writes:
> I'm trying to get reauth to work so i can drop /var/spool/mail in afs... is
> this sane or is there some consideration i did not take into account? (of
Dont do that. Done that, been there.
Consistency of the files is only guaranteed upon the closure of a
file, thus opening a wide range of race conditions. The locking by
sendmail seemed not to work on AFS. After some troubles, I ended up
installing an imap server, serving out mails from local disk.
The UW-imap is rather trivial to install; you can even use your krb5
tickets to avoid retyping your password if the mail client supports
this (not all do so, some not out of the box).
BTW I do have a reauth.pl perl script, but that is rather for the
interactive start of long-running jobs. What you are probably looking
for is the kinit -k option.
Yours,
--
Martin Schulz schulz@iwrmm.math.uni-karlsruhe.de
Uni Karlsruhe, Institut f. wissenschaftliches Rechnen u. math. Modellbildung
Engesser Str. 6, 76128 Karlsruhe
From mike@bizittech.com Tue Oct 1 16:11:57 2002
From: mike@bizittech.com (mike@bizittech.com)
Date: Tue, 1 Oct 2002 11:11:57 -0400
Subject: [OpenAFS] Configuring top level of the afs Filespace
Message-ID: <001201c2695c$e31d2140$01000001@micron>
This is a multi-part message in MIME format.
------=_NextPart_000_000F_01C2693B.5B668EB0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi
I am trying to install following Quick Beginnings for UNIX on redhat 7.3 =
system.
A server only setup .=20
[root@rh73xfs root]# fs setacl /afs system:anyuser rl
fs: Invalid argument; it is possible that /afs is not in AFS.
bos status `hostname`
bos: a pioctl failed (getting tickets)
bos: running unauthenticated
Instance kaserver, currently running normally.
Instance buserver, currently running normally.
Instance ptserver, currently running normally.
Instance vlserver, currently running normally.
Instance fs, currently running normally.
Auxiliary status is: file server running.
Instance upserver, currently running normally.
lsmod
Module Size Used by Tainted: PF
libafs-2.4.18-4SGI_XFS_1.1-i686 444576 0 (unused)
autofs 12164 0 (autoclean) (unused)
3c59x 28488 1
ext3 67104 1 (autoclean)
jbd 49320 1 (autoclean) [ext3]
usb-uhci 24452 0 (unused)
usbcore 73184 1 [usb-uhci]
/etc/sysconfig/afs =20
AFS_CLIENT=3Doff
AFS_SERVER=3Don
------=_NextPart_000_000F_01C2693B.5B668EB0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi
I am trying to install following Quick =
Beginnings=20
for UNIX on redhat 7.3 system.
A server only setup .
[root@rh73xfs root]# fs setacl /afs =
system:anyuser=20
rl
fs: Invalid argument; it is possible that /afs is not in=20
AFS.
bos status `hostname`
bos: a pioctl =
failed=20
(getting tickets)
bos: running unauthenticated
Instance kaserver,=20
currently running normally.
Instance buserver, currently running=20
normally.
Instance ptserver, currently running normally.
Instance=20
vlserver, currently running normally.
Instance fs, currently running=20
normally.
Auxiliary status is: file server=20
running.
Instance upserver, currently running =
normally.
lsmod
Module &=
nbsp; =20
Size Used by Tainted:=20
PF
libafs-2.4.18-4SGI_XFS_1.1-i686 444576 0 =20
(unused)
autofs &=
nbsp; =20
12164 0 (autoclean)=20
(unused)
3c59x &n=
bsp; =20
28488 =20
1
ext3 &nbs=
p; =20
67104 1 =20
(autoclean)
jbd &=
nbsp; =20
49320 1 (autoclean)=20
[ext3]
usb-uhci &=
nbsp; =20
24452 0 =20
(unused)
usbcore =
=20
73184 1 [usb-uhci]
/etc/sysconfig/afs =
AFS_CLIENT=3Doff
AFS_SERVER=3Don
------=_NextPart_000_000F_01C2693B.5B668EB0--
From cg2v@andrew.cmu.edu Tue Oct 1 16:30:05 2002
From: cg2v@andrew.cmu.edu (Chaskiel M Grundman)
Date: Tue, 01 Oct 2002 11:30:05 -0400
Subject: [OpenAFS] OpenAFS on RedHat 8.0
In-Reply-To: <3D99B429.2020005@fnal.gov>
References: <3D99B429.2020005@fnal.gov>
Message-ID: <72570000.1033486205@endicott>
--On Tuesday, October 01, 2002 09:41:45 -0500 Troy Dawson
wrote:
> /usr/vice/etc/modload/libafs-2.4.18-14-i686.mp.o: unresolved symbol
> sys_call_table
This is because some smart kernel developer decided that it is Evil for
modules to manipulate (and in most cases use) the system call entry points,
and decided to forbid it by removing the functionality. Various
openafs-oriented people had been in touch with kernel developers to see if
afs's needs could be met. I thought that something had been worked out....,
but at first glance, I don't see a syscall registration function in any of
the RH 8 kernel patches.
In any case, I think I this can be hacked around in redhat kernels by using
kallsyms_symbol_to_address, but it will take me a few days before I'll be
able to get to it (having to install a machine and all).
From Mitchell.D.Baker@rose-hulman.edu Tue Oct 1 17:03:22 2002
From: Mitchell.D.Baker@rose-hulman.edu (Mitchell D. Baker)
Date: 01 Oct 2002 11:03:22 -0500
Subject: [OpenAFS] Return codes..
Message-ID: <1033488202.13868.20.camel@babylon5.rose-hulman.edu>
Looking for what some of the return codes are that are being logged but
the Fileserver process... Where in the source tree would I look or some
documentation? Like
SRXAFS_FetchData
SAFS_FetchStatus
Like we will get a returns 13 for FetchStatus.. just want to be able to
look things up...
Thanks all
See-ya
Mitch
--
/####################################################################/
/# Mitchell "Buzz" Baker "To Infinity And Beyond..." #/
/# Sr. Systems/Security Admin Rose-Hulman Institute of Technology #/
/# Mitchell.D.Baker@rose-hulman.edu www.rose-hulman.edu #/
/# For PGP Public key, check out www.keyserver.net #/
/####################################################################/
From Todd_DeSantis@transarc.com Tue Oct 1 17:34:43 2002
From: Todd_DeSantis@transarc.com (Todd_DeSantis@transarc.com)
Date: Tue, 1 Oct 2002 12:34:43 -0400 (EDT)
Subject: [OpenAFS] Return codes..
In-Reply-To: <1033488202.13868.20.camel@babylon5.rose-hulman.edu>
References: <1033488202.13868.20.camel@babylon5.rose-hulman.edu>
Message-ID: <4xaQuXE99g1T5RKU1Q@transarc.com>
Hi Buzz:
> Looking for what some of the return codes are that are being logged
> but the Fileserver process... Where in the source tree would I look
> or some documentation? Like
> SRXAFS_FetchData
> SAFS_FetchStatus
> Like we will get a returns 13 for FetchStatus.. just want to be able
> to look things up...
You can use the AFS command, trnaslate_et to translate the error text
[viced] translate_et 13
13 ().13 = Permission denied
And you can always search through the OS's /usr/include/sys/errno.h
for errors with a lower number.
The user trying to Fetch this file does not have permission to do so.
Todd DeSantis
AFS Support
IBM Pittsburgh Lab
From warlord@MIT.EDU Tue Oct 1 17:54:06 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 01 Oct 2002 12:54:06 -0400
Subject: [OpenAFS] Configuring top level of the afs Filespace
In-Reply-To: <001201c2695c$e31d2140$01000001@micron>
References: <001201c2695c$e31d2140$01000001@micron>
Message-ID:
writes:
> Hi
> I am trying to install following Quick Beginnings for UNIX on redhat 7.3 system.
> A server only setup .
>
> [root@rh73xfs root]# fs setacl /afs system:anyuser rl
> fs: Invalid argument; it is possible that /afs is not in AFS.
[snip]
> AFS_CLIENT=off
> AFS_SERVER=on
'fs' is a client application, and you have the client turned off.
Fix that and try again!
-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 jgarman@wedgie.org Tue Oct 1 18:44:38 2002
From: jgarman@wedgie.org (Jason Garman)
Date: Tue, 1 Oct 2002 13:44:38 -0400
Subject: [OpenAFS] Web-based file managers that play well with AFS...?
Message-ID: <20021001134438.A35728@got.wedgie.org>
Sigh...
Does anyone know of a web-based "file manager" that works with AFS?
Unfortunately I've been tasked with finding some solution and so didn't
want to reinvent the wheel if someone else had already implemented a
solution.
Thanks
--
Jason Garman / jgarman@wedgie.org
From shadow@dementia.org Tue Oct 1 18:56:38 2002
From: shadow@dementia.org (Derrick J Brashear)
Date: Tue, 1 Oct 2002 13:56:38 -0400 (EDT)
Subject: [OpenAFS] Return codes..
In-Reply-To: <1033488202.13868.20.camel@babylon5.rose-hulman.edu>
Message-ID:
On 1 Oct 2002, Mitchell D. Baker wrote:
> Looking for what some of the return codes are that are being logged but
> the Fileserver process... Where in the source tree would I look or some
> documentation? Like
> SRXAFS_FetchData
> SAFS_FetchStatus
>
Todd's answer is probably what you want, but if you do care these
functions are in src/viced.
From shadow@dementia.org Tue Oct 1 19:00:59 2002
From: shadow@dementia.org (Derrick J Brashear)
Date: Tue, 1 Oct 2002 14:00:59 -0400 (EDT)
Subject: [OpenAFS] OpenAFS on RedHat 8.0
In-Reply-To: <72570000.1033486205@endicott>
Message-ID:
On Tue, 1 Oct 2002, Chaskiel M Grundman wrote:
> --On Tuesday, October 01, 2002 09:41:45 -0500 Troy Dawson
> wrote:
>
> > /usr/vice/etc/modload/libafs-2.4.18-14-i686.mp.o: unresolved symbol
> > sys_call_table
>
> This is because some smart kernel developer decided that it is Evil for
> modules to manipulate (and in most cases use) the system call entry points,
> and decided to forbid it by removing the functionality. Various
Across a minor version (of the kernel) update, too. "Yay".
From rtm@cert.org Tue Oct 1 19:35:44 2002
From: rtm@cert.org (Rudolph T Maceyko)
Date: Tue, 01 Oct 2002 14:35:44 -0400
Subject: [OpenAFS] OpenAFS on RedHat 8.0
In-Reply-To:
References:
Message-ID: <229240000.1033497344@vorst.blue.cert.org>
Actually...
--On Tuesday, October 01, 2002 14:00:59 -0400 Derrick J Brashear
wrote:
>> This is because some smart kernel developer decided that it is Evil
>> for modules to manipulate (and in most cases use) the system call
>> entry points, and decided to forbid it by removing the
>> functionality. Various
>
> Across a minor version (of the kernel) update, too. "Yay".
To Red Hat's "credit", they did not introduce this change in the 2.4.18
kernels for Red Hat 7.3. It was only for Red Hat 8.0 (and its betas),
which also uses 2.4.18 kernels that they stopped exporting
sys_call_list.
Another way to put it is that they made the change across "releases" of
the kernel (in RPM speak), which you would normally think of as a much
smaller interval than "versions"...
Rudy
From openafs-info@openafs.org Tue Oct 1 20:15:05 2002
From: openafs-info@openafs.org (Derek Atkins)
Date: 01 Oct 2002 15:15:05 -0400
Subject: [OpenAFS] Configuring top level of the afs Filespace
In-Reply-To: <001501c26977$690dec20$01000001@micron>
References: <001201c2695c$e31d2140$01000001@micron>
<001501c26977$690dec20$01000001@micron>
Message-ID:
hi,
Please CC all your responses to openafs-info..
writes:
> Hi
> I am trying to setup server only . is That possible without using the client
You need an AFS client in order to setup an AFS server. Once you have
the server set up you can turn off the client (and use it from other
machines). But you DO need access to _some_ AFS client in order to
set volume/directory ACLS, make mount points, etc.
-derek
> thanks
> ----- Original Message -----
> From: "Derek Atkins"
> To:
> Cc:
> Sent: Tuesday, October 01, 2002 12:54 PM
> Subject: Re: [OpenAFS] Configuring top level of the afs Filespace
>
>
> > writes:
> >
> > > Hi
> > > I am trying to install following Quick Beginnings for UNIX on redhat 7.3
> system.
> > > A server only setup .
> > >
> > > [root@rh73xfs root]# fs setacl /afs system:anyuser rl
> > > fs: Invalid argument; it is possible that /afs is not in AFS.
> > [snip]
> > > AFS_CLIENT=off
> > > AFS_SERVER=on
> >
> > 'fs' is a client application, and you have the client turned off.
> >
> > Fix that and try again!
> >
> > -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
> >
>
--
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 abuechle@fhzh.ch Wed Oct 2 09:02:01 2002
From: abuechle@fhzh.ch (Andreas Buechler)
Date: Wed, 2 Oct 2002 10:02:01 +0200
Subject: [OpenAFS] pam and openafs 1.2.7 for RH 7.2
Message-ID: <15770.43001.206989.710003@hszpc38.isz.ch>
Hello,
I just installed openafs 1.2.7 on a alpha machine. Everything worked fine
(rebuilding and installing the rpm's) and at the end I was told to change
the files cacheinfo and ThisCell. I changed both files, now I am able to
get tokens etc as root for any afs-user. To be able to login and get a
token automatically I changed /etc/pam.d/system-auth as discribed at the end of the
installation.
