > Hi! > > I' ve updated our PowerPC Mac with latest available OpenAFS version for > PPC Mac OS X Tiger, which was 1.5.26. > For bad sake, MacOS X just went bad every shutdown and OpenAFS seems to > be the reason because of which MacOS X won't shutdown clear and ends in > a kernel panic. > > Is there any 1.5.27 for MacOS Tiger PPC available? Maybe that bu is fixed. > It is, however, no build is available currently. What are you using tht's a 1.5 feature? > Btw: I clicked send that bug to apple, does OpenAFS get these error > reports just like on win ? > We don't. ------=_Part_7597_16405545.1196344607282 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On Nov 29, 2007 5:58 AM, Lars Schimmer <l.schimmer@cgv.tugraz.at> wrote:
Hi!

I' ve updated our PowerPC Mac with latest available OpenAFS version for
PPC Mac OS X Tiger, which was 1.5.26.
For bad sake, MacOS X just went bad every shutdown and OpenAFS seems to
be the reason because of which MacOS X won't shutdown clear and ends in
a kernel panic.

Is there any 1.5.27 for MacOS Tiger PPC available? Maybe that bu is fixed.

It is, however, no build is available currently. What are you using tht's a 1.5 feature?
 
Btw: I clicked send that bug to apple, does OpenAFS get these error
reports just like on win ?
We don't.

 

------=_Part_7597_16405545.1196344607282-- From lara.lloret.iglesias@cern.ch Thu Nov 29 13:56:54 2007 From: lara.lloret.iglesias@cern.ch (Lara Lloret Iglesias) Date: Thu, 29 Nov 2007 14:56:54 +0100 Subject: [OpenAFS] Automatically get AFS-token Message-ID: <9A6A62B6B84859469F3EBB5F09D818CA99416F@cernxchg50.cern.ch> Hello! I=B4ve just installed three AFS clients, and everything seems to work = fine! The only problem is that I would like to get my afs-token on = login, without having to type klog. Any idea? Thank you very much, Lara From rob@nofocus.org Thu Nov 29 13:57:30 2007 From: rob@nofocus.org (Rob Banz) Date: Thu, 29 Nov 2007 08:57:30 -0500 Subject: [OpenAFS] What's the problem with reiser In-Reply-To: <474EC2D1.1060407@rampaginggeek.com> References: <474E8F61.6030509@mpi-magdeburg.mpg.de> <20071129.132450.222201899.haba@habanero.pdc.kth.se> <200711291341.21115.dirk.heinrichs.ext@nsn.com> <474EC2D1.1060407@rampaginggeek.com> Message-ID: >>> >> >> For a cache partition, at least on other *ixes, the cache partition >> has always needed special attention because of the way its used by >> the >> AFS kernel module. Certain care has to be taken as to do operations >> in such a way that kernel deadlocks and such are avoided. For >> example, on Solaris you use ufs, however, you can't use logging ufs >> because of known deadlock problems. >> >> I'd assume that the use of ext2 on Linux is for a similar reason. >> >> -rob > Fascinating. I did not know of UFS logging issue on the cache > partition. > Strangely, I haven't heard of any issues. does ext3 have this issue > as well? I had used logging ufs as a cache partition for years without a problem as well -- but in the past couple years ran into deadlocks. I remember reliably seeing them under Solaris 10x86 on a Dell 2650 where it'd lock up right after AFS started and some automated processes were busy trying to access it. -rob From l.schimmer@cgv.tugraz.at Thu Nov 29 14:01:56 2007 From: l.schimmer@cgv.tugraz.at (Lars Schimmer) Date: Thu, 29 Nov 2007 15:01:56 +0100 Subject: [OpenAFS] MacOS X 1.4 (tiger) PPC built of 1.5.27? In-Reply-To: References: <474E9B69.4080602@cgv.tugraz.at> Message-ID: <474EC654.9040704@cgv.tugraz.at> Derrick Brashear wrote: > On Nov 29, 2007 5:58 AM, Lars Schimmer wrote= : >=20 >> Hi! >> >> I' ve updated our PowerPC Mac with latest available OpenAFS version fo= r >> PPC Mac OS X Tiger, which was 1.5.26. >> For bad sake, MacOS X just went bad every shutdown and OpenAFS seems t= o >> be the reason because of which MacOS X won't shutdown clear and ends i= n >> a kernel panic. >> >> Is there any 1.5.27 for MacOS Tiger PPC available? Maybe that bu is fi= xed. >> >=20 > It is, however, no build is available currently. What are you using tht= 's a > 1.5 feature? I just made good experiece with 1.5.27 win build and thought install=20 latest mac client to get latest usefull features. If no public 1.5.27 is available, I should go back to latest stable=20 built, or which one should be preferred? >> Btw: I clicked send that bug to apple, does OpenAFS get these error >> reports just like on win ? >> > We don't. >=20 MfG, Lars Schimmer --=20 ------------------------------------------------------------- TU Graz, Institut f=FCr ComputerGraphik & WissensVisualisierung Tel: +43 316 873-5405 E-Mail: l.schimmer@cgv.tugraz.at Fax: +43 316 873-5402 PGP-Key-ID: 0x4A9B1723 From shadow@gmail.com Thu Nov 29 14:03:32 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 29 Nov 2007 09:03:32 -0500 Subject: [OpenAFS] MacOS X 1.4 (tiger) PPC built of 1.5.27? In-Reply-To: <474EC654.9040704@cgv.tugraz.at> References: <474E9B69.4080602@cgv.tugraz.at> <474EC654.9040704@cgv.tugraz.at> Message-ID: ------=_Part_7611_16845677.1196345012314 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Nov 29, 2007 9:01 AM, Lars Schimmer wrote: > Derrick Brashear wrote: > > On Nov 29, 2007 5:58 AM, Lars Schimmer wrote: > > > >> Hi! > >> > >> I' ve updated our PowerPC Mac with latest available OpenAFS version for > >> PPC Mac OS X Tiger, which was 1.5.26. > >> For bad sake, MacOS X just went bad every shutdown and OpenAFS seems to > >> be the reason because of which MacOS X won't shutdown clear and ends in > >> a kernel panic. > >> > >> Is there any 1.5.27 for MacOS Tiger PPC available? Maybe that bu is > fixed. > >> > > > > It is, however, no build is available currently. What are you using > tht's a > > 1.5 feature? > > I just made good experiece with 1.5.27 win build and thought install > latest mac client to get latest usefull features. > If no public 1.5.27 is available, I should go back to latest stable > built, or which one should be preferred? > In general the preferred build is listed at http://www.openafs.org/macos.html A Tiger 1.5.28 build will be issued when that is released. ------=_Part_7611_16845677.1196345012314 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On Nov 29, 2007 9:01 AM, Lars Schimmer <l.schimmer@cgv.tugraz.at> wrote:
Derrick Brashear wrote:
> On Nov 29, 2007 5:58 AM, Lars Schimmer <l.schimmer@cgv.tugraz.at> wrote:
>
>> Hi!
>>
>> I' ve updated our PowerPC Mac with latest available OpenAFS version for
>> PPC Mac OS X Tiger, which was 1.5.26.
>> For bad sake, MacOS X just went bad every shutdown and OpenAFS seems to
>> be the reason because of which MacOS X won't shutdown clear and ends in
>> a kernel panic.
>>
>> Is there any 1.5.27 for MacOS Tiger PPC available? Maybe that bu is fixed.
>>
>
> It is, however, no build is available currently. What are you using tht's a
> 1.5 feature?

I just made good experiece with 1.5.27 win build and thought install
latest mac client to get latest usefull features.
If no public 1.5.27 is available, I should go back to latest stable
built, or which one should be preferred?
In general the preferred build is listed at http://www.openafs.org/macos.html
 
A Tiger 1.5.28 build will be issued when that is released.


------=_Part_7611_16845677.1196345012314-- From shadow@gmail.com Thu Nov 29 14:04:20 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 29 Nov 2007 09:04:20 -0500 Subject: [OpenAFS] What's the problem with reiser In-Reply-To: References: <474E8F61.6030509@mpi-magdeburg.mpg.de> <20071129.132450.222201899.haba@habanero.pdc.kth.se> <200711291341.21115.dirk.heinrichs.ext@nsn.com> <474EC2D1.1060407@rampaginggeek.com> Message-ID: ------=_Part_7614_14698359.1196345060502 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Nov 29, 2007 8:57 AM, Rob Banz wrote: > >>> > >> > >> For a cache partition, at least on other *ixes, the cache partition > >> has always needed special attention because of the way its used by > >> the > >> AFS kernel module. Certain care has to be taken as to do operations > >> in such a way that kernel deadlocks and such are avoided. For > >> example, on Solaris you use ufs, however, you can't use logging ufs > >> because of known deadlock problems. > >> > >> I'd assume that the use of ext2 on Linux is for a similar reason. > >> > >> -rob > > Fascinating. I did not know of UFS logging issue on the cache > > partition. > > Strangely, I haven't heard of any issues. does ext3 have this issue > > as well? > > I had used logging ufs as a cache partition for years without a > problem as well -- but in the past couple years ran into deadlocks. I > remember reliably seeing them under Solaris 10x86 on a Dell 2650 where > it'd lock up right after AFS started and some automated processes were > busy trying to access it. > For documentation purposes it might be interesting to get kernel backtraces of those if you ever get bored. ------=_Part_7614_14698359.1196345060502 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On Nov 29, 2007 8:57 AM, Rob Banz <rob@nofocus.org> wrote:
>>>
>>
>> For a cache partition, at least on other *ixes, the cache partition
>> has always needed special attention because of the way its used by
>> the
>> AFS kernel module.  Certain care has to be taken as to do operations
>> in such a way that kernel deadlocks and such are avoided.  For
>> example, on Solaris you use ufs, however, you can't use logging ufs
>> because of known deadlock problems.
>>
>> I'd assume that the use of ext2 on Linux is for a similar reason.
>>
>> -rob
> Fascinating. I did not know of UFS logging issue on the cache
> partition.
> Strangely, I haven't heard of any issues. does ext3 have this issue
> as well?

I had used logging ufs as a cache partition for years without a
problem as well -- but in the past couple years ran into deadlocks.  I
remember reliably seeing them under Solaris 10x86 on a Dell 2650 where
it'd lock up right after AFS started and some automated processes were
busy trying to access it.

For documentation purposes it might be interesting to get kernel backtraces of those if you ever get bored.
 

------=_Part_7614_14698359.1196345060502-- From jaltman@secure-endpoints.com Thu Nov 29 15:14:04 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Thu, 29 Nov 2007 16:14:04 +0100 Subject: [OpenAFS-Doc] Re: [OpenAFS] Quick Start Guide updated for Kerberos v5 In-Reply-To: <200711291339.lATDd6Pj009101@cmf.nrl.navy.mil> References: <200711291339.lATDd6Pj009101@cmf.nrl.navy.mil> Message-ID: <474ED73C.5090407@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms050201080000030201040401 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit chas williams - CONTRACTOR wrote: > In message <8763zlpuki.fsf@windlord.stanford.edu>,Russ Allbery writes: >>> ** about fsck: does solaris use inode, namei or both? Is >>> clarification needed? >> Solaris can use either, so yes, clarification is needed. I'm fairly sure >> you don't need the custom fsck if you use namei. > > you do not need the custom fsck for namei. further, namei only works > for ufs nonlogging filesystems. if you have say zfs perhaps, namei > is your only choice. given this, i think its reasonable to suggest > that people just use namei only. Except that when you use memcache you lose the benefits of the cache between restarts. There are many organizations that use cache sizes large enough so that the entire 90+% of the data needed for the operating system and applications comes from the cache. After a restart all that is then required is for a series of FetchStatus calls to be performed. Jeffrey Altman --------------ms050201080000030201040401 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMjkxNTE0MDRaMCMGCSqGSIb3DQEJBDEWBBSvUYSW maPiFgwKvOfo8Apy7p3T9DBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAaV0yLsVDu2jjsTk9N5bBcVs+ay0P5gciY8BFY/1wDEmhrJARCCjB VQmY+GBG1M/hFkydl5R3D8AEjQF9N7ec4sQRkDbVfDEn0wP4nrR10yXtIWRMiw/Ij+wafUDT H2JaXc0DdiRPzLPZczZLHV6YWAuFqvWItslfQBIEMnq9TM/FxaiLXZWbsDZYGgwknf6Vs3wL 2fZ4ntkUP4/H2n2MEO0/oS2Ha0YxMd0Rkvu5+9j4D2ur8IRrhDzU7kRn2dxHFn8vz0aknom0 fnTaGk2xO2msVwh1D2eaBYo85yTmDjhrSmmKEnpyCeCkKEDSobhlbY9zQspAGejqx2B+0mqB FwAAAAAAAA== --------------ms050201080000030201040401-- From jarausch@igpm.rwth-aachen.de Thu Nov 29 15:31:29 2007 From: jarausch@igpm.rwth-aachen.de (Helmut Jarausch) Date: Thu, 29 Nov 2007 16:31:29 +0100 (CET) Subject: [OpenAFS] openafs partition - how to increase Message-ID: Hi, I'd like to resize (enlarge) the ext2-partition on which e.g. /vicepa is mounted. I know how to enlarge an ext2 partition using Linux tools, but how does OpenAFS react to this situtation? Does it use the additional space or is there any danger OpenAFS gets confused? Many thanks for a hint, Helmut Jarausch Lehrstuhl fuer Numerische Mathematik RWTH - Aachen University D 52056 Aachen, Germany From shadow@gmail.com Thu Nov 29 15:38:44 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 29 Nov 2007 10:38:44 -0500 Subject: [OpenAFS] openafs partition - how to increase In-Reply-To: References: Message-ID: ------=_Part_7928_27812413.1196350724202 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Nov 29, 2007 10:31 AM, Helmut Jarausch wrote: > Hi, > > I'd like to resize (enlarge) the ext2-partition on which e.g. > /vicepa is mounted. > > I know how to enlarge an ext2 partition using Linux tools, > but how does OpenAFS react to this situtation? > > Does it use the additional space or is there any danger > OpenAFS gets confused? > > Many thanks for a hint, > It doesn't care, and won't use it unless you increase the number in the cacheinfo file anyway. ------=_Part_7928_27812413.1196350724202 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On Nov 29, 2007 10:31 AM, Helmut Jarausch <jarausch@igpm.rwth-aachen.de> wrote:
Hi,

I'd like to resize (enlarge) the ext2-partition on which e.g.
/vicepa is mounted.

I know how to enlarge an ext2 partition using Linux tools,
but how does OpenAFS react to this situtation?