Does anybody have an idea why I still cant login via ssh as an afs-user?
I posted my sshd and system-auth pam-files at the end of this mail.
Thanks for any help and sorry if this message was posted twice!
Andi
##### sshd #####
#%PAM-1.0
auth required /lib/security/pam_stack.so service=system-auth
auth required /lib/security/pam_nologin.so
account required /lib/security/pam_stack.so service=system-auth
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_limits.so
session optional /lib/security/pam_console.so
##### end sshd #####
#### system-auth #####
#%PAM-1.0
# This file is auto-generated.
# User changes will be destroyed the next time authconfig is run.
auth required /lib/security/pam_env.so
auth sufficient /lib/security/pam_unix.so likeauth nullok
auth required /lib/security/pam_deny.so
auth sufficient /lib/security/pam_afs.so try_first_pass ignore_root
account required /lib/security/pam_unix.so
password required /lib/security/pam_cracklib.so retry=3 type=
password sufficient /lib/security/pam_unix.so nullok use_authtok md5 shado
w
password required /lib/security/pam_deny.so
session required /lib/security/pam_limits.so
session required /lib/security/pam_unix.so
##### end system-auth #####
--
From schulz@iwrmm.math.uni-karlsruhe.de Wed Oct 2 09:10:20 2002
From: schulz@iwrmm.math.uni-karlsruhe.de (Martin Schulz)
Date: 02 Oct 2002 10:10:20 +0200
Subject: [OpenAFS] reauth
In-Reply-To:
References: <200209301824.43668.rla9216@rit.edu>
Message-ID:
Well. I hate personal mail as response to the mailing list.. thought I
had put it to the list.. so here it goes again:
Lee Damon writes:
>
>
> >BTW I do have a reauth.pl perl script, but that is rather for the
> >interactive start of long-running jobs.
>
> I'd be interested in seeing this, if you're willing to share.
This perl script is nothing really great.
Well, I just added the following to my afs-krb5 document: (tough this
is not yet online due to some Webserver-rsync issue at the moment)
------------------------------------------------------------------
How do I start a long-running job?
AFS tokens are limited to a 8 hours lifetime by default, this is a
problem for long running jobs. Suppose a user want to run a job over
night on his workstation. He/She can use the little perl script
reauth.pl to make sure the program has valid tokens all the time. The
basic function is as follows: The script asks for the users password,
goes into background and then obtains new tickets and tokens on a
regularly basis. This is done by the use of kinit and aklog.
A more thourough description is found in the comments.
------------------------------------------------------------------
You will find reauth.pl under
http://www.mathematik.uni-karlsruhe.de/~iwrmm/Persons/Schulz/Unix/afs/reauth.pl
Sorry for the incorrect mime type (webserver misconfig). To read it,
use "view Source". Comments are welcome.
Yours,
--
Martin Schulz schulz@iwrmm.math.uni-karlsruhe.de
Uni Karlsruhe, Institut f. wissenschaftliches Rechnen u. math. Modellbildung
Engesser Str. 6, 76128 Karlsruhe
From security@xauth.net Wed Oct 2 09:51:14 2002
From: security@xauth.net (Charles Clancy)
Date: Wed, 2 Oct 2002 03:51:14 -0500 (CDT)
Subject: [OpenAFS] pam and openafs 1.2.7 for RH 7.2
In-Reply-To: <15770.43001.206989.710003@hszpc38.isz.ch>
Message-ID:
> Does anybody have an idea why I still cant login via ssh as an afs-user?
> I posted my sshd and system-auth pam-files at the end of this mail.
Your configuration:
> auth required /lib/security/pam_env.so
> auth sufficient /lib/security/pam_unix.so likeauth nullok
> auth required /lib/security/pam_deny.so
> auth sufficient /lib/security/pam_afs.so try_first_pass
> ignore_root
Once it gets down to pam_deny.so, your login has failed. It never tries
pam_afs.so.
Try the following:
auth required /lib/security/pam_env.so
auth sufficient /lib/security/pam_afs.so try_first_pass
ignore_root
auth required /lib/security/pam_unix.so likeauth nullok
auth required /lib/security/pam_deny.so
[ t charles clancy ]--[ tclancy@uiuc.edu ]--[ www.uiuc.edu/~tclancy ]
From rsc@semantics.ch Wed Oct 2 10:00:20 2002
From: rsc@semantics.ch (Reto Schneider)
Date: Wed, 2 Oct 2002 11:00:20 +0200 (MEST)
Subject: [OpenAFS] multiple network interfaces in AFS
Message-ID: <37544.194.196.100.18.1033549220.squirrel@www.semantics.ch>
Hi
does anybody know, how i can prevent an network interface
beeing used by AFS ? I have a public and a private network
interface - only the public interface is reachable by the clients.
In DFS, there was the RPC_UNSUPPORTED_NETIFS environment variable.
Is there anything similar in AFS ?
Thanks
Reto
--
__________________________
Reto Schneider
Partner
Semantics AG
Toggenburgerstrasse 156
9500 Wil
Tel. +41 71 929 46 04
Fax. +41 71 929 46 01
http://www.semantics.ch
From Derrick J Brashear Wed Oct 2 10:06:39 2002
From: Derrick J Brashear (Derrick J Brashear)
Date: Wed, 2 Oct 2002 05:06:39 -0400 (EDT)
Subject: [OpenAFS] multiple network interfaces in AFS
In-Reply-To: <37544.194.196.100.18.1033549220.squirrel@www.semantics.ch>
Message-ID:
On Wed, 2 Oct 2002, Reto Schneider wrote:
> Hi
>
> does anybody know, how i can prevent an network interface
> beeing used by AFS ? I have a public and a private network
> interface - only the public interface is reachable by the clients.
> In DFS, there was the RPC_UNSUPPORTED_NETIFS environment variable.
> Is there anything similar in AFS ?
/usr/vice/etc/NetInfo and/or /usr/vice/etc/NetRestrict can be used for
this.
Quoting from documentation:
If the NetInfo file exists when the Cache Manager initializes, the Cache
Managers uses its contents as the basis for a list of the machine's
interfaces. If the file does not exist, the Cache Manager instead uses the
network interfaces configured with the operating system. If the
NetRestrict file exists, the Cache Manager removes any addresses included
in it from the list it is compiling. It records the completed list in
kernel memory.
From abuechle@fhzh.ch Wed Oct 2 10:46:20 2002
From: abuechle@fhzh.ch (Andreas Buechler)
Date: Wed, 2 Oct 2002 11:46:20 +0200
Subject: [OpenAFS] pam and openafs 1.2.7 for RH 7.2
In-Reply-To:
References: <15770.43001.206989.710003@hszpc38.isz.ch>
Message-ID: <15770.49260.4762.561962@hszpc38.isz.ch>
After modifing /etc/pam.d/system-auth as you proposed I still have
troubles login in. In the /var/log/messages I get following entries
afterwards:
Oct 2 11:29:43 burner pam_afs[5701]: AFS Authentication failed for user testuser. password was incorrect
Oct 2 11:29:43 burner pam_afs[5702]: AFS Authentication failed for user testuser. password was incorrect
Oct 2 11:29:49 burner pam_afs[5703]: AFS Authentication failed for user testuser. password was incorrect
Oct 2 11:29:49 burner pam_afs[5704]: AFS Authentication failed for user testuser. password was incorrect
Oct 2 11:29:52 burner sshd(pam_unix)[5699]: 2 more authentication failures; logname= uid=0 euid=0 tty=NODEVssh user=testuser
I don't understand this because I am sure that I didn't misstype the
password (tried it several times...). Is there a way to find out that
this is really a pam related problem?
Andi
--
From jarausch@igpm.rwth-aachen.de Wed Oct 2 12:38:28 2002
From: jarausch@igpm.rwth-aachen.de (jarausch@igpm.rwth-aachen.de)
Date: Wed, 2 Oct 2002 13:38:28 +0200
Subject: [OpenAFS] client without a server
Message-ID: <200210021138.NAA92206@numa1.igpm.rwth-aachen.de>
Hi,
I would like to have the possibility to run a client even if the
afs server is not available.
Currently I get (lots of) messages
Lost contact of volume location server ...
and even worse, I cannot login (as non-root)
here is (the relevant part of) my file /etc/pam.conf
# The PAM configuration file for the `login' service
#
login auth requisite pam_securetty.so
login auth required pam_unix.so
login auth sufficient pam_afs.so debug try_first_pass ignore_root
login auth optional pam_group.so
login account requisite pam_time.so
login account required pam_unix.so
login password required pam_cracklib.so retry=3
login password required pam_unix.so shadow md5 use_authtok
login session required pam_unix.so
Any help is very much appreciated,
Helmut Jarausch
Lehrstuhl fuer Numerische Mathematik
Institute of Technology, RWTH Aachen
D 52056 Aachen, Germany
From kerberos@northsailor.de Wed Oct 2 13:18:56 2002
From: kerberos@northsailor.de (Klaas Hagemann)
Date: Wed, 2 Oct 2002 14:18:56 +0200
Subject: [OpenAFS] scripts for automatic installation + client cache questions
Message-ID: <003a01c26a0d$e1dd6fe0$2b03a8c0@mummert.priv>
Hello,
i am currently trying to write scripts for an automatic setup for
openafs-file-servers and dbms-servers.
Therefor i use the bos exec command quite a lot.
But when adding a new server, i have to restart all servers.
So i want a "bos exec host "/etc/init.d/afs restart"
But it ends up with a communication failure. Does someone have an idea for a
workaround for this?
Then i have got another question:
what are the advances for an own partition for the client on
/usr/vice/cache?
Currently, i do not have a seperate partition for it and it works good.
Thanks for your answers
Klaas
From excds@kth.se Wed Oct 2 13:23:47 2002
From: excds@kth.se (Daniel =?ISO-8859-1?Q?Sw=E4rd?=)
Date: 02 Oct 2002 14:23:47 +0200
Subject: [OpenAFS] Unnamed groups?
Message-ID: <1033561427.575.30.camel@hybris>
I've got openafs and Kerberos up and running. But something puzzles me:
The users I've created are members of the groups cdrom,audio,floppy but
when I'm logged in as one of the users and issue the command "groups" I
get two extra gid's without groupnames.
Is this something afs/Kerberos-related?
/Daniel
btw, I found an afs-wikiwikiweb before but I seem to have lost the URL.
Anyone who knows where it is?
From kerberos@northsailor.de Wed Oct 2 13:31:32 2002
From: kerberos@northsailor.de (Klaas Hagemann)
Date: Wed, 2 Oct 2002 14:31:32 +0200
Subject: Fw: [OpenAFS] client without a server
Message-ID: <006001c26a0f$a42d96a0$2b03a8c0@mummert.priv>
Hello (or Hallo ;-) ) Helmut,
you can start the afs-client without an afs-server as far as i can see it.
But feel free to correct my.
You just cannot use any functions of the afs-client.
The pam-afs Modul you are using uses these functions .
So you should change your /etc/pam.d/login like this:
login auth requisite pam_securetty.so
login auth sufficient pam_unix.so
login auth required pam_afs.so debug try_first_pass ignore_root
login auth optional pam_group.so
login account requisite pam_time.so
login account required pam_unix.so
login password required pam_cracklib.so retry=3
login password required pam_unix.so shadow md5 use_authtok
login session required pam_unix.so
So the login-session will be terminatet, if the pam_unix is sufficient.
To make this usefull, users should have a different password in afs then on
the unix-box (otherwise it won't get to pam_afs.so).
Klaas
----- Original Message -----
> From:
> To:
> Sent: Wednesday, October 02, 2002 1:38 PM
> Subject: [OpenAFS] client without a server
>
>
> > Hi,
> >
> > I would like to have the possibility to run a client even if the
> > afs server is not available.
> > Currently I get (lots of) messages
> > Lost contact of volume location server ...
> >
> > and even worse, I cannot login (as non-root)
> >
> > here is (the relevant part of) my file /etc/pam.conf
> >
> > # The PAM configuration file for the `login' service
> > #
> > login auth requisite pam_securetty.so
> > login auth required pam_unix.so
> > login auth sufficient pam_afs.so debug try_first_pass ignore_root
> > login auth optional pam_group.so
> > login account requisite pam_time.so
> > login account required pam_unix.so
> > login password required pam_cracklib.so retry=3
> > login password required pam_unix.so shadow md5 use_authtok
> > login session required pam_unix.so
> >
> > Any help is very much appreciated,
> >
> > Helmut Jarausch
> >
> > Lehrstuhl fuer Numerische Mathematik
> > Institute of Technology, RWTH Aachen
> > D 52056 Aachen, Germany
> >
> > _______________________________________________
> > OpenAFS-info mailing list
> > OpenAFS-info@openafs.org
> > https://lists.openafs.org/mailman/listinfo/openafs-info
>
From warlord@MIT.EDU Wed Oct 2 14:17:44 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 02 Oct 2002 09:17:44 -0400
Subject: [OpenAFS] scripts for automatic installation + client cache questions
In-Reply-To: <003a01c26a0d$e1dd6fe0$2b03a8c0@mummert.priv>
References: <003a01c26a0d$e1dd6fe0$2b03a8c0@mummert.priv>
Message-ID:
"Klaas Hagemann" writes:
> Hello,
>
> i am currently trying to write scripts for an automatic setup for
> openafs-file-servers and dbms-servers.
> Therefor i use the bos exec command quite a lot.
> But when adding a new server, i have to restart all servers.
> So i want a "bos exec host "/etc/init.d/afs restart"
> But it ends up with a communication failure. Does someone have an idea for a
> workaround for this?