Does it use the additional space or is there any danger
OpenAFS gets confused?

Many thanks for a hint,
It doesn't care, and won't use it unless you increase the number in the cacheinfo file anyway.
 

------=_Part_7928_27812413.1196350724202-- From allbery@ece.cmu.edu Thu Nov 29 15:40:29 2007 From: allbery@ece.cmu.edu (Brandon S. Allbery KF8NH) Date: Thu, 29 Nov 2007 10:40:29 -0500 Subject: [OpenAFS] openafs partition - how to increase In-Reply-To: References: Message-ID: <4A4BAE4D-89C2-4DCC-97FF-7736222DD20E@ece.cmu.edu> On Nov 29, 2007, at 10:38 , Derrick Brashear wrote: > On Nov 29, 2007 10:31 AM, Helmut Jarausch aachen.de> wrote: > > I'd like to resize (enlarge) the ext2-partition on which e.g. > > /vicepa is mounted. > It doesn't care, and won't use it unless you increase the number in > the cacheinfo file anyway. Aroo? That'd be /var/vice/cache, not e.g. /vicepa? -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] allbery@kf8nh.com system administrator [openafs,heimdal,too many hats] allbery@ece.cmu.edu electrical and computer engineering, carnegie mellon university KF8NH From sdevine@msu.edu Thu Nov 29 15:42:37 2007 From: sdevine@msu.edu (Steve Devine) Date: Thu, 29 Nov 2007 10:42:37 -0500 Subject: [OpenAFS] openafs partition - how to increase In-Reply-To: References: Message-ID: <474EDDED.7030303@msu.edu> Derrick Brashear wrote: > > > On Nov 29, 2007 10:31 AM, Helmut Jarausch > > > wrote: > > Hi, > > I'd like to resize (enlarge) the ext2-partition on which e.g. > /vicepa is mounted. > > I know how to enlarge an ext2 partition using Linux tools, > but how does OpenAFS react to this situtation? > > Does it use the additional space or is there any danger > OpenAFS gets confused? > > Many thanks for a hint, > > It doesn't care, and won't use it unless you increase the number in > the cacheinfo file anyway. > > I don't think he is asking about a cache partition. -- Steve Devine Network Storage & Printing Academic Computing & Network Services Michigan State University 506 Computer Center East Lansing, MI 48824-1042 1-517-432-7327 Baseball is ninety percent mental; the other half is physical. - Yogi Berra From deengert@anl.gov Thu Nov 29 15:44:54 2007 From: deengert@anl.gov (Douglas E. Engert) Date: Thu, 29 Nov 2007 09:44:54 -0600 Subject: [OpenAFS] Automatically get AFS-token In-Reply-To: <9A6A62B6B84859469F3EBB5F09D818CA99416F@cernxchg50.cern.ch> References: <9A6A62B6B84859469F3EBB5F09D818CA99416F@cernxchg50.cern.ch> Message-ID: <474EDE76.40003@anl.gov> Lara Lloret Iglesias wrote: > Hello! >=20 > I=B4ve just installed three AFS clients, and everything seems to work f= ine! The only problem is that I would like to get my afs-token on login, = without having to type klog. > Any idea? Yes use pam_krb5 and pam_afs_session. I am assuming that your site has th= e krb5 setup and can use aklog instead of klog. Check with you afs admins. Try Google'ing with these: site:cern.ch aklog site:cern.ch pam_krb5 site:cern.ch pam_afs2 pam_afs2 is what we use, but will eventual convert to pam_afs_session. >=20 > Thank you very much, >=20 >=20 > Lara > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info >=20 >=20 --=20 Douglas E. Engert Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439 (630) 252-5444 From jaltman@secure-endpoints.com Thu Nov 29 15:59:14 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Thu, 29 Nov 2007 16:59:14 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <474CCEEC.5040103@secure-endpoints.com> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> <474C2C21.7010507@secure-endpoints.com> <474C815F.2010308@enem.nl> <474C8D0C.50500@secure-endpoints.com> <474C8FC8.5000405@enem.nl> <474CCEEC.5040103@secure-endpoints.com> Message-ID: <474EE1D2.4010407@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms030500010901000008090500 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Jeffrey Altman wrote: > If the files are truly intended for read-only use, store them in a > directory that provides only 'rl' access to the end users or store them > in a .readonly volume. In both of those cases the AFS Cache Manager > knows that the user cannot obtain a lock on the file and will issue one > locally. > > Jeffrey Altman Now that I am back in my own timezone let me take the time to explain a bit more about locking and Microsoft Office applications. Office applications will obtain an exclusive lock on a file even when the file is being opened in "read only" mode. OAFW will translate "file open for read and not write and not delete" and "request for exclusive lock" as meaning "obtain a read lock on the file". The problem that you are experiencing is that while the Office application is requesting a lock for a very small subset of the file, AFS only implements full file locks. If Office applications are two machines attempt to open the same file and the first one has the file open for write, the second one won't be able to open for read because lock requests that otherwise would be non-intersecting byte ranges collide when translated into full file locks. Therefore, if you are providing files to be used simply as read only templates, they should be stored in AFS in a manner that indicates to the AFS client that they are in fact readonly so that the cache manager knows it is safe to fake the locks locally. Jeffrey Altman --------------ms030500010901000008090500 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMjkxNTU5MTRaMCMGCSqGSIb3DQEJBDEWBBS4FKGc MegUP4VyiVC77rDIuPKdETBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAo5bsNpyYvxYVGpmREnLpmDuZSDKOLQojtDUYEikHbrDZIzdJ2CW9 c1SThdxwGRIckVwOA5Vckfu77SS/uprvhRA1ebu70z60g+40ZcOhz6IZrVc8nqShRjUieQoI eGWl1JniNV5mxVt5K2teE9B1t9UOw+IaHjSlWOh9zCuQKdfL3V4QRZCbujUQZSjPmIDDnQ0F Icc1rmzgqHIdyUSOIo4MOi9GEcB3Ynq4ZHjTv5VYe1fUM8/cCpXXvokyknW3isknLbpFImEt 5ILdeds9nyPVzFCgHIQjRxOJ2It7fSA9mwNxHRaSqpNgxq+wvVTXeq9aJKULo/PzhZL+2Od5 ZwAAAAAAAA== --------------ms030500010901000008090500-- From mcgarr@umich.edu Thu Nov 29 15:57:06 2007 From: mcgarr@umich.edu (Michael C Garrison) Date: Thu, 29 Nov 2007 10:57:06 -0500 (EST) Subject: [OpenAFS] Automatically get AFS-token In-Reply-To: <9A6A62B6B84859469F3EBB5F09D818CA99416F@cernxchg50.cern.ch> References: <9A6A62B6B84859469F3EBB5F09D818CA99416F@cernxchg50.cern.ch> Message-ID: This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. ---745681150-274484725-1196351826=:24965 Content-Type: TEXT/PLAIN; charset=iso-8859-1; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE On Thu, 29 Nov 2007, Lara Lloret Iglesias wrote: > Hello! > > I=B4ve just installed three AFS clients, and everything seems to work fin= e! The only problem is that I would like to get my afs-token on login, with= out having to type klog. > Any idea? Use a pam module, such as Russ' pam-afs-session:=20 http://www.eyrie.org/~eagle/software/pam-afs-session/ -- Mike Garrison ---745681150-274484725-1196351826=:24965-- From dirk.heinrichs@online.de Thu Nov 29 17:47:15 2007 From: dirk.heinrichs@online.de (Dirk Heinrichs) Date: Thu, 29 Nov 2007 18:47:15 +0100 Subject: [OpenAFS] openafs partition - how to increase In-Reply-To: References: Message-ID: <200711291847.22321.dirk.heinrichs@online.de> --nextPart1297564.TgferXUnXa Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Am Donnerstag 29 November 2007 schrieb Helmut Jarausch: > I'd like to resize (enlarge) the ext2-partition on which e.g. > /vicepa is mounted. > > Does it use the additional space or is there any danger > OpenAFS gets confused? It will happily use the additional space. Bye... Dirk --nextPart1297564.TgferXUnXa Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD4DBQBHTvsq8NVtnsLkZ7sRApFQAJ9YLBKRp8uHcMeiJJyi7zZsUPuK6ACXVn4U XlBRCQLBcgBbo499DqLNiw== =/sCG -----END PGP SIGNATURE----- --nextPart1297564.TgferXUnXa-- From Jerry.Normandin@dafca.com Thu Nov 29 17:52:40 2007 From: Jerry.Normandin@dafca.com (Jerry Normandin) Date: Thu, 29 Nov 2007 09:52:40 -0800 Subject: [OpenAFS] openafs upgrade from 1.4.1 to 1.5.7 In-Reply-To: <20071129.102915.181841539.haba@habanero.pdc.kth.se> Message-ID: Thanks everybody for your input! I just started my postion here and inherited this AFS deployment. It was great that the previous people chose AFS, but they deployed it wrong. I decided on using the XFS file system, Also I will be upgrading to 1.4.7 The very first day I was here I got queries about improving AFS performance. First I plan on making the fixes, next to add some redundancy. There is none here. At my previous employers I had plenty of nodes. Over there is no Redundancy in the volume servers. Sure there are multiple volume servers, but only one for AFS HOME and one for TOOLS. So.. I have my work cut out for me. =20 Once again.. Thanks everybody! -----Original Message----- From: Harald Barth [mailto:haba@kth.se]=20 Sent: Thursday, November 29, 2007 4:29 AM To: Jerry Normandin Cc: openafs-info@openafs.org Subject: Re: [OpenAFS] openafs upgrade from 1.4.1 to 1.5.7 > What's the underlying filesystem? AFS passes through the semantics of > metadata operations of the underlying filesystem, and ext* for instance is > poor at it. For example xfs on Linux has worked well for us. > > We are running an old version of AFS.. 1.4.1. Are there any > > configuration differences between 1.4.1 and 1.5.7? > > >=20 > Lots. Of course, we recommend 1.4.5, and not some random 1.5, especially not > an old one. 1.5.7 is similarly old to 1.4.1. As Derrick said, use 1.4.x for servers. > > Can I have a mixed environment of versions? > Unless you have pts supergroups enabled, yes, though there is a pending bug > regarding moving volumes between current 1.5.x and 1.4.x. Better than that. You can mix even if you have supergroups enabled, but don't create any supergroups until all your servers are upgraded. There is no need for a complete cell shutdown to perform an upgrade. There has not been such a need for years. Harald. From shadow@gmail.com Thu Nov 29 18:07:29 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 29 Nov 2007 13:07:29 -0500 Subject: [OpenAFS] openafs partition - how to increase In-Reply-To: <474EDDED.7030303@msu.edu> References: <474EDDED.7030303@msu.edu> Message-ID: ------=_Part_8490_1667803.1196359649900 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Nov 29, 2007 10:42 AM, Steve Devine wrote: > Derrick Brashear wrote: > > > > > > On Nov 29, 2007 10:31 AM, Helmut Jarausch > > > > > wrote: > > > > Hi, > > > > I'd like to resize (enlarge) the ext2-partition on which e.g. > > /vicepa is mounted. > > > > I know how to enlarge an ext2 partition using Linux tools, > > but how does OpenAFS react to this situtation? > > > > Does it use the additional space or is there any danger > > OpenAFS gets confused? > > > > Many thanks for a hint, > > > > It doesn't care, and won't use it unless you increase the number in > > the cacheinfo file anyway. > > > > > I don't think he is asking about a cache partition. > He's not, I'm just an idiot. ------=_Part_8490_1667803.1196359649900 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On Nov 29, 2007 10:42 AM, Steve Devine <sdevine@msu.edu> wrote:
Derrick Brashear wrote:
>
>
> On Nov 29, 2007 10:31 AM, Helmut Jarausch
> <jarausch@igpm.rwth-aachen.de <mailto: jarausch@igpm.rwth-aachen.de>>
> wrote:
>
>     Hi,
>
>     I'd like to resize (enlarge) the ext2-partition on which e.g.
>     /vicepa is mounted.
>
>     I know how to enlarge an ext2 partition using Linux tools,
>     but how does OpenAFS react to this situtation?
>
>     Does it use the additional space or is there any danger
>     OpenAFS gets confused?
>
>     Many thanks for a hint,
>
> It doesn't care, and won't use it unless you increase the number in
> the cacheinfo file anyway.
>
>
I don't think he is asking about a cache partition.

He's not, I'm just an idiot.
 

------=_Part_8490_1667803.1196359649900-- From shadow@gmail.com Thu Nov 29 18:08:33 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 29 Nov 2007 13:08:33 -0500 Subject: [OpenAFS] openafs upgrade from 1.4.1 to 1.5.7 In-Reply-To: References: <20071129.102915.181841539.haba@habanero.pdc.kth.se> Message-ID: ------=_Part_8493_19588521.1196359713207 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Nov 29, 2007 12:52 PM, Jerry Normandin wrote: > Thanks everybody for your input! I just started my postion here and > inherited this AFS deployment. It was great that the previous people > chose AFS, but they deployed it wrong. I decided on using the XFS file > system, Also I will be upgrading to 1.4.7 > We released 1.4.5 recently. What do you know that I don't? ------=_Part_8493_19588521.1196359713207 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On Nov 29, 2007 12:52 PM, Jerry Normandin <Jerry.Normandin@dafca.com> wrote:
Thanks everybody for your input! I just started my postion here and
inherited this AFS deployment.  It was great that the previous people
chose AFS, but they deployed it wrong.  I decided on using the XFS file
system, Also I will be upgrading to 1.4.7

We released 1.4.5 recently.

What do you know that I don't?
 