No, because the bosserver get's killed before the command completes.
Is there any particular reason you don't use "bos restart -all -bosserver"?
> Then i have got another question:
> what are the advances for an own partition for the client on
> /usr/vice/cache?
> Currently, i do not have a seperate partition for it and it works good.
Basically, if your cache partition fills, AFS is screwed. The benefit
of having your own AFS cache partition is that if only AFS uses the
partition you are guaranteed that it will never fill to capacity. If
you share the cache partition, then you have to always worry about how
much space is left, because AFS will crash ungracefully in the face of
a full cache (or worse, it could destroy data).
> Thanks for your answers
> Klaas
-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 warlord@MIT.EDU Wed Oct 2 14:18:37 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 02 Oct 2002 09:18:37 -0400
Subject: [OpenAFS] Unnamed groups?
In-Reply-To: <1033561427.575.30.camel@hybris>
References: <1033561427.575.30.camel@hybris>
Message-ID:
Daniel Sw=E4rd writes:
> I've got openafs and Kerberos up and running. But something puzzles me:
> The users I've created are members of the groups cdrom,audio,floppy but
> when I'm logged in as one of the users and issue the command "groups" I
> get two extra gid's without groupnames.
>=20
> Is this something afs/Kerberos-related?
Yes. This is the PAG.
> /Daniel
>=20
> btw, I found an afs-wikiwikiweb before but I seem to have lost the URL.
> Anyone who knows where it is?
It's off www.openafs.org.
-derek
--=20
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 excds@kth.se Wed Oct 2 14:56:35 2002
From: excds@kth.se (Daniel =?ISO-8859-1?Q?Sw=E4rd?=)
Date: 02 Oct 2002 15:56:35 +0200
Subject: [OpenAFS] Unnamed groups?
In-Reply-To:
References: <1033561427.575.30.camel@hybris>
Message-ID: <1033566995.575.48.camel@hybris>
> > Is this something afs/Kerberos-related?
>
> Yes. This is the PAG.
Shouldn't that resolve to a name? That is, exist in /etc/group or
something similar?
> > btw, I found an afs-wikiwikiweb before but I seem to have lost the URL.
> > Anyone who knows where it is?
>
> It's off www.openafs.org.
Thanks.
/Daniel
From excds@kth.se Wed Oct 2 14:59:33 2002
From: excds@kth.se (Daniel =?ISO-8859-1?Q?Sw=E4rd?=)
Date: 02 Oct 2002 15:59:33 +0200
Subject: [OpenAFS] Unnamed groups?
In-Reply-To:
References: <1033561427.575.30.camel@hybris>
Message-ID: <1033567173.571.50.camel@hybris>
Don't mind my previous post... I just RTFM:ed a little bit more now...
;-))
/Daniel
From abuechle@hszpc30.isz.ch Tue Oct 1 15:34:59 2002
From: abuechle@hszpc30.isz.ch (Andreas Buechler)
Date: Tue, 1 Oct 2002 16:34:59 +0200 (CEST)
Subject: [OpenAFS] openafs 1.2.7 for RH 7.2
Message-ID:
Hello,
I just installed openafs 1.2.7 on a alpha machine. Everything worked fine
(rebuilding and installing the rpm's) and at the end I was told to change
the files cacheinfo and ThisCell. I changed both files, now I am able to
get tokens etc as root for any afs-user. To be able to login and get a
token automatically I changed /etc/pam.d/system-auth as discribed at the
end of the installation.
Does anybody have an idea why I still can't login via ssh as an afs-user?
I posted my sshd and system-auth pam-files at the end of this mail.
Thanks for any help!
Andi
#### system-auth ####
#%PAM-1.0
# This file is auto-generated.
# User changes will be destroyed the next time authconfig is run.
auth required /lib/security/pam_env.so
auth sufficient /lib/security/pam_unix.so likeauth nullok
auth required /lib/security/pam_deny.so
auth sufficient /lib/security/pam_afs.so try_first_pass
ignore_root
account required /lib/security/pam_unix.so
password required /lib/security/pam_cracklib.so retry=3 type=
password sufficient /lib/security/pam_unix.so nullok use_authtok md5
shadow
password required /lib/security/pam_deny.so
session required /lib/security/pam_limits.so
session required /lib/security/pam_unix.so
#### end system-auth ####
#### sshd ####
#%PAM-1.0
auth required /lib/security/pam_stack.so service=system-auth
auth required /lib/security/pam_nologin.so
account required /lib/security/pam_stack.so service=system-auth
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_limits.so
session optional /lib/security/pam_console.so
#### end sshd ####
From mcintyre@ucsc.edu Tue Oct 1 19:21:48 2002
From: mcintyre@ucsc.edu (Charles McIntyre)
Date: Tue, 01 Oct 2002 11:21:48 -0700
Subject: [OpenAFS] file server preferences disappearing intermittently
Message-ID: <5.1.0.14.2.20021001111741.032ef338@cats-po-1.ucsc.edu>
Has anyone seen this? I'm using OpenAFS 1.2.2b for Windows on a W2K=20
workstation. Intermittently, I won't be able to get tokens. The error=20
says that the afsd service cannot be contacted. I check the Preferences in=
=20
the AFS Client Configuration control panel and there aren't any file server=
=20
or volume locator servers specified. If I restart the service or restart=20
the computer, they reappear. I'm also wondering if there's a way to=20
restart the service as a non-administrator.
Thanks!
Charles
=BA=B0`=B0=BA=A4=F8=A4=BA=B0`=B0=BA=A4=F8=F8=A4=BA=B0`=B0=BA=A4=F8=A4=BA=B0`=
=B0=BA=A4=F8=F8=A4=BA=B0`=B0=BA=A4
Charles McIntyre
PC/UNIX Support Engineer
Instructional Computing, UCSC
ph: 831/459-5746
got a question? see http://ic.ucsc.edu/help
From klaas@northsailor.de Wed Oct 2 13:29:51 2002
From: klaas@northsailor.de (klaas hagemann)
Date: Wed, 2 Oct 2002 14:29:51 +0200
Subject: [OpenAFS] client without a server
References: <200210021138.NAA92206@numa1.igpm.rwth-aachen.de>
Message-ID: <005101c26a0f$68a82960$2b03a8c0@mummert.priv>
Hello (or Hallo ;-) ) Helmut,
you can start the afs-client without an afs-server as far as i can see it.
But feel free to correct my.
You just cannot use any functions of the afs-client.
The pam-afs Modul you are using uses these functions .
So you should change your /etc/pam.d/login like this:
login auth requisite pam_securetty.so
login auth sufficient pam_unix.so
login auth required pam_afs.so debug try_first_pass ignore_root
login auth optional pam_group.so
login account requisite pam_time.so
login account required pam_unix.so
login password required pam_cracklib.so retry=3
login password required pam_unix.so shadow md5 use_authtok
login session required pam_unix.so
So the login-session will be terminatet, if the pam_unix is sufficient.
To make this usefull, users should have a different password in afs then on
the unix-box (otherwise it won't get to pam_afs.so).
Klaas
----- Original Message -----
From:
To:
Sent: Wednesday, October 02, 2002 1:38 PM
Subject: [OpenAFS] client without a server
> Hi,
>
> I would like to have the possibility to run a client even if the
> afs server is not available.
> Currently I get (lots of) messages
> Lost contact of volume location server ...
>
> and even worse, I cannot login (as non-root)
>
> here is (the relevant part of) my file /etc/pam.conf
>
> # The PAM configuration file for the `login' service
> #
> login auth requisite pam_securetty.so
> login auth required pam_unix.so
> login auth sufficient pam_afs.so debug try_first_pass ignore_root
> login auth optional pam_group.so
> login account requisite pam_time.so
> login account required pam_unix.so
> login password required pam_cracklib.so retry=3
> login password required pam_unix.so shadow md5 use_authtok
> login session required pam_unix.so
>
> Any help is very much appreciated,
>
> Helmut Jarausch
>
> Lehrstuhl fuer Numerische Mathematik
> Institute of Technology, RWTH Aachen
> D 52056 Aachen, Germany
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
From ian@assv.net Wed Oct 2 15:31:46 2002
From: ian@assv.net (Ian Delahorne)
Date: 02 Oct 2002 16:31:46 +0200
Subject: [OpenAFS] pam and openafs 1.2.7 for RH 7.2
In-Reply-To: <15770.49260.4762.561962@hszpc38.isz.ch>
References: <15770.43001.206989.710003@hszpc38.isz.ch>
<15770.49260.4762.561962@hszpc38.isz.ch>
Message-ID:
Andreas Buechler writes:
> After modifing /etc/pam.d/system-auth as you proposed I still have
> troubles login in. In the /var/log/messages I get following entries
> afterwards:
>
> Oct 2 11:29:43 burner pam_afs[5701]: AFS Authentication failed for user testuser. password was incorrect
> Oct 2 11:29:43 burner pam_afs[5702]: AFS Authentication failed for user testuser. password was incorrect
> Oct 2 11:29:49 burner pam_afs[5703]: AFS Authentication failed for user testuser. password was incorrect
> Oct 2 11:29:49 burner pam_afs[5704]: AFS Authentication failed for user testuser. password was incorrect
> Oct 2 11:29:52 burner sshd(pam_unix)[5699]: 2 more authentication failures; logname= uid=0 euid=0 tty=NODEVssh user=testuser
>
> I don't understand this because I am sure that I didn't misstype the
> password (tried it several times...). Is there a way to find out that
> this is really a pam related problem?
Try klog:ing as the user, that should tell you if AFS thinks you have
the right password or not.
--
/Ian D
ian@assv.net - www.assv.net
From warlord@MIT.EDU Wed Oct 2 15:38:33 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 02 Oct 2002 10:38:33 -0400
Subject: [OpenAFS] scripts for automatic installation + client cache questions
In-Reply-To: <007301c26a20$c6774c40$2b03a8c0@mummert.priv>
References: <003a01c26a0d$e1dd6fe0$2b03a8c0@mummert.priv>
<007301c26a20$c6774c40$2b03a8c0@mummert.priv>
Message-ID:
"klaas hagemann" writes:
> > Basically, if your cache partition fills, AFS is screwed. The benefit
> > of having your own AFS cache partition is that if only AFS uses the
> > partition you are guaranteed that it will never fill to capacity. If
> > you share the cache partition, then you have to always worry about how
> > much space is left, because AFS will crash ungracefully in the face of
> > a full cache (or worse, it could destroy data).
>
> Ok, i thought of something like that. But there is no performance increase?
There may be a small performance increase due to a smaller iget()
search area... I would always recommend a standalone cache partition,
mostly for stability, but just in general.
> Thanks, Klaas
-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 kerberos@northsailor.de Wed Oct 2 16:19:51 2002
From: kerberos@northsailor.de (Klaas Hagemann)
Date: Wed, 2 Oct 2002 17:19:51 +0200
Subject: [OpenAFS] client without a server
References: <200210021138.NAA92206@numa1.igpm.rwth-aachen.de> <005101c26a0f$68a82960$2b03a8c0@mummert.priv>
Message-ID: <007a01c26a27$279c03c0$2b03a8c0@mummert.priv>
Hello,
i just recognized, that the afs-client will run into a long time-out, if no
afs-server is available.
Then it won't really start, although the libafs-module is loaded (i am using
suse linux).
BTW, does anyone has a solution for this problem? To my point of view the
afs client should come up although without an afs server and should work
automatically as soon as an afs-dbms is available.
Klaas
----- Original Message -----
From: "klaas hagemann"
To: ;
Sent: Wednesday, October 02, 2002 2:29 PM
Subject: Re: [OpenAFS] client without a server
> Hello (or Hallo ;-) ) Helmut,
>
> you can start the afs-client without an afs-server as far as i can see it.
> But feel free to correct my.
> You just cannot use any functions of the afs-client.
> The pam-afs Modul you are using uses these functions .
>
> So you should change your /etc/pam.d/login like this:
>
> login auth requisite pam_securetty.so
> login auth sufficient pam_unix.so
> login auth required pam_afs.so debug try_first_pass ignore_root
> login auth optional pam_group.so
> login account requisite pam_time.so
> login account required pam_unix.so
> login password required pam_cracklib.so retry=3
> login password required pam_unix.so shadow md5 use_authtok
> login session required pam_unix.so
>
> So the login-session will be terminatet, if the pam_unix is sufficient.
> To make this usefull, users should have a different password in afs then
on
> the unix-box (otherwise it won't get to pam_afs.so).
>
> Klaas
>
> ----- Original Message -----
> From:
> To:
> Sent: Wednesday, October 02, 2002 1:38 PM
> Subject: [OpenAFS] client without a server
>
>
> > Hi,
> >
> > I would like to have the possibility to run a client even if the
> > afs server is not available.
> > Currently I get (lots of) messages
> > Lost contact of volume location server ...
> >
> > and even worse, I cannot login (as non-root)
> >
> > here is (the relevant part of) my file /etc/pam.conf
> >
> > # The PAM configuration file for the `login' service
> > #
> > login auth requisite pam_securetty.so
> > login auth required pam_unix.so
> > login auth sufficient pam_afs.so debug try_first_pass ignore_root
> > login auth optional pam_group.so
> > login account requisite pam_time.so
> > login account required pam_unix.so
> > login password required pam_cracklib.so retry=3
> > login password required pam_unix.so shadow md5 use_authtok
> > login session required pam_unix.so
> >
> > Any help is very much appreciated,
> >
> > Helmut Jarausch
> >
> > Lehrstuhl fuer Numerische Mathematik
> > Institute of Technology, RWTH Aachen
> > D 52056 Aachen, Germany
> >
> > _______________________________________________
> > 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 rees@umich.edu Wed Oct 2 16:33:33 2002
From: rees@umich.edu (Jim Rees)
Date: Wed, 02 Oct 2002 11:33:33 -0400
Subject: [OpenAFS] client without a server
In-Reply-To: "Klaas Hagemann", Wed, 02 Oct 2002 17:19:51 +0200
Message-ID: <20021002153333.BAF6C207C1@citi.umich.edu>
How is the client supposed to know the server is unavailable without using a
timeout?