------=_Part_8493_19588521.1196359713207-- From matt.smith@uconn.edu Thu Nov 29 18:49:15 2007 From: matt.smith@uconn.edu (Smith, Matt) Date: Thu, 29 Nov 2007 13:49:15 -0500 Subject: [OpenAFS] File systems on Linux, again. Message-ID: <1196362155.6464.7.camel@d80h167.uconn.edu> --=-C15JKpvLgYm9AA9X+0d1 Content-Type: text/plain Content-Transfer-Encoding: quoted-printable After the recent thread "openafs upgrade from 1.4.1 to 1.5.7", and a review of a thread[1] from July, I'm wondering if there is a definitive recommendation for which file system to use on Linux AFS file servers. Ext3, XFS, JFS, something else? Thanks all, -Matt [1] http://www.openafs.org/pipermail/openafs-info/2007-July/026798.html --=20 Matt Smith matt.smith@uconn.edu University Information Technology Services (UITS) University of Connecticut PGP Key ID: 0xE9C5244E --=-C15JKpvLgYm9AA9X+0d1 Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQBHTwmrGP63pOnFJE4RAto0AJ41351OQKW0Tbk+itlI0MGqHFJRvwCfTo/R /5G/Wp6WlXYaj5rlsTsWPQM= =2j62 -----END PGP SIGNATURE----- --=-C15JKpvLgYm9AA9X+0d1-- From rra@stanford.edu Thu Nov 29 19:33:26 2007 From: rra@stanford.edu (Russ Allbery) Date: Thu, 29 Nov 2007 11:33:26 -0800 Subject: [OpenAFS] What's the problem with reiser In-Reply-To: <474EC2D1.1060407@rampaginggeek.com> (Jason Edgecombe's message of "Thu\, 29 Nov 2007 08\:46\:57 -0500") References: <474E8F61.6030509@mpi-magdeburg.mpg.de> <20071129.132450.222201899.haba@habanero.pdc.kth.se> <200711291341.21115.dirk.heinrichs.ext@nsn.com> <474EC2D1.1060407@rampaginggeek.com> Message-ID: <878x4ghmnt.fsf@windlord.stanford.edu> Jason Edgecombe writes: > Fascinating. I did not know of UFS logging issue on the cache partition. > Strangely, I haven't heard of any issues. does ext3 have this issue as > well? No, ext3 is fine. UFS logging is also fine provided that nothing else is writing to the same partition. afsd prints out a warning about this when using UFS with logging on Solaris. -- Russ Allbery (rra@stanford.edu) From rra@stanford.edu Thu Nov 29 19:35:50 2007 From: rra@stanford.edu (Russ Allbery) Date: Thu, 29 Nov 2007 11:35:50 -0800 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: <1196362155.6464.7.camel@d80h167.uconn.edu> (Matt Smith's message of "Thu\, 29 Nov 2007 13\:49\:15 -0500") References: <1196362155.6464.7.camel@d80h167.uconn.edu> Message-ID: <874pf4hmjt.fsf@windlord.stanford.edu> "Smith, Matt" writes: > After the recent thread "openafs upgrade from 1.4.1 to 1.5.7", and a > review of a thread[1] from July, I'm wondering if there is a definitive > recommendation for which file system to use on Linux AFS file servers. > Ext3, XFS, JFS, something else? It shouldn't make much of a difference, so I think you're safe choosing your file system on whatever basis you'd choose a file system for any other file server. We use ext3 because of the stability, reliability, and "center of the mainstream" support in the kernel, which we always considered more important than a bit of additional speed, but your mileage may vary. XFS is probably the next most common choice. I would be very leery of ReiserFS. It has nice features, but the recovery tools are fairly horrific. -- Russ Allbery (rra@stanford.edu) From jason@rampaginggeek.com Thu Nov 29 20:26:03 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Thu, 29 Nov 2007 15:26:03 -0500 Subject: [OpenAFS] Slow AFS performance In-Reply-To: References: <474D78F2.40808@rampaginggeek.com> Message-ID: <474F205B.4060301@rampaginggeek.com> Dale Ghent wrote: > On Nov 28, 2007, at 9:19 AM, Jason Edgecombe wrote: > >> I'm experiencing poor AFS performance on under Sparc solaris 9 09/05HW >> running Openafs server 1.4.1 on a Sun StorageTeck 3511 Fibre channel to >> SATA array >> >> At first, I thought that having UFS logging disabled was the culprit, >> but I have enabled UFS logging and I am using the namei server, but >> performance still stinks. It took 1.5 hours to move a 3.2GB volume to >> the server. Things seem fine except on the Fibre channel disks. > > ... > >> My bonnie++ performance numbers for vicepa are here: >> http://www.coe.uncc.edu/~jwedgeco/bonnie.html >> >> What could AFS be doing that causes the performance to stink? > > Well first, your bonnie run tested sequential IO, so of course you're > going to get stellar numbers with sequential large writes and reads. > Arrays and disks in general eat that stuff right on up. > > AFS namei does writes in small batches, up to 64k, and all of it > random (think: lots of users accessing lots of small files) It would > be advantageous to tun to this environment. > > Check out your 3511s first. > > 1) Make sure write-back caching is on. This is a global as well as a > per-Logical Drive setting on the 3510/3511. > > 2) Optimized for Random I/O (be sure the latest 3511 fw is applied, it > betters the performance of this setting) > > 3) You didn't mention the type of RAID config you have on the 3511s. > RAID5? If so, is the stripe size low (64k?) > > Some 'iostat -nx 1' output would be helpful, too. The wsvc_t and %b > columns would be telling. > > /dale > ok, I'm now some better performance out of my FC array. "vos move" of a 4.4GB volume from one disk in the FC array to another disk in the array only took 16 minutes (4.6MB/s). Derrick's suggestion of upgrading to 1.4.5 with namei did the trick. Unfortunately, I don't think I can upgrade the firmware or tune the 3511 array. It's the 3511 expansion unit without the raid controller. It's a JBOD. As requested, here are the statics from "iostat -nx 1" r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device 0.0 13.0 0.0 8.0 0.0 0.1 0.0 9.8 0 2 c1t0d0 481.0 0.0 14905.1 0.0 0.0 0.6 0.0 1.3 0 50 c0t0d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t1d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t3d0 13.0 133.0 104.0 12082.3 0.0 3.6 0.0 25.0 0 86 c0t2d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t5d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t8d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t7d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t4d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t6d0 ... extended device statistics r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c1t0d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t0d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t1d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t3d0 0.0 333.0 0.0 581.0 384.2 19.0 1153.7 57.1 100 100 c0t2d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t5d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t8d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t7d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t4d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t6d0 extended device statistics r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c1t0d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t0d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t1d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t3d0 0.0 339.0 0.0 773.0 371.3 19.0 1095.4 56.0 100 100 c0t2d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t5d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t8d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t7d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t4d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c0t6d0 Jason From jason@rampaginggeek.com Thu Nov 29 20:34:24 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Thu, 29 Nov 2007 15:34:24 -0500 Subject: [OpenAFS] 1.4.5 namei on solaris 9 sparc requires AlwaysAttach for vice partitions Message-ID: <474F2250.2010403@rampaginggeek.com> Hi all, In my sordid saga to get a Sun fibre channel array working well with AFS, I found the following: When I upgraded the server to 1.4.5 namei, the fileserver would not mount the /vicep? partitions without doing a "touch /vicep?/AlwaysAttach" first. These are dedicated partitions on separate hard drives. I'm using a source-compiled openafs on solaris 9 sparc. openafs was compiled with the following options: CC=/opt/SUNWspro/bin/cc YACC="yacc -vd" ./configure \ --enable-transarc-paths \ --enable-largefile-fileserver \ --enable-supergroups \ --enable-namei-fileserver \ --with-krb5-conf=/usr/local/krb5/bin/krb5-config We're using MIT kerberos 1.4.1 on the clients & fileservers with a 1.6.x KDC # mount | grep vicep /vicepa on /dev/dsk/c0t0d0s6 read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80006 on Thu Nov 29 13:03:15 2007 /vicepd on /dev/dsk/c0t3d0s6 read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80016 on Thu Nov 29 13:03:15 2007 /vicepc on /dev/dsk/c0t2d0s6 read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d8001e on Thu Nov 29 13:03:15 2007 /vicepb on /dev/dsk/c0t1d0s6 read/write/setuid/intr/largefiles/xattr/onerror=panic/dev=1d8000e on Thu Nov 29 13:03:15 2007 # grep vicep /etc/vfstab /dev/dsk/c0t0d0s6 /dev/rdsk/c0t0d0s6 /vicepa ufs 3 yes - /dev/dsk/c0t1d0s6 /dev/rdsk/c0t1d0s6 /vicepb ufs 3 yes - /dev/dsk/c0t2d0s6 /dev/rdsk/c0t2d0s6 /vicepc ufs 3 yes - #cat SalvageLog @(#) OpenAFS 1.4.5 built 2007-11-28 11/29/2007 09:52:59 STARTING AFS SALVAGER 2.4 (/usr/afs/bin/salvager) 11/29/2007 09:52:59 No file system partitions named /vicep* found; not salvaged Does anyone know why this would be happening? Thanks, Jason From rmdyer@uncc.edu Thu Nov 29 20:37:51 2007 From: rmdyer@uncc.edu (Rodney M. Dyer) Date: Thu, 29 Nov 2007 15:37:51 -0500 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <474EE1D2.4010407@secure-endpoints.com> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> <474C2C21.7010507@secure-endpoints.com> <474C815F.2010308@enem.nl> <474C8D0C.50500@secure-endpoints.com> <474C8FC8.5000405@enem.nl> <474CCEEC.5040103@secure-endpoints.com> <474EE1D2.4010407@secure-endpoints.com> Message-ID: <6.1.2.0.1.20071129153247.02effec0@ironhost.uncc.edu.> At 10:59 AM 11/29/2007, Jeffrey Altman wrote: >Therefore, if you are providing files to be used simply as read only >templates, they should be stored in AFS in a manner that indicates to >the AFS client that they are in fact readonly so that the cache manager >knows it is safe to fake the locks locally. One small question. Historically, in virtually all DOS/Win PC networking environments, the file attribute "r" was also recognized by applications as meaning "read-only" (whole file), even if it is just "advisory" to the network client. What does AFS do in this situation, if anything at all, or is that still the applications responsibility to recognize the "r" attribute? Rodney From rob@nofocus.org Thu Nov 29 20:39:26 2007 From: rob@nofocus.org (Rob Banz) Date: Thu, 29 Nov 2007 15:39:26 -0500 Subject: [OpenAFS] Slow AFS performance In-Reply-To: <474F205B.4060301@rampaginggeek.com> References: <474D78F2.40808@rampaginggeek.com> <474F205B.4060301@rampaginggeek.com> Message-ID: >> > ok, I'm now some better performance out of my FC array. "vos move" > of a > 4.4GB volume from one disk in the FC array to another disk in the > array > only took 16 minutes (4.6MB/s). > > Derrick's suggestion of upgrading to 1.4.5 with namei did the trick. > > Unfortunately, I don't think I can upgrade the firmware or tune the > 3511 > array. It's the 3511 expansion unit without the raid controller. > It's a > JBOD. I think 1.4.5 and above has the no-fsync stuff enabled by default, which really speeds up operations that do a lot of file creations/ deletions such as volume moves. (if it doesn't, head over to the OpenAFS RT issue tracker and find the patch) -rob From shadow@gmail.com Thu Nov 29 20:45:51 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 29 Nov 2007 15:45:51 -0500 Subject: [OpenAFS] 1.4.5 namei on solaris 9 sparc requires AlwaysAttach for vice partitions In-Reply-To: <474F2250.2010403@rampaginggeek.com> References: <474F2250.2010403@rampaginggeek.com> Message-ID: ------=_Part_9131_19457510.1196369151513 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline On Nov 29, 2007 3:34 PM, Jason Edgecombe wrote: > Hi all, > > In my sordid saga to get a Sun fibre channel array working well with > AFS, I found the following: > > When I upgraded the server to 1.4.5 namei, the fileserver would not > mount the /vicep? partitions without doing a "touch > /vicep?/AlwaysAttach" first. These are dedicated partitions on separate > hard drives. > > I'm using a source-compiled openafs on solaris 9 sparc. openafs was > compiled with the following options: > CC=/opt/SUNWspro/bin/cc YACC="yacc -vd" ./configure \ > --enable-transarc-paths \ > --enable-largefile-fileserver \ > --enable-supergroups \ > --enable-namei-fileserver \ > --with-krb5-conf=/usr/local/krb5/bin/krb5-config > > We're using MIT kerberos 1.4.1 on the clients & fileservers with a 1.6.xKDC > > # mount | grep vicep > /vicepa on /dev/dsk/c0t0d0s6 > read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80006 > on Thu Nov 29 13:03:15 2007 > /vicepd on /dev/dsk/c0t3d0s6 > read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80016 > on Thu Nov 29 13:03:15 2007 > /vicepc on /dev/dsk/c0t2d0s6 > read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d8001e > on Thu Nov 29 13:03:15 2007 > /vicepb on /dev/dsk/c0t1d0s6 > read/write/setuid/intr/largefiles/xattr/onerror=panic/dev=1d8000e on Thu > Nov 29 13:03:15 2007 > > # grep vicep /etc/vfstab > /dev/dsk/c0t0d0s6 /dev/rdsk/c0t0d0s6 /vicepa ufs 3 > yes - > /dev/dsk/c0t1d0s6 /dev/rdsk/c0t1d0s6 /vicepb ufs 3 > yes - > /dev/dsk/c0t2d0s6 /dev/rdsk/c0t2d0s6 /vicepc ufs 3 > yes - > > #cat SalvageLog > @(#) OpenAFS 1.4.5 built 2007-11-28 > 11/29/2007 09:52:59 STARTING AFS SALVAGER 2.4 (/usr/afs/bin/salvager) > 11/29/2007 09:52:59 No file system partitions named /vicep* found; not > salvaged > > Does anyone know why this would be happening? > Probably a bug in the "what's acceptable as a vice partition" logic... which I thought i fixed before 1.4.5; i bet i committed the wrong thing (because i tested it) ------=_Part_9131_19457510.1196369151513 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

On Nov 29, 2007 3:34 PM, Jason Edgecombe <jason@rampaginggeek.com> wrote:
Hi all,

In my sordid saga to get a Sun fibre channel array working well with
AFS, I found the following:

When I upgraded the server to 1.4.5 namei, the fileserver would not
mount the /vicep? partitions without doing a "touch
/vicep?/AlwaysAttach" first. These are dedicated partitions on separate
hard drives.

I'm using a source-compiled openafs on solaris 9 sparc. openafs was
compiled with the following options:
CC=/opt/SUNWspro/bin/cc YACC="yacc -vd" ./configure \
 --enable-transarc-paths \
 --enable-largefile-fileserver \
 --enable-supergroups \
 --enable-namei-fileserver \
 --with-krb5-conf=/usr/local/krb5/bin/krb5-config

We're using MIT kerberos 1.4.1 on the clients & fileservers with a 1.6.x KDC

# mount | grep vicep
/vicepa on /dev/dsk/c0t0d0s6
read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80006
on Thu Nov 29 13:03:15 2007
/vicepd on /dev/dsk/c0t3d0s6
read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80016
on Thu Nov 29 13:03:15 2007
/vicepc on /dev/dsk/c0t2d0s6
read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d8001e
on Thu Nov 29 13:03:15 2007
/vicepb on /dev/dsk/c0t1d0s6
read/write/setuid/intr/largefiles/xattr/onerror=panic/dev=1d8000e on Thu
Nov 29 13:03:15 2007

# grep vicep /etc/vfstab
/dev/dsk/c0t0d0s6       /dev/rdsk/c0t0d0s6      /vicepa ufs     3
yes     -
/dev/dsk/c0t1d0s6       /dev/rdsk/c0t1d0s6      /vicepb ufs     3
yes     -
/dev/dsk/c0t2d0s6       /dev/rdsk/c0t2d0s6      /vicepc ufs     3
yes     -

#cat SalvageLog
@(#) OpenAFS 1.4.5 built  2007-11-28
11/29/2007 09:52:59 STARTING AFS SALVAGER 2.4 (/usr/afs/bin/salvager)
11/29/2007 09:52:59 No file system partitions named /vicep* found; not
salvaged

Does anyone know why this would be happening?
Probably a bug in the "what's acceptable as a vice partition" logic... which I thought i fixed before 1.4.5; i bet i committed the wrong thing (because i tested it)
 

------=_Part_9131_19457510.1196369151513-- From jason@rampaginggeek.com Thu Nov 29 21:04:41 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Thu, 29 Nov 2007 16:04:41 -0500 Subject: [OpenAFS] 1.4.5 namei on solaris 9 sparc requires AlwaysAttach for vice partitions In-Reply-To: References: <474F2250.2010403@rampaginggeek.com> Message-ID: <474F2969.7010102@rampaginggeek.com> Derrick Brashear wrote: > > > On Nov 29, 2007 3:34 PM, Jason Edgecombe > wrote: > > Hi all, > > In my sordid saga to get a Sun fibre channel array working well with > AFS, I found the following: > > When I upgraded the server to 1.4.5 namei, the fileserver would not > mount the /vicep? partitions without doing a "touch > /vicep?/AlwaysAttach" first. These are dedicated partitions on > separate > hard drives. > > I'm using a source-compiled openafs on solaris 9 sparc. openafs was > compiled with the following options: > CC=/opt/SUNWspro/bin/cc YACC="yacc -vd" ./configure \ > --enable-transarc-paths \ > --enable-largefile-fileserver \ > --enable-supergroups \ > --enable-namei-fileserver \ > --with-krb5-conf=/usr/local/krb5/bin/krb5-config > > We're using MIT kerberos 1.4.1 on the clients & fileservers with a > 1.6.x KDC > > # mount | grep vicep > /vicepa on /dev/dsk/c0t0d0s6 > read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80006 > on Thu Nov 29 13:03:15 2007 > /vicepd on /dev/dsk/c0t3d0s6 > read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80016 > on Thu Nov 29 13:03:15 2007 > /vicepc on /dev/dsk/c0t2d0s6 > read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d8001e > > on Thu Nov 29 13:03:15 2007 > /vicepb on /dev/dsk/c0t1d0s6 > read/write/setuid/intr/largefiles/xattr/onerror=panic/dev=1d8000e > on Thu > Nov 29 13:03:15 2007 > > # grep vicep /etc/vfstab > /dev/dsk/c0t0d0s6 /dev/rdsk/c0t0d0s6 /vicepa ufs 3 > yes - > /dev/dsk/c0t1d0s6 /dev/rdsk/c0t1d0s6 /vicepb ufs 3 > yes - > /dev/dsk/c0t2d0s6 /dev/rdsk/c0t2d0s6 /vicepc ufs 3 > yes - > > #cat SalvageLog > @(#) OpenAFS 1.4.5 built 2007-11-28 > 11/29/2007 09:52:59 STARTING AFS SALVAGER 2.4 (/usr/afs/bin/salvager) > 11/29/2007 09:52:59 No file system partitions named /vicep* found; not > salvaged > > Does anyone know why this would be happening? > > Probably a bug in the "what's acceptable as a vice partition" logic... > which I thought i fixed before 1.4.5; i bet i committed the wrong > thing (because i tested it) > > Is it safe to run like this? Should I file a bug? Jason From shadow@gmail.com Thu Nov 29 21:11:58 2007 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 29 Nov 2007 16:11:58 -0500 Subject: [OpenAFS] 1.4.5 namei on solaris 9 sparc requires AlwaysAttach for vice partitions In-Reply-To: <474F2969.7010102@rampaginggeek.com> References: <474F2250.2010403@rampaginggeek.com> <474F2969.7010102@rampaginggeek.com> Message-ID: ------=_Part_9261_1669255.1196370718908 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline > > > Does anyone know why this would be happening? > > > > Probably a bug in the "what's acceptable as a vice partition" logic... > > which I thought i fixed before 1.4.5; i bet i committed the wrong > > thing (because i tested it) > > > > > Is it safe to run like this? > yup > > Should I file a bug? > probably ------=_Part_9261_1669255.1196370718908 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

>     Does anyone know why this would be happening?
>
> Probably a bug in the "what's acceptable as a vice partition" logic...
> which I thought i fixed before 1.4.5; i bet i committed the wrong
> thing (because i tested it)
>
>
Is it safe to run like this?

yup
 

Should I file a bug?

probably

------=_Part_9261_1669255.1196370718908-- From reuter@rzg.mpg.de Thu Nov 29 21:17:22 2007 From: reuter@rzg.mpg.de (Hartmut Reuter) Date: Thu, 29 Nov 2007 22:17:22 +0100 Subject: [OpenAFS] 1.4.5 namei on solaris 9 sparc requires AlwaysAttach for vice partitions In-Reply-To: <474F2250.2010403@rampaginggeek.com> References: <474F2250.2010403@rampaginggeek.com> Message-ID: <474F2C62.3000104@rzg.mpg.de> I also was today surprised when I started the freshly compiled 1.4.5 fileserver on Solaris and it didn't attach any partition. There was a change between 1.4.4 and 1.4.5 in favour of zfs, but a unfortunately broken: /* Ignore non ufs or non read/write partitions */ if ((strcmp(mnt.mnt_fstype, "ufs") != 0) || (strncmp(mnt.mnt_mntopts, "ro,ignore", 9) == 0)) #else (strcmp(mnt.mnt_fstype, "ufs") != 0) #endif || (strncmp(mnt.mnt_mntopts, "ro,ignore", 9) == 0)) continue; was changed to /* Ignore non ufs or non read/write partitions */ /* but allow zfs too if we're in the NAMEI environment */ if ( #ifdef AFS_NAMEI_ENV (((!strcmp(mnt.mnt_fstype, "ufs") && strcmp(mnt.mnt_fstype, "zfs")))) || (strncmp(mnt.mnt_mntopts, "ro,ignore", 9) == 0)) continue; } #else continue; #endif The "!" in front of strcmp in the new version lets him exactly ignore ufs. Just remove it! Hartmut Jason Edgecombe wrote: > Hi all, > > In my sordid saga to get a Sun fibre channel array working well with > AFS, I found the following: > > When I upgraded the server to 1.4.5 namei, the fileserver would not > mount the /vicep? partitions without doing a "touch > /vicep?/AlwaysAttach" first. These are dedicated partitions on separate > hard drives. > > I'm using a source-compiled openafs on solaris 9 sparc. openafs was > compiled with the following options: > CC=/opt/SUNWspro/bin/cc YACC="yacc -vd" ./configure \ > --enable-transarc-paths \ > --enable-largefile-fileserver \ > --enable-supergroups \ > --enable-namei-fileserver \ > --with-krb5-conf=/usr/local/krb5/bin/krb5-config > > We're using MIT kerberos 1.4.1 on the clients & fileservers with a 1.6.x KDC > > # mount | grep vicep > /vicepa on /dev/dsk/c0t0d0s6 > read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80006 > on Thu Nov 29 13:03:15 2007 > /vicepd on /dev/dsk/c0t3d0s6 > read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d80016 > on Thu Nov 29 13:03:15 2007 > /vicepc on /dev/dsk/c0t2d0s6 > read/write/setuid/intr/largefiles/logging/xattr/onerror=panic/dev=1d8001e > on Thu Nov 29 13:03:15 2007 > /vicepb on /dev/dsk/c0t1d0s6 > read/write/setuid/intr/largefiles/xattr/onerror=panic/dev=1d8000e on Thu > Nov 29 13:03:15 2007 > > # grep vicep /etc/vfstab > /dev/dsk/c0t0d0s6 /dev/rdsk/c0t0d0s6 /vicepa ufs 3 > yes - > /dev/dsk/c0t1d0s6 /dev/rdsk/c0t1d0s6 /vicepb ufs 3 > yes - > /dev/dsk/c0t2d0s6 /dev/rdsk/c0t2d0s6 /vicepc ufs 3 > yes - > > #cat SalvageLog > @(#) OpenAFS 1.4.5 built 2007-11-28 > 11/29/2007 09:52:59 STARTING AFS SALVAGER 2.4 (/usr/afs/bin/salvager) > 11/29/2007 09:52:59 No file system partitions named /vicep* found; not > salvaged > > Does anyone know why this would be happening? > > Thanks, > Jason > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info -- ----------------------------------------------------------------- Hartmut Reuter e-mail reuter@rzg.mpg.de phone +49-89-3299-1328 RZG (Rechenzentrum Garching) fax +49-89-3299-1301 Computing Center of the Max-Planck-Gesellschaft (MPG) and the Institut fuer Plasmaphysik (IPP) ----------------------------------------------------------------- From reuter@rzg.mpg.de Thu Nov 29 21:18:30 2007 From: reuter@rzg.mpg.de (Hartmut Reuter) Date: Thu, 29 Nov 2007 22:18:30 +0100 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: <1196362155.6464.7.camel@d80h167.uconn.edu> References: <1196362155.6464.7.camel@d80h167.uconn.edu> Message-ID: <474F2CA6.3050100@rzg.mpg.de> Smith, Matt wrote: > After the recent thread "openafs upgrade from 1.4.1 to 1.5.7", and a > review of a thread[1] from July, I'm wondering if there is a definitive > recommendation for which file system to use on Linux AFS file servers. > Ext3, XFS, JFS, something else? > > Thanks all, > -Matt > > [1] http://www.openafs.org/pipermail/openafs-info/2007-July/026798.html We are using exclusively xfs since many years. It is performant and you can enlarge partitions on the fly doing "lvextend" and "xfs_growfs". Hartmut ----------------------------------------------------------------- Hartmut Reuter e-mail reuter@rzg.mpg.de phone +49-89-3299-1328 RZG (Rechenzentrum Garching) fax +49-89-3299-1301 Computing Center of the Max-Planck-Gesellschaft (MPG) and the Institut fuer Plasmaphysik (IPP) ----------------------------------------------------------------- From jaltman@secure-endpoints.com Thu Nov 29 21:30:02 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Thu, 29 Nov 2007 22:30:02 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <6.1.2.0.1.20071129153247.02effec0@ironhost.uncc.edu.> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> <474C2C21.7010507@secure-endpoints.com> <474C815F.2010308@enem.nl> <474C8D0C.50500@secure-endpoints.com> <474C8FC8.5000405@enem.nl> <474CCEEC.5040103@secure-endpoints.com> <474EE1D2.4010407@secure-endpoints.com> <6.1.2.0.1.20071129153247.02effec0@ironhost.uncc.edu.> Message-ID: <474F2F5A.60403@secure-endpoints.com> Rodney M. Dyer wrote: > At 10:59 AM 11/29/2007, Jeffrey Altman wrote: >> Therefore, if you are providing files to be used simply as read only >> templates, they should be stored in AFS in a manner that indicates to >> the AFS client that they are in fact readonly so that the cache manager >> knows it is safe to fake the locks locally. > > One small question. Historically, in virtually all DOS/Win PC > networking environments, the file attribute "r" was also recognized by > applications as meaning "read-only" (whole file), even if it is just > "advisory" to the network client. What does AFS do in this situation, > if anything at all, or is that still the applications responsibility to > recognize the "r" attribute? > > Rodney The "Windows/DOS" Read-only attribute is interpreted by the application and is separate from the AFS "r" permission. When set Office applications open documents in "shared read only" mode which means that they still obtain locks on the file. Jeffrey Altman From hans@enem.nl Fri Nov 30 01:18:05 2007 From: hans@enem.nl (Hans Melgers) Date: Fri, 30 Nov 2007 02:18:05 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <474EE1D2.4010407@secure-endpoints.com> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> <474C2C21.7010507@secure-endpoints.com> <474C815F.2010308@enem.nl> <474C8D0C.50500@secure-endpoints.com> <474C8FC8.5000405@enem.nl> <474CCEEC.5040103@secure-endpoints.com> <474EE1D2.4010407@secure-endpoints.com> Message-ID: <474F64CD.4080904@enem.nl> This is a multi-part message in MIME format. --------------070805080608030703090400 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Thanks Jeffrey, Therefore, if you are providing files to be used simply as read only templates, they should be stored in AFS in a manner that indicates to the AFS client that they are in fact readonly so that the cache manager knows it is safe to fake the locks locally. Would this also be achieved when we vos release this volume to a ro clone ? (not necessarily on another fileserver?) I wonder what common practice is for this kind of shared volumes; This one is a "projects" folder where everybody stores his or her additions to several projects, so, alot of files are opened as templates to make new ones that are also stored in the same folder or subfolders. If a readonly copy could solve this locking problems (or at least will bring the number of occurences down) we'll have to do alot of vos release commands throughout the day to avoid the locking problems as much as possible. How do you handle this ? Hans Jeffrey Altman wrote: > Jeffrey Altman wrote: > >> If the files are truly intended for read-only use, store them in a >> directory that provides only 'rl' access to the end users or store them >> in a .readonly volume. In both of those cases the AFS Cache Manager >> knows that the user cannot obtain a lock on the file and will issue one >> locally. >> >> Jeffrey Altman >> > > Now that I am back in my own timezone let me take the time to explain a > bit more about locking and Microsoft Office applications. Office > applications will obtain an exclusive lock on a file even when the file > is being opened in "read only" mode. OAFW will translate "file open for > read and not write and not delete" and "request for exclusive lock" as > meaning "obtain a read lock on the file". > > The problem that you are experiencing is that while the Office > application is requesting a lock for a very small subset of the file, > AFS only implements full file locks. If Office applications are two > machines attempt to open the same file and the first one has the file > open for write, the second one won't be able to open for read because > lock requests that otherwise would be non-intersecting byte ranges > collide when translated into full file locks. > > Therefore, if you are providing files to be used simply as read only > templates, they should be stored in AFS in a manner that indicates to > the AFS client that they are in fact readonly so that the cache manager > knows it is safe to fake the locks locally. > > Jeffrey Altman > --------------070805080608030703090400 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Thanks Jeffrey,

<snip>
Therefore, if you are providing files to be used simply as read only
templates, they should be stored in AFS in a manner that indicates to
the AFS client that they are in fact readonly so that the cache manager
knows it is safe to fake the locks locally.
Would this also be achieved when we vos release this volume to a ro clone ? (not necessarily on another fileserver?)