From security@xauth.net Wed Oct 2 16:42:23 2002
From: security@xauth.net (Charles Clancy)
Date: Wed, 2 Oct 2002 10:42:23 -0500 (CDT)
Subject: [OpenAFS] pam and openafs 1.2.7 for RH 7.2
In-Reply-To: <15770.49260.4762.561962@hszpc38.isz.ch>
Message-ID:
> After modifing /etc/pam.d/system-auth as you proposed I still have
> troubles login in. In the /var/log/messages I get following entries
> afterwards:
>
> Oct 2 11:29:43 burner pam_afs[5701]: AFS Authentication failed for user
> testuser. password was incorrect
Sorry -- there was a mistake in my previous post.
> auth required /lib/security/pam_env.so
> auth sufficient /lib/security/pam_afs.so try_first_pass
> ignore_root
> auth required /lib/security/pam_unix.so likeauth nullok
> auth required /lib/security/pam_deny.so
If you'll notice, there is no first pass to try (as pam_unix is called
after pam_afs), hence you should remove the try_first_pass option. This
shouldn't completely prevent you from logging in, but it will make the
entries you noticed show up in your logs.
Try removing that option, and see if it works then.
[ t charles clancy ]--[ tclancy@uiuc.edu ]--[ www.uiuc.edu/~tclancy ]
From warlord@MIT.EDU Wed Oct 2 16:49:25 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 02 Oct 2002 11:49:25 -0400
Subject: [OpenAFS] client without a server
In-Reply-To: <007a01c26a27$279c03c0$2b03a8c0@mummert.priv>
References: <200210021138.NAA92206@numa1.igpm.rwth-aachen.de>
<005101c26a0f$68a82960$2b03a8c0@mummert.priv>
<007a01c26a27$279c03c0$2b03a8c0@mummert.priv>
Message-ID:
Run afs with -dynroot
-derek
"Klaas Hagemann" writes:
> Hello,
>
> i just recognized, that the afs-client will run into a long time-out, if no
> afs-server is available.
> Then it won't really start, although the libafs-module is loaded (i am using
> suse linux).
>
> BTW, does anyone has a solution for this problem? To my point of view the
> afs client should come up although without an afs server and should work
> automatically as soon as an afs-dbms is available.
>
> Klaas
> ----- Original Message -----
> From: "klaas hagemann"
> To: ;
> Sent: Wednesday, October 02, 2002 2:29 PM
> Subject: Re: [OpenAFS] client without a server
>
>
> > Hello (or Hallo ;-) ) Helmut,
> >
> > you can start the afs-client without an afs-server as far as i can see it.
> > But feel free to correct my.
> > You just cannot use any functions of the afs-client.
> > The pam-afs Modul you are using uses these functions .
> >
> > So you should change your /etc/pam.d/login like this:
> >
> > login auth requisite pam_securetty.so
> > login auth sufficient pam_unix.so
> > login auth required pam_afs.so debug try_first_pass ignore_root
> > login auth optional pam_group.so
> > login account requisite pam_time.so
> > login account required pam_unix.so
> > login password required pam_cracklib.so retry=3
> > login password required pam_unix.so shadow md5 use_authtok
> > login session required pam_unix.so
> >
> > So the login-session will be terminatet, if the pam_unix is sufficient.
> > To make this usefull, users should have a different password in afs then
> on
> > the unix-box (otherwise it won't get to pam_afs.so).
> >
> > Klaas
> >
> > ----- Original Message -----
> > From:
> > To:
> > Sent: Wednesday, October 02, 2002 1:38 PM
> > Subject: [OpenAFS] client without a server
> >
> >
> > > Hi,
> > >
> > > I would like to have the possibility to run a client even if the
> > > afs server is not available.
> > > Currently I get (lots of) messages
> > > Lost contact of volume location server ...
> > >
> > > and even worse, I cannot login (as non-root)
> > >
> > > here is (the relevant part of) my file /etc/pam.conf
> > >
> > > # The PAM configuration file for the `login' service
> > > #
> > > login auth requisite pam_securetty.so
> > > login auth required pam_unix.so
> > > login auth sufficient pam_afs.so debug try_first_pass ignore_root
> > > login auth optional pam_group.so
> > > login account requisite pam_time.so
> > > login account required pam_unix.so
> > > login password required pam_cracklib.so retry=3
> > > login password required pam_unix.so shadow md5 use_authtok
> > > login session required pam_unix.so
> > >
> > > Any help is very much appreciated,
> > >
> > > Helmut Jarausch
> > >
> > > Lehrstuhl fuer Numerische Mathematik
> > > Institute of Technology, RWTH Aachen
> > > D 52056 Aachen, Germany
> > >
> > > _______________________________________________
> > > 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
>
> _______________________________________________
> 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 tino.schwarze@informatik.tu-chemnitz.de Wed Oct 2 16:50:24 2002
From: tino.schwarze@informatik.tu-chemnitz.de (Tino Schwarze)
Date: Wed, 2 Oct 2002 17:50:24 +0200
Subject: [OpenAFS] Web-based file managers that play well with AFS...?
In-Reply-To: <20021001134438.A35728@got.wedgie.org>; from jgarman@wedgie.org on Tue, Oct 01, 2002 at 01:44:38PM -0400
References: <20021001134438.A35728@got.wedgie.org>
Message-ID: <20021002175024.A21701@informatik.tu-chemnitz.de>
Hi there,
On Tue, Oct 01, 2002 at 01:44:38PM -0400, Jason Garman wrote:
> Does anyone know of a web-based "file manager" that works with AFS?
> Unfortunately I've been tasked with finding some solution and so didn't
> want to reinvent the wheel if someone else had already implemented a
> solution.
Here at Chemnitz University of Technology, a self-developed too called
"wfm" is used. It's available at
http://www-user.tu-chemnitz.de/~fri/wfm/
Unfortunately, it's documentation is German-only. I'm going to roll this
thing out myself and could probably translate it in the process.
You also need a modified mod_auth_pam for Apache which allows the server
to acquire tokens on behalf of the user. It works quite well and is
available at:
http://www.tu-chemnitz.de/urz/afs/openafs/download/AddOn/mod_auth_pam/
(or /afs/tu-chemnitz.de/openafs/AddOn/mod_auth_pam )
HTH! Tino.
PS: Rough translation of INSTALL:
1. Prerequisites
- Apache (with mod_ssl for security) and PHP3 or PHP4
- mod_auth_pam and /etc/pam.d/httpd:
auth sufficient /lib/security/pam_afs.so.1 ignore_root dont_fork
session required /lib/security/pam_afs.so.1
account required pam_pwdb.so
2. Installation
Modify the following files:
- htaccess
- config.php3, help, maybe index.php3
- copy everything into $DOCUMENT_ROOT/wfm
- cp htaccess $DOCUMENT_ROOT/wfm/.htaccess
3. Test
https://server-name/wfm
--
* LINUX - Where do you want to be tomorrow? *
http://www.tu-chemnitz.de/linux/tag/
From Mitchell.D.Baker@rose-hulman.edu Wed Oct 2 21:06:47 2002
From: Mitchell.D.Baker@rose-hulman.edu (Mitchell D. Baker)
Date: 02 Oct 2002 15:06:47 -0500
Subject: [OpenAFS] AFS_BulkStatus info...
Message-ID: <1033589207.28295.5.camel@babylon5.rose-hulman.edu>
On on the servers, we are just started to get several seconds at a time
the following messages:
ed Oct 2 15:00:40 2002 SAFS_BulkStatus returns 0
Wed Oct 2 15:00:40 2002 SAFS_BulkStatus
Wed Oct 2 15:00:40 2002 SAFS_BulkStatus returns 0
Wed Oct 2 15:00:40 2002 SAFS_BulkStatus
Wed Oct 2 15:00:40 2002 SAFS_BulkStatus returns 0
fileserver CPU time has also jumped up... We upgraded to 1.2.7 last
night to get over the callback issue..
What does the AFS_BulkStatus call do? Does that many in a short about
of time indicate a probem or potential problem? After that goes on for
several seconds, then I get the normal flow of FETCHDATA FETCHSTATUS etc
calls and the load and CPU usage for fileserver settles down..
Thanks
See-ya
Mitch
--
/####################################################################/
/# Mitchell "Buzz" Baker "To Infinity And Beyond..." #/
/# Sr. Systems/Security Admin Rose-Hulman Institute of Technology #/
/# Mitchell.D.Baker@rose-hulman.edu www.rose-hulman.edu #/
/# For PGP Public key, check out www.keyserver.net #/
/####################################################################/
From adler@bnl.gov Wed Oct 2 23:05:24 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 02 Oct 2002 18:05:24 -0400
Subject: [OpenAFS] redhat 8.0
Message-ID: <1033596324.1902.22.camel@newadler.phy.bnl.gov>
I've also gone through the exersize to build afs 1.2.7 on my red hat
8.0 system. I've attached the .spec and a patch file. But
as with the other posts, I get the unresolved sys_call_table symbol
error as well when I go to load the module. I'll be glad to
add an account on my system to someone from the openafs development
team who wishes to work on this problem.
Cheers. Steve.
From adler@bnl.gov Wed Oct 2 23:07:56 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 02 Oct 2002 18:07:56 -0400
Subject: [OpenAFS] forgot the attachements.. duhhhh...
Message-ID: <1033596476.1902.25.camel@newadler.phy.bnl.gov>
--=-OoaDeI6iwlHwgGVdTi0k
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
long day...
--=-OoaDeI6iwlHwgGVdTi0k
Content-Disposition: attachment; filename=openafs-1.2.7.spec
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; name=openafs-1.2.7.spec; charset=ISO-8859-1
%define afsvers 1.2.7
%define pkgrel 2
# Define your particular Red Hat and kernel versions:
# For Linux 2.2: 22
# For Linux 2.4: 24
#
%define osvers rh8.0
%define kernvers 24
# This is where to look for kernel-build includes files.
# Most likely you don't want to change this, but
# depending on your situation you may want:
# Linux 2.2:
# kbase =3D /usr/src/linux-
# kend =3D ""
# Linux 2.4:
# kbase =3D /lib/modules/
# kend =3D /build
#
%define kbase /usr/src/linux-
%define kend ""
# Set 'debugspec' to 1 if you want to debug the spec file. This will
# not remove the installed tree as part of the %clean operation
%define debugspec 0
# Set 'enterprisekernelsupport' to 1 if you want to build the
# kernel module for the enterprise kernel
# Note: This will only work for kernvers =3D=3D 24 on i686
%define enterprisekernelsupport 1
# Set 'bigmemkernelsupport' to 1 if you want to build the
# kernel module for the bigmem kernel
# Note: This will only work for kernvers =3D=3D 24 on i686
%define bigmemkernelsupport 1
# Set 'krb5support' to 1 if you want to build the openafs-krb5 package
# to distribute aklog and asetkey
%define krb5support 1
# OpenAFS configuration options
%define enable_bitmap_later 0
%define enable_bos_restricted_mode 0
%define enable_fast_restart 0
#######################################################################
# You probably don't need to change anything beyond this line
# NOTE: If you do, please email me!!!
Summary: OpenAFS distributed filesystem
Name: openafs
Version: %{afsvers}
Release: %{osvers}.%{pkgrel}
Copyright: IPL
BuildRoot: %{_tmppath}/%{name}-%{version}-root
Packager: Derek Atkins
Group: Networking/Filesystems
BuildRequires: kernel-source
%if "%{osvers}" !=3D "rh6.2"
# Newer versions of Red Hat require pam-devel in order to build
BuildRequires: pam-devel
%endif
Source0: =
http://www.openafs.org/dl/openafs/${afsvers}/openafs-%{afsvers}-src.tar.=
gz
Source1: =
http://www.openafs.org/dl/openafs/${afsvers}/openafs-%{afsvers}-doc.tar.=
gz
Source2: openafs-ThisCell
# http://grand.central.org/dl/cellservdb/CellServDB
Source3: openafs-CellServDB
Source4: openafs-SuidCells
Source5: openafs-cacheinfo
Source6: openafs-afsmodname
Source7: openafs-LICENSE.Sun
Source8: openafs-README
Source10: =
http://www.openafs.org/dl/openafs/${afsvers}/RELNOTES-%{afsvers}
Source11: http://www.openafs.org/dl/openafs/${afsvers}/ChangeLog
Source20: openafs-krb5-1.3.tar.gz
Patch0: openafs-%{afsvers}-rc.patch
Patch1: openafs-gcc-32.patch
Patch20: openafs-krb5-1.3-1.2.1.diff.gz
Patch21: openafs-krb5-1.3-configure.patch
%description
The AFS distributed filesystem. AFS is a distributed filesystem
allowing cross-platform sharing of files among multiple computers.
Facilities are provided for access control, authentication, backup and
administrative management.
This package provides common files shared across all the various
OpenAFS packages but are not necessarily tied to a client or server.