I wonder what common practice is for this kind of shared volumes;  This one is a "projects" folder where everybody stores his or her additions to several projects, so, alot of files are opened as templates to make new ones that are also stored in the same folder or subfolders.
If a readonly copy could solve this locking problems (or at least will bring the number of occurences down) we'll have to do alot of vos release commands throughout the day to avoid the locking problems as much as possible. How do you handle this ?

Hans



Jeffrey Altman wrote:
Jeffrey Altman wrote:
  
If the files are truly intended for read-only use, store them in a
directory that provides only 'rl' access to the end users or store them
in a .readonly volume.   In both of those cases the AFS Cache Manager
knows that the user cannot obtain a lock on the file and will issue one
locally.

Jeffrey Altman
    

Now that I am back in my own timezone let me take the time to explain a
bit more about locking and Microsoft Office applications.  Office
applications will obtain an exclusive lock on a file even when the file
is being opened in "read only" mode.  OAFW will translate "file open for
read and not write and not delete" and "request for exclusive lock" as
meaning "obtain a read lock on the file".

The problem that you are experiencing is that while the Office
application is requesting a lock for a very small subset of the file,
AFS only implements full file locks.  If Office applications are two
machines attempt to open the same file and the first one has the file
open for write, the second one won't be able to open for read because
lock requests that otherwise would be non-intersecting byte ranges
collide when translated into full file locks.

Therefore, if you are providing files to be used simply as read only
templates, they should be stored in AFS in a manner that indicates to
the AFS client that they are in fact readonly so that the cache manager
knows it is safe to fake the locks locally.