%package client
Requires: binutils, openafs-kernel, openafs =3D %{PACKAGE_VERSION}
Summary: OpenAFS Filesystem Client
Group: Networking/Filesystem
%description client
The AFS distributed filesystem. AFS is a distributed filesystem
allowing cross-platform sharing of files among multiple computers.
Facilities are provided for access control, authentication, backup and
administrative management.
This package provides basic client support to mount and manipulate
AFS.
%package server
Requires: openafs-kernel, openafs =3D %{PACKAGE_VERSION}
Summary: OpenAFS Filesystem Server
Group: Networking/Filesystems
%description server
The AFS distributed filesystem. AFS is a distributed filesystem
allowing cross-platform sharing of files among multiple computers.
Facilities are provided for access control, authentication, backup and
administrative management.
This package provides basic server support to host files in an AFS
Cell.
%package devel
Summary: OpenAFS Development Libraries and Headers
Group: Development/Filesystems
%description devel
The AFS distributed filesystem. AFS is a distributed filesystem
allowing cross-platform sharing of files among multiple computers.
Facilities are provided for access control, authentication, backup and
administrative management.
This package provides static development libraries and headers needed
to compile AFS applications. Note: AFS currently does not provide
shared libraries.
%package kernel
Summary: OpenAFS Kernel Module(s)
Requires: openafs =3D %{PACKAGE_VERSION}
Group: Networking/Filesystems
%description kernel
The AFS distributed filesystem. AFS is a distributed filesystem
allowing cross-platform sharing of files among multiple computers.
Facilities are provided for access control, authentication, backup and
administrative management.
This package provides precompiled AFS kernel modules for various
kernels.
%package kernel-source
Summary: OpenAFS Kernel Module source tree
Group: Networking/Filesystems
%description kernel-source
The AFS distributed filesystem. AFS is a distributed filesystem
allowing cross-platform sharing of files among multiple computers.
Facilities are provided for access control, authentication, backup and
administrative management.
This package provides the source code to build your own AFS kernel
module.
%package compat
Summary: OpenAFS client compatibility symlinks
Requires: openafs =3D %{PACKAGE_VERSION}, openafs-client =3D =
%{PACKAGE_VERSION}
Group: Networking/Filesystems
Obsoletes: openafs-client-compat
%description compat
The AFS distributed filesystem. AFS is a distributed filesystem
allowing cross-platform sharing of files among multiple computers.
Facilities are provided for access control, authentication, backup and
administrative management.
This package provides compatibility symlinks in /usr/afsws. It is
completely optional, and is only necessary to support legacy
applications and scripts that hard-code the location of AFS client
programs.
%package kpasswd
Summary: OpenAFS KA kpasswd support
Requires: openafs
Group: Networking/Filesystems
%description kpasswd
The AFS distributed filesystem. AFS is a distributed filesystem
allowing cross-platform sharing of files among multiple computers.
Facilities are provided for access control, authentication, backup and
administrative management.
This package provides the compatibility symlink for kpasswd, in case
you are using KAserver instead of Krb5.
%if %{krb5support}
%package krb5
Summary: OpenAFS programs to use with krb5
Requires: openafs =3D %{PACKAGE_VERSION}
Group: Networking/Filesystems
BuildRequires: krb5-devel
%description krb5
The AFS distributed filesystem. AFS is a distributed filesystem
allowing cross-platform sharing of files among multiple computers.
Facilities are provided for access control, authentication, backup and
administrative management.
This package provides compatibility programs so you can use krb5
to authenticate to AFS services, instead of using AFS's homegrown
krb4 lookalike services.
%endif
#
# PREP
#
%prep
%setup -q -b 1
%setup -q -T -D -a 20
%patch0 -p0
%patch1 -p0
%patch20 -p0
%patch21 -p0
###
### build
###
%build
### set an alias for autoconf to use autoconf-2.13.. I hope this
### works. Stephen Adler, Wed, Oct 02, 2002
alias autoconf=3Dautoconf-2.13
%ifarch i386 i486 i586 i686 athlon
sysbase=3Di386
%else
sysbase=3D%{_arch}
%endif
%ifarch alpha
sysname=3D${sysbase}_linux_%{kernvers}
%else
sysname=3D${sysbase}_linux%{kernvers}
%endif
if [ %{kernvers} =3D 22 ]; then
kv=3D'2\.2\.'
elif [ %{kernvers} =3D 24 ]; then
kv=3D'2\.4\.'
else
echo "I don't know how to build $sysname"
exit 1
fi
%ifarch i386 i486 i586 i686 athlon
archlist=3D"i386 i586 i686 athlon"
%else
archlist=3D${sysbase}
%endif
#
# PrintDefine var value statements file
#
PrintDefine() {
case $3 in
*ifn*)
echo "#ifndef $1" >> $4
;;
esac
case $3 in
*und*)
echo "#undef $1" >> $4
;;
esac
case $3 in
*def*)
echo "#define $1 $2" >> $4
;;
esac
case $3 in
*end*)
echo "#endif" >> $4
;;
esac
case $3 in
*inc*)
echo "#include $1" >> $4
;;
esac
case $3 in
*nl*)
echo "" >> $4
;;
esac
}
# PrintRedhatKernelFix arch mp file
PrintRedhatKernelFix() {
arch=3D"$1"
up=3D0
smp=3D0
ent=3D0
bm=3D0
if [ "$2" =3D "MP" ]; then
smp=3D1
elif [ "$2" =3D "EP" ]; then
ent=3D1
elif [ "$2" =3D "BM" ]; then
bm=3D1
else
up=3D1
fi
file=3D"$3"
# deal with the various boot kernels
boot=3D0
bootsmp=3D0
# arch of 'BOOT' =3D=3D 386
if [ "$arch" =3D "BOOT" ]; then
if [ "$up" =3D 1 ]; then
boot=3D1
up=3D0
elif [ "$smp" =3D 1 ]; then
bootsmp=3D1
smp=3D0
fi
arch=3Di386
fi
rm -f $file
touch $file
PrintDefine "REDHAT_FIX_H" "" ifn,def,nl $file
PrintDefine "__BOOT_KERNEL_ENTERPRISE" $ent und,def,nl $file
PrintDefine "__BOOT_KERNEL_BIGMEM" $bm und,def,nl $file
PrintDefine "__BOOT_KERNEL_SMP" $smp und,def,nl $file
PrintDefine "__BOOT_KERNEL_UP" $up und,def,nl $file
PrintDefine "__BOOT_KERNEL_BOOT" $boot und,def,nl $file
PrintDefine "__BOOT_KERNEL_BOOTSMP" $bootsmp und,def,nl $file
PrintDefine \"/boot/kernel.h\" "" inc,nl $file # include file
for ar in $archlist ; do
if [ "$ar" =3D "$arch" ]; then
PrintDefine "__MODULE_KERNEL_$ar" "1" ifn,def,end $file
else
PrintDefine "__MODULE_KERNEL_$ar" "" und $file # undef
fi
done
echo "" >> $file
PrintDefine "" "" end $file
if [ %{debugspec} =3D 1 ] ; then
echo "Kernel Configuration File for Red Hat kernels:"
cat $file
fi
}
# Pick up all the 'appropriate' kernels
kvers=3D`ls -d %{kbase}* | sed 's^%{kbase}^^g' | grep $kv`
# Choose the last one for now.. It doesn't really matter, really.
hdrdir=3D`ls -d %{kbase}*%{kend} | grep $kv | tail -1`
config_opts=3D"--enable-redhat-buildsys \
%if %{enable_bitmap_later}
--enable-bitmap-later \
%endif
%if %{enable_bos_restricted_mode}
--enable-bos-restricted-mode \
%endif
%if %{enable_fast_restart}
--enable-fast-restart \
%endif
--enable-transarc-paths"
# Configure AFS
./configure --with-afs-sysname=3D${sysname} \
--with-linux-kernel-headers=3D$hdrdir $config_opts
# Build the user-space AFS stuff
make dest_nolibafs
# Build the libafs tree
make only_libafs_tree
# Now build all the kernel modules
for vers in $kvers ; do
# Reconfigure sources for this kernel version, to catch various
# kernel params in the configure script. Yes. this takes more time,
# but it's worth it in the long run.. But first remove config.cache
# to be sure we get a clean configuration.
rm -f config.cache
./configure --with-afs-sysname=3D${sysname} \
--with-linux-kernel-headers=3D%{kbase}$vers%{kend} \
$config_opts
KTL=3D"SP MP"
%if %{enterprisekernelsupport}
# See if we should build EP support
if grep -q -r __BOOT_KERNEL_ENTERPRISE %{kbase}$vers%{kend}/include
then
KTL=3D"${KTL} EP"
fi
%endif
%if %{bigmemkernelsupport}
# See if we should build BM support
if grep -q -r __BOOT_KERNEL_BIGMEM %{kbase}$vers%{kend}/include
then
KTL=3D"${KTL} BM"
fi
%endif
=20
for mp in $KTL; do
# ... for all appropriate 'architectures'...
if [ %{kernvers} =3D 22 ]; then
# For 2.2 kernels, just do MP and SP kernels; force EP into i686
arch=3D${sysbase}
if [ $mp =3D EP -a ${sysbase} =3D i386 ]; then
arch=3Di686
fi
PrintRedhatKernelFix $arch $mp src/config/redhat-fix.h
make dest_only_libafs LOCAL_SMP_DEF=3D-DREDHAT_FIX MPS=3D$mp
elif [ %{kernvers} =3D 24 ]; then
# For 2.4 kernels, need to build modules for each architecture!
for arch in $archlist ; do
# build SP and MP on all architectures.
# build EP and BM only on i686
if [ $mp =3D SP -o $mp =3D MP -o \
\( $mp =3D EP -a $arch =3D i686 \) -o \
\( $mp =3D BM -a $arch =3D i686 \) ]; then
PrintRedhatKernelFix $arch $mp src/config/redhat-fix.h
make dest_only_libafs LOCAL_SMP_DEF=3D-DREDHAT_FIX \
LINUX_MODULE_NAME=3D"-$arch" MPS=3D$mp
fi
done
else =20
echo "I don't know how to build $sysname"
exit 1
fi
done
done
rm -f src/config/redhat-fix.h
%if %{krb5support}
# Now build aklog/asetkey
(cd openafs-krb5-1.3/src &&
autoconf &&
./configure --prefix=3D/usr --with-krb5=3D/usr/kerberos \
--with-afs=3D`pwd`/../../${sysname}/dest/ && \
make all && \
make install DESTDIR=3D`pwd`/../../${sysname}/dest/ INSTALL_BIN=3D/bin =
\
INSTALL_SBIN=3D/etc)
%endif
###
### install
###
%install
[ $RPM_BUILD_ROOT !=3D / ] && rm -rf $RPM_BUILD_ROOT
%ifarch i386 i486 i586 i686 athlon
sysbase=3Di386
%else
sysbase=3D%{_arch}
%endif
%ifarch alpha
sysname=3D${sysbase}_linux_%{kernvers}
%else
sysname=3D${sysbase}_linux%{kernvers}
%endif
# Build install tree
mkdir -p $RPM_BUILD_ROOT/usr/sbin
mkdir -p $RPM_BUILD_ROOT/etc/sysconfig
mkdir -p $RPM_BUILD_ROOT/etc/rc.d/init.d
mkdir -p $RPM_BUILD_ROOT/etc/openafs
mkdir -p $RPM_BUILD_ROOT/lib/security
mkdir -p $RPM_BUILD_ROOT/usr/afs/logs
mkdir -p $RPM_BUILD_ROOT/usr/vice/etc
mkdir -p $RPM_BUILD_ROOT/usr/vice/cache
chmod 700 $RPM_BUILD_ROOT/usr/vice/cache
# Copy files from dest to the appropriate places in BuildRoot
tar cf - -C ${sysname}/dest bin include lib | tar xf - -C =
$RPM_BUILD_ROOT/usr
tar cf - -C ${sysname}/dest/etc . | tar xf - -C =
$RPM_BUILD_ROOT/usr/sbin
tar cf - -C ${sysname}/dest/root.server/usr/afs bin | tar xf - -C =
$RPM_BUILD_ROOT/usr/afs
tar cf - -C ${sysname}/dest/root.client/usr/vice/etc afsd modload | tar =
xf - -C $RPM_BUILD_ROOT/usr/vice/etc
# Link kpasswd to kapasswd
ln -f $RPM_BUILD_ROOT/usr/bin/kpasswd $RPM_BUILD_ROOT/usr/bin/kapasswd
# Copy root.client config files
install -m 755 ${sysname}/dest/root.client/usr/vice/etc/afs.conf =
$RPM_BUILD_ROOT/etc/sysconfig/afs
install -m 755 ${sysname}/dest/root.client/usr/vice/etc/afs.rc =
$RPM_BUILD_ROOT/etc/rc.d/init.d/afs
# Copy PAM modules
install -m 755 ${sysname}/dest/lib/pam* $RPM_BUILD_ROOT/lib/security
# PAM symlinks
ln -sf pam_afs.so.1 $RPM_BUILD_ROOT/lib/security/pam_afs.so
ln -sf pam_afs.krb.so.1 $RPM_BUILD_ROOT/lib/security/pam_afs.krb.so
# Populate /usr/vice/etc
uve=3D$RPM_BUILD_ROOT/usr/vice/etc
install -p -m 644 $RPM_SOURCE_DIR/openafs-CellServDB $uve/CellServDB
install -p -m 644 $RPM_SOURCE_DIR/openafs-SuidCells $uve/SuidCells
install -p -m 644 $RPM_SOURCE_DIR/openafs-ThisCell $uve/ThisCell
install -p -m 644 $RPM_SOURCE_DIR/openafs-cacheinfo $uve/cacheinfo
install -p -m 755 $RPM_SOURCE_DIR/openafs-afsmodname $uve/afsmodname
#
# Build the SymTable
symtable=3D$RPM_BUILD_ROOT/usr/vice/etc/modload/SymTable
rm -f $symtable
echo "# SymTable, automatically generated" > $symtable
echo "# symbol version cpu module" >> $symtable
echo "" >> $symtable
$RPM_BUILD_ROOT/usr/vice/etc/afsmodname -x -f $symtable \
$RPM_BUILD_ROOT/usr/vice/etc/modload/libafs*.o
#
# install kernel-source
#
# Install the kernel module source tree
mkdir -p $RPM_BUILD_ROOT/usr/src/openafs-kernel-%{afsvers}/src
tar cf - -C libafs_tree . | \
tar xf - -C $RPM_BUILD_ROOT/usr/src/openafs-kernel-%{afsvers}/src
# Next, copy the LICENSE Files, README
install -m 644 src/LICENSE =
$RPM_BUILD_ROOT/usr/src/openafs-kernel-%{afsvers}/LICENSE.IBM
install -m 644 $RPM_SOURCE_DIR/openafs-LICENSE.Sun =
$RPM_BUILD_ROOT/usr/src/openafs-kernel-%{afsvers}/LICENSE.Sun
install -m 644 $RPM_SOURCE_DIR/openafs-README =
$RPM_BUILD_ROOT/usr/src/openafs-kernel-%{afsvers}/README
#
# Install DOCUMENTATION
#
# Build the DOC directory
mkdir -p $RPM_BUILD_ROOT/$RPM_DOC_DIR/openafs-%{afsvers}
tar cf - -C doc LICENSE html pdf | \
tar xf - -C $RPM_BUILD_ROOT/$RPM_DOC_DIR/openafs-%{afsvers}
install -m 644 $RPM_SOURCE_DIR/RELNOTES-%{afsvers} =
$RPM_BUILD_ROOT/$RPM_DOC_DIR/openafs-%{afsvers}
install -m 644 $RPM_SOURCE_DIR/ChangeLog =
$RPM_BUILD_ROOT/$RPM_DOC_DIR/openafs-%{afsvers}
#
# create filelist
#
grep -v "^#" >openafs-file-list <
CCFLAGS =3D $(KDEBUG) -O2 $(FOMIT) \
- -fno-strength-reduce -pipe -march=3Di486 -malign-loops=3D2 =
-malign-jumps=3D2 \
- -malign-functions=3D2
+ -fno-strength-reduce -pipe -march=3Di486 -falign-loops=3D2 =
-falign-jumps=3D2 \
+ -falign-functions=3D2
DEFINES =3D -D__KERNEL__ -DCPU=3D586 -DKERNEL -D_KERNEL -DMODULE =
${SMP_DEF} ${KDEFINES}
CCFLAGS =3D $(KDEBUG) -O2 $(FOMIT) -fno-strength-reduce -pipe =
-mno-fp-regs -ffixed-8
--=-OoaDeI6iwlHwgGVdTi0k--
From nneul@umr.edu Wed Oct 2 23:23:07 2002
From: nneul@umr.edu (Nathan Neulinger)
Date: 02 Oct 2002 17:23:07 -0500
Subject: [OpenAFS] forgot the attachements.. duhhhh...
In-Reply-To: <1033596476.1902.25.camel@newadler.phy.bnl.gov>
References: <1033596476.1902.25.camel@newadler.phy.bnl.gov>
Message-ID: <1033597387.27570.24.camel@cessna.rollanet.org>
The gcc32 patch won't be accepted, as it breaks compiles for older
gcc's.
It will need to say with the warnings, or add a configure check for what
syntax to use.
-- Nathan
On Wed, 2002-10-02 at 17:07, Adler, Stephen wrote:
> long day...
--
------------------------------------------------------------
Nathan Neulinger EMail: nneul@umr.edu
University of Missouri - Rolla Phone: (573) 341-4841
Computing Services Fax: (573) 341-4216
From warlord@MIT.EDU Wed Oct 2 23:34:57 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 02 Oct 2002 18:34:57 -0400
Subject: [OpenAFS] redhat 8.0
In-Reply-To: <1033596324.1902.22.camel@newadler.phy.bnl.gov>
References: <1033596324.1902.22.camel@newadler.phy.bnl.gov>
Message-ID:
Any chance you could actually send a diff of the spec rather
than the whole spec? At a glance you:
changed the version number
added the patch file
However I may have missed something else.
-derek
PS: Considering the kernel module doesn't load, I don't see a reason
to worry about the packaging yet. I'm sure that, in the process of
getting AFS to work with the lack of the sys_call_table they will get
the modules to build.
"Adler, Stephen" writes:
> I've also gone through the exersize to build afs 1.2.7 on my red hat
> 8.0 system. I've attached the .spec and a patch file. But
> as with the other posts, I get the unresolved sys_call_table symbol
> error as well when I go to load the module. I'll be glad to
> add an account on my system to someone from the openafs development
> team who wishes to work on this problem.
>
> Cheers. Steve.
>
>
>
> _______________________________________________
> 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 jds@soltis.cc Thu Oct 3 02:53:32 2002
From: jds@soltis.cc (Jesus Delgado)
Date: Wed, 2 Oct 2002 19:53:32 -0600
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
In-Reply-To: <20021002151128.M26033@soltis.cc>
References: <20021002151128.M26033@soltis.cc>
Message-ID: <20021002195332.M26176@soltis.cc>
---------- Forwarded Message -----------
Hi:
Problems when try the build openafs source rpm in redhat 8.0
the errors is the same with openafs-1.2.6 y openafs-1.2.7:
Errors in openafs-1.2.7
.4.1eq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1 rpmlib(CompressedFileNames) <=
3.0.4-1
Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
rpmlib(CompressedFileNames) <= 3.0.4-1
Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
Processing files: openafs-krb5-1.2.7-rh7.3.1
error: File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
error: File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
Requires: openafs = 1.2.7
RPM build errors:
File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
Errors with openafs-1.2.6
Requires: openafs = 1.2.6 openafs-client = 1.2.6
Obsoletes: openafs-client-compat
Processing files: openafs-kpasswd-1.2.6-rh7.3.1
Finding Provides: /usr/lib/rpm/find-provides
Finding Requires: /usr/lib/rpm/find-requires
PreReq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1 rpmlib(CompressedFileNames) <=
3.0.4-1
Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
rpmlib(CompressedFileNames) <= 3.0.4-1
Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
Processing files: openafs-krb5-1.2.6-rh7.3.1
error: File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
error: File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
Requires: openafs = 1.2.6
RPM build errors:
File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
Help me plase
Regards.
------- End of Forwarded Message -------
From warlord@MIT.EDU Thu Oct 3 03:11:21 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 02 Oct 2002 22:11:21 -0400
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
In-Reply-To: <20021002195332.M26176@soltis.cc>
References: <20021002151128.M26033@soltis.cc>
<20021002195332.M26176@soltis.cc>
Message-ID:
Hi,
We do not support RedHat 8.0 yet.
We will release an RPM for RH8.0 once it is supported.
-derek
"Jesus Delgado" writes:
> Hi:
>
> Problems when try the build openafs source rpm in redhat 8.0
> the errors is the same with openafs-1.2.6 y openafs-1.2.7:
>
> Errors in openafs-1.2.7
[snip]
> Help me plase
>
> Regards.
--
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 mitch@ccmr.cornell.edu Thu Oct 3 06:41:56 2002
From: mitch@ccmr.cornell.edu (Mitch Collinsworth)
Date: Thu, 3 Oct 2002 01:41:56 -0400 (EDT)
Subject: [OpenAFS] AFS Training ?
In-Reply-To:
Message-ID:
Following up to my own followup: There is currently only one person
registered for the December class. One of my group would also like
to attend, making a total of 2 so far. We couldn't get an exact
answer from IBM as to what constitutes sufficient critical mass to
be sure the class won't be cancelled, but it seems to be in the
neighborhood of 4 or 5. Are there others here who might be interested
in this? Write to me off-list and I can give you the address of the
teacher if you want more direct info. My selfish motivation is in
recruiting enough additional attendees to see the class not get
canceled again. If you'd like to go but the December dates are bad,
when is better? Maybe we can work with them in scheduling it for a
time when we can succeed in achieving critical mass.
It's been over a decade since I took this class, but it was a good
class then and would be quite useful today even if little has changed
in the presentation since then.
-Mitch
On Sun, 29 Sep 2002, Mitch Collinsworth wrote:
> External web site is:
> http://www7b.software.ibm.com/wsdd/education/enablement/curriculum/sw800.html
>
> A word of caution: Just because it's listed here doesn't mean it will
> actually be held. The last class that was scheduled for Pittsburgh was
> cancelled at the last minute due to insufficient number of registrants.
>
> There was supposed to be another one this fall in California (San Jose?).
> It no longer seems to be on the schedule. I take it it was cancelled, too.
>
> -Mitch
>
>
> On Sun, 29 Sep 2002, Daniel Clark/Cambridge/IBM wrote:
>
> > Paul Blackburn wrote:
> > > Transarc used to run some excellent AFS administrator courses.
> > > Does anyone know of AFS training available today?
> >
> > According to w3.education.ibm.com (internal IBM site, there is some
> > external equivalent but I don't know what it is) Transarc - now "IBM
> > Pittsburgh Lab" - still runs the "AFS Administration" course. It's course
> > code SW800 and will next occur in Pittsburgh, PA 2002-12-10 to 2002-12-13.
> > Below are the full details, which include a contact number/email to get
> > more info and future course dates. I have taken this class and can vouch
> > that it is excellent. My class was taught by someone who had been
> > administering AFS since it was a research project at CMU, so in addition to
> > access to deep technical knowledge there were also interesting historical
> > asides.
> >
> > Course Details: AFS Administration
> >
> > 2002-12-10 to 2002-12-13
> >
> > Course Code:SW800
> > Section:K8II
> >
> > Course Location:
> > Room: TBA 2
> > IBM Pittsburgh Lab 9Fl
> > 11 Stanwix Street
> > Pittsburgh
> > United States
> > 15222
> >
> > Course Contact:
> > Tracy Linza
> > linza at IBMUS
> > 412-667-4477
> > TL 989-4477
> >
> > Course Schedule:
> > 2002-12-10 to 2002-12-13
> > 09:00:00 to 18:00:00
> > Last Day End Time: 17:30:00
> >
> > Last Cancellation Date without penalty:
> > 2002-11-25
> >
> >
> > Equipment Required:
> > None
> >
> > Course Enrollments:
> > Total Students Confirmed: 1 of 14
> > Total on Standby: 0
> >
> > Additional Course Information:
> > None
> >
> > Comments:
> > None
> >
> > Location Specific Info:
> >
> > LOCATION:
> > IBM Pittsburgh Lab
> > 11 Stanwix Street
> > 9th FL
> > Pittsburgh, PA 15222
> >
> > CONTACT:
> > 412-667-4433
> > TL 989-4433
> >
> > BADGES:
> > See lobby receptionist on 9th Floor
> > All visitors are required to sign-in
> > (8:00am - 6:00pm)
> >
> > LOCAL HOTEL INFORMATION
> >
> > Pittsburgh Hilton
> > 600 Commonwealth Place
> > Gateway Center
> > Pittsburgh, PA 15222
> > 412-391-4600
> > Ask for IBM/Transarc rate of $92 per night
> > (Subject to change without notice)
> > 5 minute walk to Lab
> >
> > Sheraton Hotel
> > Carson & SMithfield Streets
> > Pittsburgh, PA 15219
> > 412-261-2000
> > 5-10 minute subway ride to Lab
> > Ask for IBM/Transarc rate of $116 per night
> > (Subject to change without notice)
> >
> >
> > DIRECTIONS
> >
> > FROM AIRPORT:
> > * From Airport follow the signs toward Pittsburgh
> > (Approx 15 miles) to the Fort Pitt Tunnel
> > * Stay in the left lane.
> > * After the tunnel, you will be on the Fort Pitt Bridge
> > * Take the Boulevard of the Allies Exit
> > * Proceed to Stanwix St and turn right.
> > * Proceed one block to 11 Stanwix
> > (Look for Blattner Bruner revolving sign)
> > * Parking garage is in the building
> > * Parking in the garage is $14 per day - student expense.
> >
> > --
> > Daniel Clark # Sys Admin & Release Engineer
> > IBM > Lotus > Messaging Technology Group
> >
> > _______________________________________________
> > 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 mpb@est.ibm.com Thu Oct 3 08:53:52 2002
From: mpb@est.ibm.com (Paul Blackburn)
Date: Thu, 03 Oct 2002 08:53:52 +0100
Subject: [OpenAFS] AFS Training ?
References:
Message-ID: <3D9BF790.3020000@est.ibm.com>
I have also attended AFS classes presented by Transarc staff some years ago.
In my experience, these were some of the best computer education courses
I have been lucky enough to go to. The course presenter was very
knowledgeable
with a good experience in the practical issues of administering AFS.
If you have the opportunity, then I highly recommend booking this course.
Good training makes all the difference to having a successful
implementation of AFS at your site.
Don't miss this!