Jeffrey Altman
  
--------------070805080608030703090400-- From jaltman@secure-endpoints.com Fri Nov 30 06:42:48 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Fri, 30 Nov 2007 07:42:48 +0100 Subject: [OpenAFS] MS Word crashing when opening files, 1.5.27 client In-Reply-To: <474F64CD.4080904@enem.nl> References: <964b3af230c97f4d8434525a3b6fa7e8@enem.nl> <474AE7CB.3040504@secure-endpoints.com> <474C0D5B.2060802@enem.nl> <474C2C21.7010507@secure-endpoints.com> <474C815F.2010308@enem.nl> <474C8D0C.50500@secure-endpoints.com> <474C8FC8.5000405@enem.nl> <474CCEEC.5040103@secure-endpoints.com> <474EE1D2.4010407@secure-endpoints.com> <474F64CD.4080904@enem.nl> Message-ID: <474FB0E8.1050004@secure-endpoints.com> Hans Melgers wrote: > Thanks Jeffrey, > > > > Therefore, if you are providing files to be used simply as read only > templates, they should be stored in AFS in a manner that indicates to > the AFS client that they are in fact readonly so that the cache manager > knows it is safe to fake the locks locally. > > Would this also be achieved when we vos release this volume to a ro > clone ? (not necessarily on another fileserver?) yes. that is a .readonly volume. > I wonder what common practice is for this kind of shared volumes; This > one is a "projects" folder where everybody stores his or her additions > to several projects, so, alot of files are opened as templates to make > new ones that are also stored in the same folder or subfolders. > If a readonly copy could solve this locking problems (or at least will > bring the number of occurences down) we'll have to do alot of vos > release commands throughout the day to avoid the locking problems as > much as possible. How do you handle this ? I can very easily imagine a "publish as template" button that gets implemented in Office applications as a button on a custom toolbar. The button uses the macro language to save the file locally. Copy the file to a drop box location in AFS. Then post a message to a web form with the name of the file that instructs a privileged service to move the copy into a templates directory that normal users only have 'rl' on. There is nothing wrong with users saving files in the same directory. Its just that only one of them at a time can open them. This is just one of the limitations that the gatekeepers would like to address but for which we have no resources. Contributions are welcome. Jeffrey Altman From Xavier.Canehan@in2p3.fr Fri Nov 30 10:40:33 2007 From: Xavier.Canehan@in2p3.fr (Xavier Canehan) Date: Fri, 30 Nov 2007 11:40:33 +0100 Subject: [OpenAFS] Migration of DB servers Message-ID: <20071130104032.GD6085@in2p3.fr> We'll have a complete site shutdown next week. I'm planning to use this opportunity to physically replace 2 afsdb servers. Same name and IP but host and OpenAFS version upgrades. Considering that clients and fileservers will be down, I guess that I'll just have to pass through the steps of managing the change on the afsdb servers. Am I right or is there some tracking of DB hosts by the fileservers that I should care of ? Thanks, Xavier -- Xavier Canehan | Centre de Calcul de l'IN2P3 Equipe Administration Systeme | Institut National de Physique Tel. +33 4.78.93.08.80 | Nucleaire et de Physique des http://cc.in2p3.fr | Particules. From reuter@rzg.mpg.de Fri Nov 30 10:51:54 2007 From: reuter@rzg.mpg.de (Hartmut Reuter) Date: Fri, 30 Nov 2007 11:51:54 +0100 Subject: [OpenAFS] Migration of DB servers In-Reply-To: <20071130104032.GD6085@in2p3.fr> References: <20071130104032.GD6085@in2p3.fr> Message-ID: <474FEB4A.3020509@rzg.mpg.de> Xavier Canehan wrote: > We'll have a complete site shutdown next week. I'm planning to use this > opportunity to physically replace 2 afsdb servers. Same name and IP but host > and OpenAFS version upgrades. > > Considering that clients and fileservers will be down, I guess that I'll > just have to pass through the steps of managing the change on the afsdb > servers. > > Am I right or is there some tracking of DB hosts by the fileservers that I > should care of ? Should be ok. The fileservers don't have any tracking of DB servers. As logn as you don't change the IP-addresses nobody should see any difference. Hartmut > > > Thanks, > > Xavier -- ----------------------------------------------------------------- Hartmut Reuter e-mail reuter@rzg.mpg.de phone +49-89-3299-1328 RZG (Rechenzentrum Garching) fax +49-89-3299-1301 Computing Center of the Max-Planck-Gesellschaft (MPG) and the Institut fuer Plasmaphysik (IPP) ----------------------------------------------------------------- From HraskyT@seznam.cz Fri Nov 30 12:14:33 2007 From: HraskyT@seznam.cz (=?iso-8859-2?Q?Tom=E1=B9=20Hr=E1sk=FD?=) Date: Fri, 30 Nov 2007 13:14:33 +0100 (CET) Subject: [OpenAFS] =?us-ascii?Q?Re=3A=20=5BOpenAFS=5D=20Automatically=20get=20AFS=2Dtoken?= In-Reply-To: Message-ID: <2685.3860-23944-1432108018-1196424873@seznam.cz> Or maybe try to put aklog (or your klog) to /etc/profile -- Tomas Hrasky > ------------ P=F9vodn=ED zpr=E1va ------------ > Od: Michael C Garrison > P=F8edm=ECt: Re: [OpenAFS] Automatically get AFS-token > Datum: 30.11.2007 03:32:01 > ---------------------------------------- > On Thu, 29 Nov 2007, Lara Lloret Iglesias wrote: > > > Hello! > > > > I=B4ve just installed three AFS clients, and everything seems to wo= rk fine! The > only problem is that I would like to get my afs-token on login, witho= ut having > to type klog. > > Any idea? > > Use a pam module, such as Russ' pam-afs-session: > http://www.eyrie.org/~eagle/software/pam-afs-session/ > > -- > Mike Garrison > > From Jerry.Normandin@dafca.com Fri Nov 30 13:52:06 2007 From: Jerry.Normandin@dafca.com (Jerry Normandin) Date: Fri, 30 Nov 2007 05:52:06 -0800 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: <874pf4hmjt.fsf@windlord.stanford.edu> Message-ID: AFS on EXT3? No there are Metadata issues. EXT3 was intended for this. I inherited a mess here that I am fixing. My predecessor built is using Ext3 for the /vicepa filesystems. It takes a hell of a long time to create, Delete, or rename files. I tested with Bonnie++... here are my stats: Initial file performance baselines taken from ENG02 with 4096kbyte cache: read performance on disk is 6.7 x faster than AFS read performance on nfs is 2.0 x faser than AFS write performance is actually impressive. file creation and deletion are very slow on afs. file creation and deletion times is horrible. I will be working on a solution to solve this. Once file creation and deletion times are comparable to NFS, then AFS should perform well. Bonnie++ Benchmarks ENG02 AFS: Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP eng02.dafca.loca 8G 5296 47 6115 40 4074 34 19518 63 22494 7 125.6 4 ------Sequential Create------ --------Random Create-------- -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete-- files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP 16 23 1 7777 79 23 0 23 1 4546 66 18 0 eng02.dafca.local,8G,5296,47,6115,40,4074,34,19518,63,22494,7,125.6,4,16 ,23,1,7777,79,23,0,23,1,4546,66,18,0 ENG02 Local Disk: Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP eng02.dafca.loca 8G 35438 95 42686 23 16027 6 13516 31 24135 4 604.7 1 ------Sequential Create------ --------Random Create-------- -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete-- files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP 16 2046 77 +++++ +++ +++++ +++ 2230 84 +++++ +++ 1461 17 eng02.dafca.local,8G,35438,95,42686,23,16027,6,13516,31,24135,4,604.7,1, 16,2046,77,+++++,+++,+++++,+++,2230,84,+++++,+++,1461,17 ENG02 NFS: Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP eng02.dafca.loca 8G 11051 27 11126 3 8944 5 11179 27 11186 2 118.1 0 ------Sequential Create------ --------Random Create-------- -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete-- files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP 16 934 4 6092 12 2951 6 949 4 7922 12 2359 5 eng02.dafca.local,8G,11051,27,11126,3,8944,5,11179,27,11186,2,118.1,0,16 ,934,4,6092,12,2951,6,949,4,7922,12,2359,5 So I have to backup the vicepa filesystems. Create a xfs filesystem. And restore. I've got 3TB of data to deal with. Trust me you do not want to use ext3 for the /vicepa file system! -----Original Message----- From: openafs-info-admin@openafs.org [mailto:openafs-info-admin@openafs.org] On Behalf Of Russ Allbery Sent: Thursday, November 29, 2007 2:36 PM To: openafs-info@openafs.org Subject: Re: [OpenAFS] File systems on Linux, again. "Smith, Matt" writes: > After the recent thread "openafs upgrade from 1.4.1 to 1.5.7", and a > review of a thread[1] from July, I'm wondering if there is a definitive > recommendation for which file system to use on Linux AFS file servers. > Ext3, XFS, JFS, something else? It shouldn't make much of a difference, so I think you're safe choosing your file system on whatever basis you'd choose a file system for any other file server. We use ext3 because of the stability, reliability, and "center of the mainstream" support in the kernel, which we always considered more important than a bit of additional speed, but your mileage may vary. XFS is probably the next most common choice. I would be very leery of ReiserFS. It has nice features, but the recovery tools are fairly horrific. --=20 Russ Allbery (rra@stanford.edu) _______________________________________________ OpenAFS-info mailing list OpenAFS-info@openafs.org https://lists.openafs.org/mailman/listinfo/openafs-info From stephen@physics.unc.edu Fri Nov 30 14:26:11 2007 From: stephen@physics.unc.edu (Stephen Joyce) Date: Fri, 30 Nov 2007 09:26:11 -0500 (EST) Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: References: Message-ID: When you have bonnie++ numbers for xfs /vicep partitions, please post them. Mine are currently ext3, but I've warmed considerably to xfs since they were installed. On Fri, 30 Nov 2007, Jerry Normandin wrote: > AFS on EXT3? No there are Metadata issues. EXT3 was intended for this. > I inherited a mess here that I am fixing. My predecessor built is using > Ext3 for the /vicepa filesystems. It takes a hell of a long time to > create, > Delete, or rename files. I tested with Bonnie++... here are my stats: > > Initial file performance baselines taken from ENG02 with 4096kbyte > cache: > > read performance on disk is 6.7 x faster than AFS > read performance on nfs is 2.0 x faser than AFS > > write performance is actually impressive. file creation and deletion > are very slow on afs. > > file creation and deletion times is horrible. I will be working on a > solution to solve this. Once file creation and deletion times are > comparable to NFS, then AFS should perform well. > > Bonnie++ Benchmarks > > ENG02 AFS: > Version 1.03 ------Sequential Output------ --Sequential Input- > --Random- > -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- > --Seeks-- > Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP > /sec %CP > eng02.dafca.loca 8G 5296 47 6115 40 4074 34 19518 63 22494 7 > 125.6 4 > ------Sequential Create------ --------Random > Create-------- > -Create-- --Read--- -Delete-- -Create-- --Read--- > -Delete-- > files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP > /sec %CP > 16 23 1 7777 79 23 0 23 1 4546 66 > 18 0 > eng02.dafca.local,8G,5296,47,6115,40,4074,34,19518,63,22494,7,125.6,4,16 > ,23,1,7777,79,23,0,23,1,4546,66,18,0 > > > > ENG02 Local Disk: > > Version 1.03 ------Sequential Output------ --Sequential Input- > --Random- > -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- > --Seeks-- > Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP > /sec %CP > eng02.dafca.loca 8G 35438 95 42686 23 16027 6 13516 31 24135 4 > 604.7 1 > ------Sequential Create------ --------Random > Create-------- > -Create-- --Read--- -Delete-- -Create-- --Read--- > -Delete-- > files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP > /sec %CP > 16 2046 77 +++++ +++ +++++ +++ 2230 84 +++++ +++ > 1461 17 > eng02.dafca.local,8G,35438,95,42686,23,16027,6,13516,31,24135,4,604.7,1, > 16,2046,77,+++++,+++,+++++,+++,2230,84,+++++,+++,1461,17 > > > > ENG02 NFS: > Version 1.03 ------Sequential Output------ --Sequential Input- > --Random- > -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- > --Seeks-- > Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP > /sec %CP > eng02.dafca.loca 8G 11051 27 11126 3 8944 5 11179 27 11186 2 > 118.1 0 > ------Sequential Create------ --------Random > Create-------- > -Create-- --Read--- -Delete-- -Create-- --Read--- > -Delete-- > files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP > /sec %CP > 16 934 4 6092 12 2951 6 949 4 7922 12 > 2359 5 > eng02.dafca.local,8G,11051,27,11126,3,8944,5,11179,27,11186,2,118.1,0,16 > ,934,4,6092,12,2951,6,949,4,7922,12,2359,5 > > So I have to backup the vicepa filesystems. Create a xfs filesystem. > And restore. I've got 3TB of data to deal with. > > Trust me you do not want to use ext3 for the /vicepa file system! > > -----Original Message----- > From: openafs-info-admin@openafs.org > [mailto:openafs-info-admin@openafs.org] On Behalf Of Russ Allbery > Sent: Thursday, November 29, 2007 2:36 PM > To: openafs-info@openafs.org > Subject: Re: [OpenAFS] File systems on Linux, again. > > "Smith, Matt" writes: > >> After the recent thread "openafs upgrade from 1.4.1 to 1.5.7", and a >> review of a thread[1] from July, I'm wondering if there is a > definitive >> recommendation for which file system to use on Linux AFS file servers. >> Ext3, XFS, JFS, something else? > > It shouldn't make much of a difference, so I think you're safe choosing > your file system on whatever basis you'd choose a file system for any > other file server. We use ext3 because of the stability, reliability, > and > "center of the mainstream" support in the kernel, which we always > considered more important than a bit of additional speed, but your > mileage > may vary. > > XFS is probably the next most common choice. > > I would be very leery of ReiserFS. It has nice features, but the > recovery > tools are fairly horrific. > > -- > Russ Allbery (rra@stanford.edu) > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > From jason@rampaginggeek.com Fri Nov 30 15:58:52 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Fri, 30 Nov 2007 10:58:52 -0500 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? Message-ID: <4750333C.3030706@rampaginggeek.com> Hi everyone, Traditionally, we have used direct-attached scsi disk packs on Sun Sparc servers running Solaria 9 for OpenAFS. This has given us the most bang for the buck. We forgo RAID because we have the backup capabilities of AFS. What types of storage technologies are other AFS sites using for their AFS vicep partitions? We need to figure our future direction for the next couple of years. Fibre channel seems all the rage, but it's quite expensive. I'm open to any and all feedback. What works? What doesn't? What offers the best bang for the buck on an OpenAFS server? This is for an academic environment that fills both academic and research needs. Researchers are asking for lots of AFS space (200GB+). Of course this needs to be backed up as well. Thanks, Jason From sdevine@msu.edu Fri Nov 30 16:27:40 2007 From: sdevine@msu.edu (Steve Devine) Date: Fri, 30 Nov 2007 11:27:40 -0500 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: <4750333C.3030706@rampaginggeek.com> References: <4750333C.3030706@rampaginggeek.com> Message-ID: <475039FC.60303@msu.edu> Jason Edgecombe wrote: > Hi everyone, > > Traditionally, we have used direct-attached scsi disk packs on Sun Sparc > servers running Solaria 9 for OpenAFS. This has given us the most bang > for the buck. We forgo RAID because we have the backup capabilities of AFS. > > What types of storage technologies are other AFS sites using for their > AFS vicep partitions? We need to figure our future direction for the > next couple of years. Fibre channel seems all the rage, but it's quite > expensive. I'm open to any and all feedback. What works? What doesn't? > What offers the best bang for the buck on an OpenAFS server? > > This is for an academic environment that fills both academic and > research needs. Researchers are asking for lots of AFS space (200GB+). > Of course this needs to be backed up as well. > > Thanks, > Jason > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > We are using Iscsi for user vols. We put volumes that require higher performance on a Fiber Based San. -- Steve Devine Network Storage & Printing Academic Computing & Network Services Michigan State University 506 Computer Center East Lansing, MI 48824-1042 1-517-432-7327 Baseball is ninety percent mental; the other half is physical. - Yogi Berra From jason@rampaginggeek.com Fri Nov 30 16:34:22 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Fri, 30 Nov 2007 11:34:22 -0500 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: <475039FC.60303@msu.edu> References: <4750333C.3030706@rampaginggeek.com> <475039FC.60303@msu.edu> Message-ID: <47503B8E.2060308@rampaginggeek.com> Steve Devine wrote: > Jason Edgecombe wrote: >> Hi everyone, >> >> Traditionally, we have used direct-attached scsi disk packs on Sun Sparc >> servers running Solaria 9 for OpenAFS. This has given us the most bang >> for the buck. We forgo RAID because we have the backup capabilities >> of AFS. >> >> What types of storage technologies are other AFS sites using for their >> AFS vicep partitions? We need to figure our future direction for the >> next couple of years. Fibre channel seems all the rage, but it's quite >> expensive. I'm open to any and all feedback. What works? What doesn't? >> What offers the best bang for the buck on an OpenAFS server? >> >> This is for an academic environment that fills both academic and >> research needs. Researchers are asking for lots of AFS space (200GB+). >> Of course this needs to be backed up as well. >> >> Thanks, >> Jason >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info >> > We are using Iscsi for user vols. We put volumes that require higher > performance on a Fiber Based San. > What's serving the iSCSI? Is it a storage appliance of some sort? Thanks, Jason From