--
cheers
paul http://acm.org/~mpb
"The defense of individual rights has reached such extremes as to make
society as a whole defenseless agains certain individuals. It is
time, in the West, to defend not so much human rights as human
obligations." -- Aleksandr Solzhenisyn
Mitch Collinsworth wrote:
>Following up to my own followup: There is currently only one person
>registered for the December class. One of my group would also like
>to attend, making a total of 2 so far. We couldn't get an exact
>answer from IBM as to what constitutes sufficient critical mass to
>be sure the class won't be cancelled, but it seems to be in the
>neighborhood of 4 or 5. Are there others here who might be interested
>in this? Write to me off-list and I can give you the address of the
>teacher if you want more direct info. My selfish motivation is in
>recruiting enough additional attendees to see the class not get
>canceled again. If you'd like to go but the December dates are bad,
>when is better? Maybe we can work with them in scheduling it for a
>time when we can succeed in achieving critical mass.
>
>It's been over a decade since I took this class, but it was a good
>class then and would be quite useful today even if little has changed
>in the presentation since then.
>
>-Mitch
>
>
>On Sun, 29 Sep 2002, Mitch Collinsworth wrote:
>
>>External web site is:
>>http://www7b.software.ibm.com/wsdd/education/enablement/curriculum/sw800.html
>>
>>A word of caution: Just because it's listed here doesn't mean it will
>>actually be held. The last class that was scheduled for Pittsburgh was
>>cancelled at the last minute due to insufficient number of registrants.
>>
>>There was supposed to be another one this fall in California (San Jose?).
>>It no longer seems to be on the schedule. I take it it was cancelled, too.
>>
>>-Mitch
>>
>>
>>On Sun, 29 Sep 2002, Daniel Clark/Cambridge/IBM wrote:
>>
>>>Paul Blackburn wrote:
>>>
>>>>Transarc used to run some excellent AFS administrator courses.
>>>>Does anyone know of AFS training available today?
>>>>
>>>According to w3.education.ibm.com (internal IBM site, there is some
>>>external equivalent but I don't know what it is) Transarc - now "IBM
>>>Pittsburgh Lab" - still runs the "AFS Administration" course. It's course
>>>code SW800 and will next occur in Pittsburgh, PA 2002-12-10 to 2002-12-13.
>>>Below are the full details, which include a contact number/email to get
>>>more info and future course dates. I have taken this class and can vouch
>>>that it is excellent. My class was taught by someone who had been
>>>administering AFS since it was a research project at CMU, so in addition to
>>>access to deep technical knowledge there were also interesting historical
>>>asides.
>>>
>>>Course Details: AFS Administration
>>>
>>>2002-12-10 to 2002-12-13
>>>
>>>Course Code:SW800
>>>Section:K8II
>>>
>>>Course Location:
>>> Room: TBA 2
>>> IBM Pittsburgh Lab 9Fl
>>> 11 Stanwix Street
>>> Pittsburgh
>>> United States
>>> 15222
>>>
>>>Course Contact:
>>> Tracy Linza
>>> linza at IBMUS
>>> 412-667-4477
>>> TL 989-4477
>>>
>>>Course Schedule:
>>> 2002-12-10 to 2002-12-13
>>> 09:00:00 to 18:00:00
>>> Last Day End Time: 17:30:00
>>>
>>>Last Cancellation Date without penalty:
>>> 2002-11-25
>>>
>>>
>>>Equipment Required:
>>> None
>>>
>>>Course Enrollments:
>>> Total Students Confirmed: 1 of 14
>>> Total on Standby: 0
>>>
>>>Additional Course Information:
>>> None
>>>
>>>Comments:
>>> None
>>>
>>>Location Specific Info:
>>>
>>>LOCATION:
>>> IBM Pittsburgh Lab
>>> 11 Stanwix Street
>>> 9th FL
>>> Pittsburgh, PA 15222
>>>
>>>CONTACT:
>>> 412-667-4433
>>> TL 989-4433
>>>
>>>BADGES:
>>> See lobby receptionist on 9th Floor
>>> All visitors are required to sign-in
>>> (8:00am - 6:00pm)
>>>
>>>LOCAL HOTEL INFORMATION
>>>
>>>Pittsburgh Hilton
>>>600 Commonwealth Place
>>>Gateway Center
>>>Pittsburgh, PA 15222
>>>412-391-4600
>>>Ask for IBM/Transarc rate of $92 per night
>>>(Subject to change without notice)
>>>5 minute walk to Lab
>>>
>>>Sheraton Hotel
>>>Carson & SMithfield Streets
>>>Pittsburgh, PA 15219
>>>412-261-2000
>>>5-10 minute subway ride to Lab
>>>Ask for IBM/Transarc rate of $116 per night
>>>(Subject to change without notice)
>>>
>>>
>>>DIRECTIONS
>>>
>>>FROM AIRPORT:
>>>* From Airport follow the signs toward Pittsburgh
>>> (Approx 15 miles) to the Fort Pitt Tunnel
>>>* Stay in the left lane.
>>>* After the tunnel, you will be on the Fort Pitt Bridge
>>>* Take the Boulevard of the Allies Exit
>>>* Proceed to Stanwix St and turn right.
>>>* Proceed one block to 11 Stanwix
>>> (Look for Blattner Bruner revolving sign)
>>>* Parking garage is in the building
>>>* Parking in the garage is $14 per day - student expense.
>>>
>>>--
>>>Daniel Clark # Sys Admin & Release Engineer
>>>IBM > Lotus > Messaging Technology Group
>>>
>>>_______________________________________________
>>>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
>>
>
>_______________________________________________
>OpenAFS-info mailing list
>OpenAFS-info@openafs.org
>https://lists.openafs.org/mailman/listinfo/openafs-info
>
From traxtopel@HotPOP.com Thu Oct 3 13:32:06 2002
From: traxtopel@HotPOP.com (trax)
Date: 03 Oct 2002 14:32:06 +0200
Subject: [OpenAFS] openafs 1.2.7 - mandrake9.0
Message-ID: <1033647143.2400.91.camel@localhost.localdomain>
I am trying to recompile the openafs1.2.7 RPM on Mandrake 9.0.
The problem I am having is building the openafs-krb5 package.
All other packages build correctly.
I have modified the spec so that the krb5./configure to kerberos path to
/usr as opposed to /usr/kereberos(which is for rh7.3).
Processing files: openafs-krb5-1.2.7-mdk9.0.1
error: File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
error: File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
After rpm is finished recompiling, it bottles out as the aklog file is
not built.
All other packages build correctly
Running the make all manually from openafs-krb5-1.3/, produces the
following.
[root@localhost src]# make all
gcc -o aklog aklog.o aklog_main.o aklog_param.o krb_util.o
linked_list.o adderrtable.o -lkrb524
-L/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib -L/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs -lsys -lprot -lubik -lauth -lrxkad -lrx -llwp -ldes -lsys /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/util.a -lresolv -lkrb5 -lk5crypto -lcom_err
/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/libcom_err.a(error_msg.o): In function `add_to_error_table':
error_msg.o(.text+0x220): multiple definition of `add_to_error_table'
adderrtable.o:/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/adderrtable.c:31: first defined here
/usr/bin/ld: Warning: size of symbol `add_to_error_table' changed from
16 to 49 in
/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/libcom_err.a(error_msg.o)
collect2: ld returned 1 exit status
make: *** [aklog] Error 1
config.log shows the following ...
This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.
configure:564: checking for C compiler
configure:576: checking for gcc
configure:689: checking whether the C compiler (gcc ) works
configure:705: gcc -o conftest conftest.c 1>&5
configure:731: checking whether the C compiler (gcc ) is a
cross-compiler
configure:736: checking whether we are using GNU C
configure:745: gcc -E conftest.c
configure:764: checking whether gcc accepts -g
configure:899: checking for a BSD compatible install
configure:979: checking for getDirPath in
/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/util.a
configure:997: gcc -o conftest -g -O2
-I/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//include conftest.c /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/util.a 1>&5
configure:1055: checking for socket in -lsocket
configure:1074: gcc -o conftest -g -O2
-I/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//include conftest.c -lsocket -lnsl 1>&5
/usr/bin/ld: cannot find -lsocket
collect2: ld returned 1 exit status
configure: failed program was:
#line 1063 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error. */
/* We use char because int might match the return type of a gcc2
builtin and then its argument prototype would still apply. */
char socket();
int main() {
socket()
; return 0; }
configure:1095: checking for t_bind in -lnsl
configure:1114: gcc -o conftest -g -O2
-I/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//include conftest.c -lnsl 1>&5
/tmp/ccg62sYD.o: In function `main':
/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/configure:1110:
undefined reference to `t_bind'
collect2: ld returned 1 exit status
configure: failed program was:
#line 1103 "configure"
#include "confdefs.h"
/* Override any gcc2 internal prototype to avoid an error. */
/* We use char because int might match the return type of a gcc2
builtin and then its argument prototype would still apply. */
char t_bind();
int main() {
t_bind()
; return 0; }
configure:1140: checking for compile
configure:1168: gcc -o conftest -g -O2
-I/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//include conftest.c -lgen 1>&5
/usr/bin/ld: cannot find -lgen
collect2: ld returned 1 exit status
configure: failed program was:
#line 1145 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
which can conflict with char compile(); below. */
#include
/* Override any gcc2 internal prototype to avoid an error. */
/* We use char because int might match the return type of a gcc2
builtin and then its argument prototype would still apply. */
char compile();
int main() {
/* The GNU C library defines this for functions which it implements
to always fail with ENOSYS. Some functions are actually named
something starting with __ and the normal name is an alias. */
#if defined (__stub_compile) || defined (__stub___compile)
choke me
#else
compile();
#endif
; return 0; }
configure:1140: checking for step
configure:1168: gcc -o conftest -g -O2
-I/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//include conftest.c -lgen 1>&5
/usr/bin/ld: cannot find -lgen
collect2: ld returned 1 exit status
configure: failed program was:
#line 1145 "configure"
#include "confdefs.h"
/* System header to define __stub macros and hopefully few prototypes,
which can conflict with char step(); below. */
#include
/* Override any gcc2 internal prototype to avoid an error. */
/* We use char because int might match the return type of a gcc2
buchar step();
int main() {
/* The GNU C library defines this for functions which it implements
to always fail with ENOSYS. Some functions are actually named
something starting with __ and the normal name is an alias. */
#if defined (__stub_step) || defined (__stub___step)
choke me
#else
step();
#endif
; return 0; }
iltin and then its argument prototype would still apply. */
I have tried recompiling with gcc2.96 aswell as 3.2, both give the same
result.
From adler@bnl.gov Thu Oct 3 13:55:27 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 08:55:27 -0400
Subject: [OpenAFS] redhat 8.0
In-Reply-To:
References: <1033596324.1902.22.camel@newadler.phy.bnl.gov>
Message-ID: <1033649727.15413.2.camel@newadler.phy.bnl.gov>
one more tweek, I aliased autoconf to autoconf-2.13. It looks like
redhat 8.0 ships with 2 versions of autoconf and the default is
2.53. Trying to build openafs with autoconf 2.53 leads to problems
in AC_LANG definition when configing one of the sub systems of open
afs. (Sorry, don't remember which one....)
Summary: 3 changes.
1) the version
2) patch2:
3) alias autoconf=autoconf-2.13
Cheers. Steve.
On Wed, 2002-10-02 at 18:34, Derek Atkins wrote:
> Any chance you could actually send a diff of the spec rather
> than the whole spec? At a glance you:
>
> changed the version number
> added the patch file
>
> However I may have missed something else.
>
> -derek
>
> PS: Considering the kernel module doesn't load, I don't see a reason
> to worry about the packaging yet. I'm sure that, in the process of
> getting AFS to work with the lack of the sys_call_table they will get
> the modules to build.
>
> "Adler, Stephen" writes:
>
> > I've also gone through the exersize to build afs 1.2.7 on my red hat
> > 8.0 system. I've attached the .spec and a patch file. But
> > as with the other posts, I get the unresolved sys_call_table symbol
> > error as well when I go to load the module. I'll be glad to
> > add an account on my system to someone from the openafs development
> > team who wishes to work on this problem.
> >
> > Cheers. Steve.
> >
> >
> >
> > _______________________________________________
> > 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 adler@bnl.gov Thu Oct 3 14:00:20 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 09:00:20 -0400
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
In-Reply-To:
References: <20021002151128.M26033@soltis.cc>
<20021002195332.M26176@soltis.cc>
Message-ID: <1033650020.15413.7.camel@newadler.phy.bnl.gov>
How can we help?
On Wed, 2002-10-02 at 22:11, Derek Atkins wrote:
> Hi,
>
> We do not support RedHat 8.0 yet.
>
> We will release an RPM for RH8.0 once it is supported.
>
> -derek
>
> "Jesus Delgado" writes:
>
> > Hi:
> >
> > Problems when try the build openafs source rpm in redhat 8.0
> > the errors is the same with openafs-1.2.6 y openafs-1.2.7:
> >
> > Errors in openafs-1.2.7
>
> [snip]
>
> > Help me plase
> >
> > Regards.
>
> --
> 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
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
From adler@bnl.gov Thu Oct 3 14:06:08 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 09:06:08 -0400
Subject: [OpenAFS] forgot the attachements.. duhhhh...
In-Reply-To: <1033597387.27570.24.camel@cessna.rollanet.org>
References: <1033596476.1902.25.camel@newadler.phy.bnl.gov>
<1033597387.27570.24.camel@cessna.rollanet.org>
Message-ID: <1033650368.15413.10.camel@newadler.phy.bnl.gov>
I wouldn't expect the patch to be permanently integrated into
the rpm build for the reasons you stated. Basically it's a quick
work around so that I can get up and running. I'm very much a
novice with m4/autoconf and all that.
Steve.
On Wed, 2002-10-02 at 18:23, Nathan Neulinger wrote:
> The gcc32 patch won't be accepted, as it breaks compiles for older
> gcc's.