chas@cmf.nrl.navy.mil Fri Nov 30 16:37:09 2007 From: chas@cmf.nrl.navy.mil (chas williams - CONTRACTOR) Date: Fri, 30 Nov 2007 11:37:09 -0500 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: <4750333C.3030706@rampaginggeek.com> Message-ID: <200711301637.lAUGb9Kt026133@cmf.nrl.navy.mil> In message <4750333C.3030706@rampaginggeek.com>,Jason Edgecombe writes: >next couple of years. Fibre channel seems all the rage, but it's quite >expensive. I'm open to any and all feedback. What works? What doesn't? >What offers the best bang for the buck on an OpenAFS server? i went pretty cheap last time. i am using solaris volume manager raid1's on some built-in sata drives on solaris 10. a hardware raid might handle disk failures easier (just replace the drive) but nice hardware raids can be fairly expensive. svm lets you round-robin read accesses (per thread) between the mirrors which can potentially boost read performance a bit. its cheap enough that adding another x86 server isnt very expensive. each server here fits into 1U and provides +400G. of course, we are only supporting a couple hundred users. >This is for an academic environment that fills both academic and >research needs. Researchers are asking for lots of AFS space (200GB+). >Of course this needs to be backed up as well. sata drives just keep getting bigger, so you can pick up a 350GB or 500GB drive fairly cheap. you can finally get a 5-year warranty on some sata drives which makes them fairly attractive. From rob@nofocus.org Fri Nov 30 16:39:05 2007 From: rob@nofocus.org (Rob Banz) Date: Fri, 30 Nov 2007 11:39:05 -0500 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: <4750333C.3030706@rampaginggeek.com> References: <4750333C.3030706@rampaginggeek.com> Message-ID: Don't forgo some sort of live data protection -- most likely via RAID. If these volumes are RW, and you have a storage failure, you're going to be putting yourself and your users through hell waiting for stuff to be restored. If you're not looking for super performance, go for a RAID5ish solution (or something like a JBOD + raidz using ZFS on Solaris), or if performance is an issue, hardware or software mirroring is the way to go. Storage is cheap -- your time and your users isn't. There are affordable FC-attached options out there, such as the Apple XRAID*, and other similar options from smaller vendors. You can probably find yourself some similar-priced options that do iSCSI from some of the smaller storage vendors out there. Also -- avoid the Linux game. Go with Solaris 10 and ZFS. You've got a solid storage architecture, and a top-of-the-line filesystem, and you can skip these silly ext3/ext2/reiser/xfs/xvm discussions on the list ;) -rob * Apple basically gives these things away to higher-ed, and they offer pretty damn nice pricing on Qlogic's stackable switches. Talk to your apple rep. They work great with Solaris too ;) On Nov 30, 2007, at 10:58, Jason Edgecombe wrote: > Hi everyone, > > Traditionally, we have used direct-attached scsi disk packs on Sun > Sparc > servers running Solaria 9 for OpenAFS. This has given us the most bang > for the buck. We forgo RAID because we have the backup capabilities > of AFS. > > What types of storage technologies are other AFS sites using for their > AFS vicep partitions? We need to figure our future direction for the > next couple of years. Fibre channel seems all the rage, but it's quite > expensive. I'm open to any and all feedback. What works? What doesn't? > What offers the best bang for the buck on an OpenAFS server? > > This is for an academic environment that fills both academic and > research needs. Researchers are asking for lots of AFS space (200GB+). > Of course this needs to be backed up as well. > > Thanks, > Jason > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info From chas@cmf.nrl.navy.mil Fri Nov 30 16:42:14 2007 From: chas@cmf.nrl.navy.mil (chas williams - CONTRACTOR) Date: Fri, 30 Nov 2007 11:42:14 -0500 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: Message-ID: <200711301642.lAUGgEuF026262@cmf.nrl.navy.mil> In message ,"Jerry Normandin" writes: >write performance is actually impressive. file creation and deletion >are very slow on afs. because writing is easier than reading. the afs cache manager can group the outgoing writes together and send them in a single message. while the cache manager has readahead it doesnt work because the afs global locks blocks any progress the readahead thread might make. From matt@linuxbox.com Fri Nov 30 17:10:15 2007 From: matt@linuxbox.com (Matt Benjamin) Date: Fri, 30 Nov 2007 12:10:15 -0500 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: <200711301642.lAUGgEuF026262@cmf.nrl.navy.mil> References: <200711301642.lAUGgEuF026262@cmf.nrl.navy.mil> Message-ID: <475043F7.1010101@linuxbox.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Hey, Chas, Sorry to bug. I've been looking at this, tangentially, because I've been working with bypassing dcache/memcache, writing direct into page cache. Pretty far along on that side of things. Is rx locking so coarse that in general only one read makes progress even independent of the cache--to your knowledge? Matt chas williams - CONTRACTOR wrote: > In message ia.net>,"Jerry Normandin" writes: >> write performance is actually impressive. file creation and deletion >> are very slow on afs. > > because writing is easier than reading. the afs cache manager can > group the outgoing writes together and send them in a single message. > while the cache manager has readahead it doesnt work because the afs > global locks blocks any progress the readahead thread might make. > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info - -- Matt Benjamin The Linux Box 206 South Fifth Ave. Suite 150 Ann Arbor, MI 48104 http://linuxbox.com tel. 734-761-4689 fax. 734-769-8938 cel. 734-216-5309 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFHUEJRJiSUUSaRdSURCA2lAJ4p19x8OH8y2cYO3WERwapwdZ9gYwCffFeR t96QUFYv5yHK6wlFTbn9TSE= =BAnD -----END PGP SIGNATURE----- From chas@cmf.nrl.navy.mil Fri Nov 30 17:58:18 2007 From: chas@cmf.nrl.navy.mil (chas williams - CONTRACTOR) Date: Fri, 30 Nov 2007 12:58:18 -0500 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: <475043F7.1010101@linuxbox.com> Message-ID: <200711301758.lAUHwIVR027540@cmf.nrl.navy.mil> no i dont think rx locking is the problem. the rx locking is actually pretty good. i had tracked this down with fstrace at one point but i seem to have lost the trace at the moment. i will dig around and see if i can find it. In message <475043F7.1010101@linuxbox.com>,Matt Benjamin writes: >-----BEGIN PGP SIGNED MESSAGE----- >Hash: SHA256 > >Hey, Chas, > >Sorry to bug. > >I've been looking at this, tangentially, because I've been working with >bypassing dcache/memcache, writing direct into page cache. Pretty far >along on that side of things. Is rx locking so coarse that in general >only one read makes progress even independent of the cache--to your >knowledge? > >Matt > >chas williams - CONTRACTOR wrote: >> In message med >> ia.net>,"Jerry Normandin" writes: >>> write performance is actually impressive. file creation and deletion >>> are very slow on afs. >> >> because writing is easier than reading. the afs cache manager can >> group the outgoing writes together and send them in a single message. >> while the cache manager has readahead it doesnt work because the afs >> global locks blocks any progress the readahead thread might make. >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info > > > > >- -- > >Matt Benjamin > >The Linux Box >206 South Fifth Ave. Suite 150 >Ann Arbor, MI 48104 > >http://linuxbox.com > >tel. 734-761-4689 >fax. 734-769-8938 >cel. 734-216-5309 > >-----BEGIN PGP SIGNATURE----- >Version: GnuPG v1.4.7 (GNU/Linux) >Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > >iD8DBQFHUEJRJiSUUSaRdSURCA2lAJ4p19x8OH8y2cYO3WERwapwdZ9gYwCffFeR >t96QUFYv5yHK6wlFTbn9TSE= >=BAnD >-----END PGP SIGNATURE----- >_______________________________________________ >OpenAFS-info mailing list >OpenAFS-info@openafs.org >https://lists.openafs.org/mailman/listinfo/openafs-info > From sdevine@msu.edu Fri Nov 30 18:10:15 2007 From: sdevine@msu.edu (Steve Devine) Date: Fri, 30 Nov 2007 13:10:15 -0500 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: <47503B8E.2060308@rampaginggeek.com> References: <4750333C.3030706@rampaginggeek.com> <475039FC.60303@msu.edu> <47503B8E.2060308@rampaginggeek.com> Message-ID: <47505207.5020906@msu.edu> Jason Edgecombe wrote: > Steve Devine wrote: > >> Jason Edgecombe wrote: >> >>> Hi everyone, >>> >>> Traditionally, we have used direct-attached scsi disk packs on Sun Sparc >>> servers running Solaria 9 for OpenAFS. This has given us the most bang >>> for the buck. We forgo RAID because we have the backup capabilities >>> of AFS. >>> >>> What types of storage technologies are other AFS sites using for their >>> AFS vicep partitions? We need to figure our future direction for the >>> next couple of years. Fibre channel seems all the rage, but it's quite >>> expensive. I'm open to any and all feedback. What works? What doesn't? >>> What offers the best bang for the buck on an OpenAFS server? >>> >>> This is for an academic environment that fills both academic and >>> research needs. Researchers are asking for lots of AFS space (200GB+). >>> Of course this needs to be backed up as well. >>> >>> Thanks, >>> Jason >>> _______________________________________________ >>> OpenAFS-info mailing list >>> OpenAFS-info@openafs.org >>> https://lists.openafs.org/mailman/listinfo/openafs-info >>> >>> >> We are using Iscsi for user vols. We put volumes that require higher >> performance on a Fiber Based San. >> >> > What's serving the iSCSI? Is it a storage appliance of some sort? > > Thanks, > Jason > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > We are using a unit from infortrend.com its ISCSi to SATA. (16) 500G drives in a raid6 config. So we are getting about 6.5 T out of it. Some things about this scare me I won't deny but it is a lot of disk space for reasonable prices. (seems like it was around 7~8k) Seek time is an issue when you have 350G partitions. I also worry about salvage times if we suffer a crash .. but I compiled the fileservers with fast restart and so far I have not had to test it. We have been online with this device about 10 months. So far so good. -- Steve Devine Network Storage & Printing Academic Computing & Network Services Michigan State University 506 Computer Center East Lansing, MI 48824-1042 1-517-432-7327 Baseball is ninety percent mental; the other half is physical. - Yogi Berra From matt@linuxbox.com Fri Nov 30 18:16:00 2007 From: matt@linuxbox.com (Matt Benjamin) Date: Fri, 30 Nov 2007 13:16:00 -0500 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: <200711301758.lAUHwIVR027540@cmf.nrl.navy.mil> References: <200711301758.lAUHwIVR027540@cmf.nrl.navy.mil> Message-ID: <47505360.5010800@linuxbox.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 woot, chas :) chas williams - CONTRACTOR wrote: > no i dont think rx locking is the problem. the rx locking is > actually pretty good. i had tracked this down with fstrace > at one point but i seem to have lost the trace at the moment. > i will dig around and see if i can find it. > > In message <475043F7.1010101@linuxbox.com>,Matt Benjamin writes: > Hey, Chas, > - -- Matt Benjamin The Linux Box 206 South Fifth Ave. Suite 150 Ann Arbor, MI 48104 http://linuxbox.com tel. 734-761-4689 fax. 734-769-8938 cel. 734-216-5309 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFHUFNgJiSUUSaRdSURCH7/AJ9eTtSta1JCYEcL4jt2kaPH4rm25QCePpXS 8ozxvvFRngC1EMZz1vXY/0E= =koC6 -----END PGP SIGNATURE----- From stephen@physics.unc.edu Fri Nov 30 18:51:25 2007 From: stephen@physics.unc.edu (Stephen Joyce) Date: Fri, 30 Nov 2007 13:51:25 -0500 (EST) Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: <4750333C.3030706@rampaginggeek.com> References: <4750333C.3030706@rampaginggeek.com> Message-ID: I don't have money for FC or a SAN, so I've stuck with DAS. I've had good experience with building many smallish servers rather than one big or expensive one. I'm currently using cheap Dell PowerEdge servers running linux. I think we got them for about $800/ea, and they support console redirection (critical when you have lots of physical servers). We added a 2-port 3ware raid1 for the OS and a 4-port 3ware for the data (raid1 or raid5 depending on the requirements). Right now I'm keeping the servers to around 1TB each, but they're capable of hosting 2-4TB each (depending on raid level) with the largest current drives. If money were no object, I'd have opted for hot-swappable drives, but with under 1TB of data on each, any time I've needed to replace a drive I've just moved the volumes to another server. These systems are cheap enough (under about $1.5K each for everything) that I keep a spare of everything just in case (spare fully configured and running server plus spare raid cards and drives on the shelf). I _strongly_ advise raid. Raid1 for the OS and raid1, 5, or 6 for the data, depending on your requirements. I know some people have reported impressive results with linux software raid, but I swear by 3ware hardware raid controllers; they "just work." Just avoid "fakeraid" controller cards (promise, low-end adaptec, etc) like the plague. They're far more trouble than they're worth. I really like solaris, but this setup is MUCH cheaper and faster than our old solaris setup. On Fri, 30 Nov 2007, Jason Edgecombe wrote: > Hi everyone, > > Traditionally, we have used direct-attached scsi disk packs on Sun Sparc > servers running Solaria 9 for OpenAFS. This has given us the most bang > for the buck. We forgo RAID because we have the backup capabilities of AFS. > > What types of storage technologies are other AFS sites using for their > AFS vicep partitions? We need to figure our future direction for the > next couple of years. Fibre channel seems all the rage, but it's quite > expensive. I'm open to any and all feedback. What works? What doesn't? > What offers the best bang for the buck on an OpenAFS server? > > This is for an academic environment that fills both academic and > research needs. Researchers are asking for lots of AFS space (200GB+). > Of course this needs to be backed up as well. > > Thanks, > Jason Cheers, Stephen -- Stephen Joyce Systems Administrator P A N I C Physics & Astronomy Department Physics & Astronomy University of North Carolina at Chapel Hill Network Infrastructure voice: (919) 962-7214 and Computing fax: (919) 962-0480 http://www.panic.unc.edu Don't judge a book by its movie. From rra@stanford.edu Fri Nov 30 19:17:39 2007 From: rra@stanford.edu (Russ Allbery) Date: Fri, 30 Nov 2007 11:17:39 -0800 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: (Jerry Normandin's message of "Fri\, 30 Nov 2007 05\:52\:06 -0800") References: Message-ID: <87bq9b35m4.fsf@windlord.stanford.edu> "Jerry Normandin" writes: > AFS on EXT3? No there are Metadata issues. EXT3 was intended for this. > I inherited a mess here that I am fixing. My predecessor built is using > Ext3 for the /vicepa filesystems. It takes a hell of a long time to > create, Delete, or rename files. *shrug*. It works just fine for us. -- Russ Allbery (rra@stanford.edu) From deej@duke.edu Fri Nov 30 19:22:03 2007 From: deej@duke.edu (Dj Merrill) Date: Fri, 30 Nov 2007 14:22:03 -0500 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: <87bq9b35m4.fsf@windlord.stanford.edu> References: <87bq9b35m4.fsf@windlord.stanford.edu> Message-ID: <475062DB.1070505@duke.edu> Russ Allbery wrote: > "Jerry Normandin" writes: > >> AFS on EXT3? No there are Metadata issues. EXT3 was intended for this. >> I inherited a mess here that I am fixing. My predecessor built is using >> Ext3 for the /vicepa filesystems. It takes a hell of a long time to >> create, Delete, or rename files. > > *shrug*. It works just fine for us. > We've been using ext3 for /vicepX partitions for over 3 years. Works just fine for us, too. Jerry, any chance there might be something else that is causing your slowness? -Dj From jaltman@secure-endpoints.com Fri Nov 30 19:37:30 2007 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Fri, 30 Nov 2007 14:37:30 -0500 Subject: [OpenAFS] File systems on Linux, again. In-Reply-To: References: Message-ID: <4750667A.5060209@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms080406030306090304040505 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Jerry Normandin wrote: > AFS on EXT3? No there are Metadata issues. EXT3 was intended for this. > I inherited a mess here that I am fixing. My predecessor built is using > Ext3 for the /vicepa filesystems. It takes a hell of a long time to > create, > Delete, or rename files. I tested with Bonnie++... here are my stats: > > Initial file performance baselines taken from ENG02 with 4096kbyte > cache: > > read performance on disk is 6.7 x faster than AFS > read performance on nfs is 2.0 x faser than AFS > > write performance is actually impressive. file creation and deletion > are very slow on afs. > > file creation and deletion times is horrible. I will be working on a > solution to solve this. Once file creation and deletion times are > comparable to NFS, then AFS should perform well. Upgrade to 1.4.5 and check your performance again. You should see a measurable improvement on EXT3. Jeffrey Altman --------------ms080406030306090304040505 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMzAxOTM3MzBaMCMGCSqGSIb3DQEJBDEWBBSbsXfk QuZr83QsIXp3kXd8dxg0VDBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEAXmIQlYSSUeS8AHfHSyxc+zQJGO9p2qrEcHdng0dNZYCshIiswyJx npbYR9/x4/zootoKeT9CF4A0EAUGuXWsiZT0uZSvGuTkarzx6dadUcKXE9Sd2xWE9j/dX8Cc mQFjcqfddByggDOPrKx19nMy0JOX32YMj4XRaw7mHMvEaGehnBAkPNTcWcT/tTlvNSW6HGT5 oKJnmstZJca9xyoq1z2FlS62xZecTZ7WEKiBVAHXbI5AHL6bhuCgw9AcSnHvm/CiJUVJnEcE U5s4FmC8qMB+br5I7PVDHsglKs/C4M8jNI9vpDkfcZC4tG4pSC9i1pltxZrh3Y0MG5imtSnI qgAAAAAAAA== --------------ms080406030306090304040505-- From rra@stanford.edu Fri Nov 30 19:52:14 2007 From: rra@stanford.edu (Russ Allbery) Date: Fri, 30 Nov 2007 11:52:14 -0800 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: <4750333C.3030706@rampaginggeek.com> (Jason Edgecombe's message of "Fri\, 30 Nov 2007 10\:58\:52 -0500") References: <4750333C.3030706@rampaginggeek.com> Message-ID: <87y7cfzf2p.fsf@windlord.stanford.edu> Jason Edgecombe writes: > What types of storage technologies are other AFS sites using for their > AFS vicep partitions? We need to figure our future direction for the > next couple of years. Fibre channel seems all the rage, but it's quite > expensive. I'm open to any and all feedback. What works? What doesn't? > What offers the best bang for the buck on an OpenAFS server? We're currently using fibre-channel SAN to Nexsan disk and some fibre-channel to EMC disk, and soon will be experimenting with fibre-channel to Netapp. We're also considering experimenting with NFS-mounting Netapp and using that as an AFS vice partition, but that's more in the way of a science experiment rather than something we'll probably run in production. -- Russ Allbery (rra@stanford.edu) From rob@nofocus.org Fri Nov 30 19:54:02 2007 From: rob@nofocus.org (Rob Banz) Date: Fri, 30 Nov 2007 14:54:02 -0500 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: References: <4750333C.3030706@rampaginggeek.com> Message-ID: <24C08CC8-03CF-4F4F-9350-C62A956312C0@nofocus.org> Run solaris x86! The site I used to work at has been using it exclusively on its afs servers since sol 10 came out. -rob On Nov 30, 2007, at 13:51, Stephen Joyce wrote: > I don't have money for FC or a SAN, so I've stuck with DAS. I've had > good experience with building many smallish servers rather than one > big or expensive one. > > I'm currently using cheap Dell PowerEdge servers running linux. I > think we got them for about $800/ea, and they support console > redirection (critical when you have lots of physical servers). We > added a 2-port 3ware raid1 for the OS and a 4-port 3ware for the > data (raid1 or raid5 depending on the requirements). Right now I'm > keeping the servers to around 1TB each, but they're capable of > hosting 2-4TB each (depending on raid level) with the largest > current drives. > > If money were no object, I'd have opted for hot-swappable drives, > but with under 1TB of data on each, any time I've needed to replace > a drive I've just moved the volumes to another server. > > These systems are cheap enough (under about $1.5K each for > everything) that I keep a spare of everything just in case (spare > fully configured and running server plus spare raid cards and drives > on the shelf). > > I _strongly_ advise raid. Raid1 for the OS and raid1, 5, or 6 for > the data, depending on your requirements. I know some people have > reported impressive results with linux software raid, but I swear by > 3ware hardware raid controllers; they "just work." Just avoid > "fakeraid" controller cards (promise, low-end adaptec, etc) like the > plague. They're far more trouble than they're worth. > > I really like solaris, but this setup is MUCH cheaper and faster > than our old solaris setup. > > On Fri, 30 Nov 2007, Jason Edgecombe wrote: > >> Hi everyone, >> >> Traditionally, we have used direct-attached scsi disk packs on Sun >> Sparc >> servers running Solaria 9 for OpenAFS. This has given us the most >> bang >> for the buck. We forgo RAID because we have the backup capabilities >> of AFS. >> >> What types of storage technologies are other AFS sites using for >> their >> AFS vicep partitions? We need to figure our future direction for the >> next couple of years. Fibre channel seems all the rage, but it's >> quite >> expensive. I'm open to any and all feedback. What works? What >> doesn't? >> What offers the best bang for the buck on an OpenAFS server? >> >> This is for an academic environment that fills both academic and >> research needs. Researchers are asking for lots of AFS space (200GB >> +). >> Of course this needs to be backed up as well. >> >> Thanks, >> Jason > > Cheers, Stephen > -- > Stephen Joyce > Systems Administrator P A > N I C > Physics & Astronomy Department Physics & > Astronomy > University of North Carolina at Chapel Hill Network > Infrastructure > voice: (919) 962-7214 and > Computing > fax: (919) 962-0480 http://www.panic.unc.edu > > Don't judge a book by its movie. > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info From rob@nofocus.org Fri Nov 30 19:54:02 2007 From: rob@nofocus.org (Rob Banz) Date: Fri, 30 Nov 2007 14:54:02 -0500 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: References: <4750333C.3030706@rampaginggeek.com> Message-ID: <24C08CC8-03CF-4F4F-9350-C62A956312C0@nofocus.org> Run solaris x86! The site I used to work at has been using it exclusively on its afs servers since sol 10 came out. -rob On Nov 30, 2007, at 13:51, Stephen Joyce wrote: > I don't have money for FC or a SAN, so I've stuck with DAS. I've had > good experience with building many smallish servers rather than one > big or expensive one. > > I'm currently using cheap Dell PowerEdge servers running linux. I > think we got them for about $800/ea, and they support console > redirection (critical when you have lots of physical servers). We > added a 2-port 3ware raid1 for the OS and a 4-port 3ware for the > data (raid1 or raid5 depending on the requirements). Right now I'm > keeping the servers to around 1TB each, but they're capable of > hosting 2-4TB each (depending on raid level) with the largest > current drives. > > If money were no object, I'd have opted for hot-swappable drives, > but with under 1TB of data on each, any time I've needed to replace > a drive I've just moved the volumes to another server. > > These systems are cheap enough (under about $1.5K each for > everything) that I keep a spare of everything just in case (spare > fully configured and running server plus spare raid cards and drives > on the shelf). > > I _strongly_ advise raid. Raid1 for the OS and raid1, 5, or 6 for > the data, depending on your requirements. I know some people have > reported impressive results with linux software raid, but I swear by > 3ware hardware raid controllers; they "just work." Just avoid > "fakeraid" controller cards (promise, low-end adaptec, etc) like the > plague. They're far more trouble than they're worth. > > I really like solaris, but this setup is MUCH cheaper and faster > than our old solaris setup. > > On Fri, 30 Nov 2007, Jason Edgecombe wrote: > >> Hi everyone, >> >> Traditionally, we have used direct-attached scsi disk packs on Sun >> Sparc >> servers running Solaria 9 for OpenAFS. This has given us the most >> bang >> for the buck. We forgo RAID because we have the backup capabilities >> of AFS. >> >> What types of storage technologies are other AFS sites using for >> their >> AFS vicep partitions? We need to figure our future direction for the >> next couple of years. Fibre channel seems all the rage, but it's >> quite >> expensive. I'm open to any and all feedback. What works? What >> doesn't? >> What offers the best bang for the buck on an OpenAFS server? >> >> This is for an academic environment that fills both academic and >> research needs. Researchers are asking for lots of AFS space (200GB >> +). >> Of course this needs to be backed up as well. >> >> Thanks, >> Jason > > Cheers, Stephen > -- > Stephen Joyce > Systems Administrator P A > N I C > Physics & Astronomy Department Physics & > Astronomy > University of North Carolina at Chapel Hill Network > Infrastructure > voice: (919) 962-7214 and > Computing > fax: (919) 962-0480 http://www.panic.unc.edu > > Don't judge a book by its movie. > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info From john@iastate.edu Fri Nov 30 20:27:09 2007 From: john@iastate.edu (John Hascall) Date: Fri, 30 Nov 2007 14:27:09 CST Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: Your message of Fri, 30 Nov 2007 13:51:25 -0500. Message-ID: <13639.1196454429@malison.ait.iastate.edu> Stephen Joyce replies > I don't have money for FC or a SAN, so I've stuck with DAS. I've had good > experience with building many smallish servers rather than one big or > expensive one. DAS is our approach as well. We are using "white boxes" built from a hot-swapable Chenbro case with 3ware's 12 port SATA card in a raid-10 config. We're getting down pretty close to about $1K/TB. John From jason@rampaginggeek.com Fri Nov 30 20:46:01 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Fri, 30 Nov 2007 15:46:01 -0500 Subject: [OpenAFS] 1.4.5 namei on solaris 9 sparc requires AlwaysAttach for vice partitions In-Reply-To: References: <474F2250.2010403@rampaginggeek.com> <474F2969.7010102@rampaginggeek.com> Message-ID: <47507689.5070902@rampaginggeek.com> Derrick Brashear wrote: > > > > Does anyone know why this would be happening? > > > > Probably a bug in the "what's acceptable as a vice partition" > logic... > > which I thought i fixed before 1.4.5; i bet i committed the wrong > > thing (because i tested it) > > > > > Is it safe to run like this? > > > yup > > > > Should I file a bug? > > > probably > I applied the STABLE14-namei-allow-ufs-20071129 patch to 1.4.5 and recompiled. It's working now. Consider the issue #78813 fixed. Thanks, Jason From jason@rampaginggeek.com Fri Nov 30 20:53:47 2007 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Fri, 30 Nov 2007 15:53:47 -0500 Subject: [OpenAFS] Deletion of clones & source volumes takes a while with vos move Message-ID: <4750785B.5010102@rampaginggeek.com> When running a vos move, deleting a clone or the source volume seems to take almost as much time as the copying of the data itself. Does AFS scrub the disk when deleting volumes or does it just release the inodes? Platform: Solaris 9 SPARC on ufs filesystem. Just a curious question, no issue, per se. Thanks, Jason From cclausen@acm.org Fri Nov 30 21:05:08 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Fri, 30 Nov 2007 15:05:08 -0600 Subject: [OpenAFS] Deletion of clones & source volumes takes a while with vos move References: <4750785B.5010102@rampaginggeek.com> Message-ID: <9B69C77531164500ADA669687BA36284@CDCHOME> Jason Edgecombe wrote: > When running a vos move, deleting a clone or the source volume seems > to take almost as much time as the copying of the data itself. > > Does AFS scrub the disk when deleting volumes or does it just release > the inodes? Not that I am aware of. > Platform: Solaris 9 SPARC on ufs filesystem. Apply the no-fsync patch or reduce the number of files in your volumes. < References: <4750785B.5010102@rampaginggeek.com> <9B69C77531164500ADA669687BA36284@CDCHOME> Message-ID: <47507D3B.8050402@secure-endpoints.com> This is a cryptographically signed message in MIME format. --------------ms030407020005070502090106 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Christopher D. Clausen wrote: > Apply the no-fsync patch There is no patch to apply. Just update to 1.4.5. --------------ms030407020005070502090106 Content-Type: application/x-pkcs7-signature; name="smime.p7s" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="smime.p7s" Content-Description: S/MIME Cryptographic Signature MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIJeTCC AxcwggKAoAMCAQICEALr5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UE BhMCWkExJTAjBgNVBAoTHFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMT I1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oX DTA4MDUzMDA2MTM1N1owczEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVy aWMxHDAaBgNVBAMTE0plZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIB AQCsoz/0+s4Cn65n/3bU3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5c bhSnJPmb2OOb1frx64dlDgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCF PKhZfcrGAXoVCrFBadnLAokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu184 9Cj6I4nY5bt+r94kyDHM94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d 1ZXepBXyexQbwOYvftpVlMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcG A1UdEQQgMB6BHGphbHRtYW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADAN BgkqhkiG9w0BAQUFAAOBgQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1 AZAvCv+ZMMRRKWCsrG5tItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk 0oiBQcQD8A+UHwrcmaj3UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAxcwggKAoAMCAQICEALr 5BE3U6n+HWCoLbyhohMwDQYJKoZIhvcNAQEFBQAwYjELMAkGA1UEBhMCWkExJTAjBgNVBAoT HFRoYXd0ZSBDb25zdWx0aW5nIChQdHkpIEx0ZC4xLDAqBgNVBAMTI1RoYXd0ZSBQZXJzb25h bCBGcmVlbWFpbCBJc3N1aW5nIENBMB4XDTA3MDUzMTA2MTM1N1oXDTA4MDUzMDA2MTM1N1ow czEPMA0GA1UEBBMGQWx0bWFuMRUwEwYDVQQqEwxKZWZmcmV5IEVyaWMxHDAaBgNVBAMTE0pl ZmZyZXkgRXJpYyBBbHRtYW4xKzApBgkqhkiG9w0BCQEWHGphbHRtYW5Ac2VjdXJlLWVuZHBv aW50cy5jb20wggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCsoz/0+s4Cn65n/3bU 3shXw4y5u1uEMEsBOiqNU0PfIKGYQe95b1FKNbNAkctSdQT6GF5cbhSnJPmb2OOb1frx64dl DgskaG561xa8XPA1aP8Cc+33dgsSLIxGEh97lyUYHEfWBC03KMCFPKhZfcrGAXoVCrFBadnL AokQbUTFahVg/qQx2IT3wSj1sCIfV5UDuXcEKHCvRtEZIsSzu1849Cj6I4nY5bt+r94kyDHM 94MHYBJi+6tWLFRy2gkIB3HEPmxAiQrKljNpH9bOffiBLIAgmJ6d1ZXepBXyexQbwOYvftpV lMEFHHQmdiwH3tj69hE78XvM5X9J+SbjbuNpAgMBAAGjOTA3MCcGA1UdEQQgMB6BHGphbHRt YW5Ac2VjdXJlLWVuZHBvaW50cy5jb20wDAYDVR0TAQH/BAIwADANBgkqhkiG9w0BAQUFAAOB gQB8FShDN2Ig034Y5eyadiFDEtOvsIJ3Z2xV9aTL4u8xMlz1gZR1AZAvCv+ZMMRRKWCsrG5t ItV8DFPSfWAGMpInmMarA4f76JRLQEUhkRUg8GpkJM5ryk5EDakk0oiBQcQD8A+UHwrcmaj3 UWxQ9zCjDgU+1mY9nEQxZZyp4eeUfzCCAz8wggKooAMCAQICAQ0wDQYJKoZIhvcNAQEFBQAw gdExCzAJBgNVBAYTAlpBMRUwEwYDVQQIEwxXZXN0ZXJuIENhcGUxEjAQBgNVBAcTCUNhcGUg VG93bjEaMBgGA1UEChMRVGhhd3RlIENvbnN1bHRpbmcxKDAmBgNVBAsTH0NlcnRpZmljYXRp b24gU2VydmljZXMgRGl2aXNpb24xJDAiBgNVBAMTG1RoYXd0ZSBQZXJzb25hbCBGcmVlbWFp bCBDQTErMCkGCSqGSIb3DQEJARYccGVyc29uYWwtZnJlZW1haWxAdGhhd3RlLmNvbTAeFw0w MzA3MTcwMDAwMDBaFw0xMzA3MTYyMzU5NTlaMGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxU aGF3dGUgQ29uc3VsdGluZyAoUHR5KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwg RnJlZW1haWwgSXNzdWluZyBDQTCBnzANBgkqhkiG9w0BAQEFAAOBjQAwgYkCgYEAxKY8VXNV +065yplaHmjAdQRwnd/p/6Me7L3N9VvyGna9fww6YfK/Uc4B1OVQCjDXAmNaLIkVcI7dyfAr hVqqP3FWy688Cwfn8R+RNiQqE88r1fOCdz0Dviv+uxg+B79AgAJk16emu59l0cUqVIUPSAR/ p7bRPGEEQB5kGXJgt/sCAwEAAaOBlDCBkTASBgNVHRMBAf8ECDAGAQH/AgEAMEMGA1UdHwQ8 MDowOKA2oDSGMmh0dHA6Ly9jcmwudGhhd3RlLmNvbS9UaGF3dGVQZXJzb25hbEZyZWVtYWls Q0EuY3JsMAsGA1UdDwQEAwIBBjApBgNVHREEIjAgpB4wHDEaMBgGA1UEAxMRUHJpdmF0ZUxh YmVsMi0xMzgwDQYJKoZIhvcNAQEFBQADgYEASIzRUIPqCy7MDaNmrGcPf6+svsIXoUOWlJ1/ TCG4+DYfqi2fNi/A9BxQIJNwPP2t4WFiw9k6GX6EsZkbAMUaC4J0niVQlGLH2ydxVyWN3amc OY6MIE9lX5Xa9/eH1sYITq726jTlEBpbNU1341YheILcIRk13iSx0x1G/11fZU8xggNkMIID YAIBATB2MGIxCzAJBgNVBAYTAlpBMSUwIwYDVQQKExxUaGF3dGUgQ29uc3VsdGluZyAoUHR5 KSBMdGQuMSwwKgYDVQQDEyNUaGF3dGUgUGVyc29uYWwgRnJlZW1haWwgSXNzdWluZyBDQQIQ AuvkETdTqf4dYKgtvKGiEzAJBgUrDgMCGgUAoIIBwzAYBgkqhkiG9w0BCQMxCwYJKoZIhvcN AQcBMBwGCSqGSIb3DQEJBTEPFw0wNzExMzAyMTE0MzVaMCMGCSqGSIb3DQEJBDEWBBS293Y9 gjXzfqgpMvF0fJ0ctu/KhzBSBgkqhkiG9w0BCQ8xRTBDMAoGCCqGSIb3DQMHMA4GCCqGSIb3 DQMCAgIAgDANBggqhkiG9w0DAgIBQDAHBgUrDgMCBzANBggqhkiG9w0DAgIBKDCBhQYJKwYB BAGCNxAEMXgwdjBiMQswCQYDVQQGEwJaQTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcg KFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhhd3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3Vpbmcg Q0ECEALr5BE3U6n+HWCoLbyhohMwgYcGCyqGSIb3DQEJEAILMXigdjBiMQswCQYDVQQGEwJa QTElMCMGA1UEChMcVGhhd3RlIENvbnN1bHRpbmcgKFB0eSkgTHRkLjEsMCoGA1UEAxMjVGhh d3RlIFBlcnNvbmFsIEZyZWVtYWlsIElzc3VpbmcgQ0ECEALr5BE3U6n+HWCoLbyhohMwDQYJ KoZIhvcNAQEBBQAEggEATkcs4ezY4k9l6yy9v8keQHSlSIQEmXP8UE1OLS8z03EQ8E3gcILc iNaIZ5Wa/XrUNQJFZXEa+lbaK28Xpa7fjD/1KyXvweXVrHYva/iO+O5IUaYvLtAeeHnBckc+ 1F6wVMxajRZTr4VCD0X7Idz7r9OXEz4bqQ3pwf6MVX3mpM8CxbJZ0cD4DvZGmMyZZtmp6zyW OyFFjW8s2Z92QvCmo0LEdSFUB6t0j3xNnyJuI5eLTKIjglW6Z/SdLwK9rMk1MjxFHcRd9MZu cKVaMChRIvpue4RnAytZ5msA4uZHcTF0v8o6P7Kke/+tP2+fdknxOMiiK/yEZvxWZ8UpG3NH 6QAAAAAAAA== --------------ms030407020005070502090106-- From cclausen@acm.org Fri Nov 30 21:23:07 2007 From: cclausen@acm.org (Christopher D. Clausen) Date: Fri, 30 Nov 2007 15:23:07 -0600 Subject: [OpenAFS] Deletion of clones & source volumes takes a while with vos move References: <4750785B.5010102@rampaginggeek.com> <9B69C77531164500ADA669687BA36284@CDCHOME> <47507D3B.8050402@secure-endpoints.com> Message-ID: Jeffrey Altman wrote: > Christopher D. Clausen wrote: > >> Apply the no-fsync patch > > There is no patch to apply. Just update to 1.4.5. Hmm... Jason's previous email seemed to indicate that he was already running 1.4.5. I guess we need to know how many files are in the volume and how long the clone actually takes. B/c things should get considerably faster with the included no-fsync patch in 1.4.5. < (Jason Edgecombe's message of "Fri\, 30 Nov 2007 15\:53\:47 -0500") References: <4750785B.5010102@rampaginggeek.com> Message-ID: <87ir3jmmcx.fsf@windlord.stanford.edu> Jason Edgecombe writes: > When running a vos move, deleting a clone or the source volume seems to > take almost as much time as the copying of the data itself. Deleting the source volume requires dropping all the callbacks, right? That was one of my guesses as to why this sometimes takes a while. Doesn't really explain the clone part, though. -- Russ Allbery (rra@stanford.edu) From daleg@umbc.edu Fri Nov 30 22:36:51 2007 From: daleg@umbc.edu (Dale Ghent) Date: Fri, 30 Nov 2007 17:36:51 -0500 Subject: [OpenAFS] Which storage technology to use for terabytes of storage with AFS? In-Reply-To: <13639.1196454429@malison.ait.iastate.edu> References: <13639.1196454429@malison.ait.iastate.edu> Message-ID: On Nov 30, 2007, at 3:27 PM, John Hascall wrote: > > Stephen Joyce replies >> I don't have money for FC or a SAN, so I've stuck with DAS. I've >> had good >> experience with building many smallish servers rather than one big or >> expensive one. > > DAS is our approach as well. We are using "white boxes" built from > a hot-swapable Chenbro case with 3ware's 12 port SATA card in a > raid-10 config. > We're getting down pretty close to about $1K/TB. I imagine your Apple edu discount is similar to ours. We get the Xserve RAIDs for the same cost/space ratio. And those can be DAS or SAN'd. /dale -- Dale Ghent Specialist, Storage and UNIX Systems UMBC - Office of Information Technology ECS 201 - x51705