>
> It will need to say with the warnings, or add a configure check for what
> syntax to use.
>
> -- Nathan
>
> On Wed, 2002-10-02 at 17:07, Adler, Stephen wrote:
> > long day...
> --
>
> ------------------------------------------------------------
> Nathan Neulinger EMail: nneul@umr.edu
> University of Missouri - Rolla Phone: (573) 341-4841
> Computing Services Fax: (573) 341-4216
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
From nneul@umr.edu Thu Oct 3 14:14:47 2002
From: nneul@umr.edu (Neulinger, Nathan)
Date: Thu, 3 Oct 2002 08:14:47 -0500
Subject: [OpenAFS] forgot the attachements.. duhhhh...
Message-ID: <2B45A04D8F18D947A400F0850CE3B53B060D69@umr-mail7.umr.edu>
Problem is - right now, redhat's kernel will not support afs at all due
to changes they made that removed support for some facilities that afs
requires in order to function properly. If you switch to a
virgin/standard kernel build instead of the kernel rpm, it will work ok.
You shouldn't need that patch to get it built anyway - they are just
warnings, they shouldn't prevent the compile from finishing.=20
-- Nathan
------------------------------------------------------------
Nathan Neulinger EMail: nneul@umr.edu
University of Missouri - Rolla Phone: (573) 341-4841
Computing Services Fax: (573) 341-4216
> -----Original Message-----
> From: Adler, Stephen [mailto:adler@bnl.gov]=20
> Sent: Thursday, October 03, 2002 8:06 AM
> To: Neulinger, Nathan
> Cc: openafs-info@openafs.org
> Subject: Re: [OpenAFS] forgot the attachements.. duhhhh...
>=20
>=20
> I wouldn't expect the patch to be permanently integrated into
> the rpm build for the reasons you stated. Basically it's a quick
> work around so that I can get up and running. I'm very much a
> novice with m4/autoconf and all that.
>=20
> Steve.
>=20
> On Wed, 2002-10-02 at 18:23, Nathan Neulinger wrote:
> > The gcc32 patch won't be accepted, as it breaks compiles for older
> > gcc's.=20
> >=20
> > It will need to say with the warnings, or add a configure=20
> check for what
> > syntax to use.
> >=20
> > -- Nathan
> >=20
> > On Wed, 2002-10-02 at 17:07, Adler, Stephen wrote:
> > > long day...
> > --=20
> >=20
> > ------------------------------------------------------------
> > Nathan Neulinger EMail: nneul@umr.edu
> > University of Missouri - Rolla Phone: (573) 341-4841
> > Computing Services Fax: (573) 341-4216
> >=20
> > _______________________________________________
> > OpenAFS-info mailing list
> > OpenAFS-info@openafs.org
> > https://lists.openafs.org/mailman/listinfo/openafs-info
>=20
>=20
From adler@bnl.gov Thu Oct 3 15:08:28 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 10:08:28 -0400
Subject: [OpenAFS] patching the red hat kernel
Message-ID: <1033654108.15481.67.camel@newadler.phy.bnl.gov>
Guys,
I would like to get afs up and running on my red hat 8.0 system and
I'm rather dead in the water right now. One thing I could do is
patch the red hat kernel so that call_sys_table is visible again.
Does anyone know what it would take to do this? At least I could
get going again until an official version comes out from the
openafs group. (or is this a waist of my time and I should just
be patient...)
Cheers. Steve.
From nneul@umr.edu Thu Oct 3 15:16:19 2002
From: nneul@umr.edu (Neulinger, Nathan)
Date: Thu, 3 Oct 2002 09:16:19 -0500
Subject: [OpenAFS] patching the red hat kernel
Message-ID: <2B45A04D8F18D947A400F0850CE3B53B060D6F@umr-mail7.umr.edu>
If you're willing to do that, you're better off just installing a virgin
kernel from kernel.org.=20
-- Nathan
------------------------------------------------------------
Nathan Neulinger EMail: nneul@umr.edu
University of Missouri - Rolla Phone: (573) 341-4841
Computing Services Fax: (573) 341-4216
> -----Original Message-----
> From: Adler, Stephen [mailto:adler@bnl.gov]=20
> Sent: Thursday, October 03, 2002 9:08 AM
> To: openafs-info@openafs.org
> Subject: [OpenAFS] patching the red hat kernel
>=20
>=20
> Guys,
>=20
> I would like to get afs up and running on my red hat 8.0 system and
> I'm rather dead in the water right now. One thing I could do is
> patch the red hat kernel so that call_sys_table is visible again.
> Does anyone know what it would take to do this? At least I could
> get going again until an official version comes out from the
> openafs group. (or is this a waist of my time and I should just
> be patient...)
>=20
> Cheers. Steve.
>=20
>=20
>=20
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
>=20
From warlord@MIT.EDU Thu Oct 3 15:31:02 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 03 Oct 2002 10:31:02 -0400
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
In-Reply-To: <1033650020.15413.7.camel@newadler.phy.bnl.gov>
References: <20021002151128.M26033@soltis.cc>
<20021002195332.M26176@soltis.cc>
<1033650020.15413.7.camel@newadler.phy.bnl.gov>
Message-ID:
"Adler, Stephen" writes:
> How can we help?
By being patient... The developers know the problem exists and
are trying to determine a workaround. This will take _time_, and
the more time we spend responding to email the less time we have
to work on the problem. ;)
-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 warlord@MIT.EDU Thu Oct 3 15:29:59 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 03 Oct 2002 10:29:59 -0400
Subject: [OpenAFS] patching the red hat kernel
In-Reply-To: <1033654108.15481.67.camel@newadler.phy.bnl.gov>
References: <1033654108.15481.67.camel@newadler.phy.bnl.gov>
Message-ID:
You're always welcome to tread off into undiscovered country,
but if you choose to do so you are on your own. If it were
_ME_ I would wait, but honestly I don't know how long it will
be. It might be a couple days, it might be a couple weeks.
Is there any particular reason you _NEED_ to use RH8 in your
environment NOW NOW NOW?
-derek
"Adler, Stephen" writes:
> Guys,
>
> I would like to get afs up and running on my red hat 8.0 system and
> I'm rather dead in the water right now. One thing I could do is
> patch the red hat kernel so that call_sys_table is visible again.
> Does anyone know what it would take to do this? At least I could
> get going again until an official version comes out from the
> openafs group. (or is this a waist of my time and I should just
> be patient...)
>
> Cheers. Steve.
>
>
>
> _______________________________________________
> 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 schmitt@inf.ethz.ch Thu Oct 3 15:34:09 2002
From: schmitt@inf.ethz.ch (Marc Schmitt)
Date: Thu, 03 Oct 2002 16:34:09 +0200
Subject: [OpenAFS] patching the red hat kernel
References: <2B45A04D8F18D947A400F0850CE3B53B060D6F@umr-mail7.umr.edu>
Message-ID: <3D9C5561.6010804@inf.ethz.ch>
What about rebuilding the 2.4.18-10 kernel SRPM, installing the
resulting RPMs and then rebuilding the OpenAFS SRPM and installing them?
Greetz
Marc
Neulinger, Nathan wrote:
> If you're willing to do that, you're better off just installing a virgin
> kernel from kernel.org.
>
> -- Nathan
From nneul@umr.edu Thu Oct 3 15:35:30 2002
From: nneul@umr.edu (Neulinger, Nathan)
Date: Thu, 3 Oct 2002 09:35:30 -0500
Subject: [OpenAFS] patching the red hat kernel
Message-ID: <2B45A04D8F18D947A400F0850CE3B53B060D70@umr-mail7.umr.edu>
If 2.4.18-10 was what was previously working with 7.x, yeah, that would
likely work just fine.=20
-- Nathan
------------------------------------------------------------
Nathan Neulinger EMail: nneul@umr.edu
University of Missouri - Rolla Phone: (573) 341-4841
Computing Services Fax: (573) 341-4216
> -----Original Message-----
> From: Marc Schmitt [mailto:schmitt@inf.ethz.ch]=20
> Sent: Thursday, October 03, 2002 9:34 AM
> To: Neulinger, Nathan
> Cc: Adler, Stephen; openafs-info@openafs.org
> Subject: Re: [OpenAFS] patching the red hat kernel
>=20
>=20
> What about rebuilding the 2.4.18-10 kernel SRPM, installing the=20
> resulting RPMs and then rebuilding the OpenAFS SRPM and=20
> installing them?
>=20
> Greetz
> Marc
>=20
> Neulinger, Nathan wrote:
> > If you're willing to do that, you're better off just=20
> installing a virgin
> > kernel from kernel.org.=20
> >=20
> > -- Nathan
>=20
>=20
From warlord@MIT.EDU Thu Oct 3 15:55:33 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 03 Oct 2002 10:55:33 -0400
Subject: [OpenAFS] openafs 1.2.7 - mandrake9.0
In-Reply-To: <1033647143.2400.91.camel@localhost.localdomain>
References: <1033647143.2400.91.camel@localhost.localdomain>
Message-ID:
trax writes:
> gcc -o aklog aklog.o aklog_main.o aklog_param.o krb_util.o
> linked_list.o adderrtable.o -lkrb524
> -L/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib -L/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs -lsys -lprot -lubik -lauth -lrxkad -lrx -llwp -ldes -lsys /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/util.a -lresolv -lkrb5 -lk5crypto -lcom_err
> /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/libcom_err.a(error_msg.o): In function `add_to_error_table':
> error_msg.o(.text+0x220): multiple definition of `add_to_error_table'
> adderrtable.o:/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/adderrtable.c:31: first defined here
> /usr/bin/ld: Warning: size of symbol `add_to_error_table' changed from
> 16 to 49 in
> /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/libcom_err.a(error_msg.o)
> collect2: ld returned 1 exit status
> make: *** [aklog] Error 1
[configure snipped snipped]
> I have tried recompiling with gcc2.96 aswell as 3.2, both give the same
> result.
The problem is clearly a redefinition of add_to_error_table(). The
question is why this is happening. Your configure snippet did not
show anything interesting.
This doesn't happen to me on any of my RH builds. The reason is that
on my RH system it's pulling in /usr/kerberos/lib/libcom_err.so
instead of .../lib/afs/libcom_err.a! So the problem is how you've
changed the build process due to kerberos living in /usr instead of
/usr/kerberos. You need to keep the -L/usr (as opposed to
-L/usr/kerberos) early in the link path so it pulls in the "right"
com_err library.
-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 jasper@plainspace.com Thu Oct 3 16:40:18 2002
From: jasper@plainspace.com (jasper@plainspace.com)
Date: Thu, 3 Oct 2002 17:40:18 +0200
Subject: [OpenAFS] cannot mount root.cell on /afs/cellname
Message-ID: <3D9C8102.18337.1681D54@localhost>
Hi *,
I'm trying to install openafs-1.2.3final2-6 (debian package version) on a GNU/Linux
debian 3.0 server with kernel 2.2.20.
With the afs-newcell script and a lot of manual configuration I managed to get the
servers and the client running and to create root.afs and root.cell
When I try to run the afs-rootvol script it stops when it tries to fs mkmount the root.cell
volume:
fs: File '/afs/ardiden.plainspace.com' doesn't exist
also I can't write anything to the /afs directory :
/afs# touch x
touch: creating `x': No such file or directory
My BosLog reads:
Thu Oct 3 15:34:18 2002: Server directory access is okay
Thu Oct 3 15:34:20 2002: fs:salv exited with code 0
My FileLog reads:
Thu Oct 3 15:34:21 2002 File server starting
Thu Oct 3 15:34:21 2002 afs_krb_get_lrealm failed, using ardiden.plainspace.com.
Thu Oct 3 15:34:21 2002 VL_RegisterAddrs rpc failed; will retry periodically
(code=5376, err=4)
Thu Oct 3 15:34:22 2002 Partition /vicepa: attached 2 volumes; 0 volumes not
attached
Thu Oct 3 15:34:22 2002 Getting FileServer name...
Thu Oct 3 15:34:22 2002 FileServer host name is 'ardiden'
Thu Oct 3 15:34:22 2002 Getting FileServer address...
Thu Oct 3 15:34:22 2002 FileServer ardiden has address 192.168.12.2 (0x20ca8c0 or
0xc0a80c02 in host byte order)
Thu Oct 3 15:34:22 2002 File Server started Thu Oct 3 15:34:22 2002
fs examine /afs returns:
Volume status for vid = 536870948 named root.afs
Current disk quota is 5000
Current blocks used are 2
The partition has 30051 blocks available out of 30091
vos examine root.afs returns:
root.afs 536870948 RW 2 K On-line
ardiden.plainspace.com /vicepa
RWrite 536870948 ROnly 0 Backup 0
MaxQuota 5000 K
Creation Thu Oct 3 12:19:53 2002
Last Update Thu Oct 3 17:16:39 2002
23 accesses in the past day (i.e., vnode references)
RWrite: 536870948
number of sites -> 1
server ardiden.plainspace.com partition /vicepa RW Site
vos examine root.cell returns:
root.cell 536870951 RW 2 K On-line
ardiden.plainspace.com /vicepa
RWrite 536870951 ROnly 0 Backup 0
MaxQuota 5000 K
Creation Thu Oct 3 12:44:25 2002
Last Update Thu Oct 3 12:44:25 2002
0 accesses in the past day (i.e., vnode references)
RWrite: 536870951
number of sites -> 1
server ardiden.plainspace.com partition /vicepa RW Site
permissions on the /afs dir are 0775. I guess Kerberos is configured right, tokens
returns
User's (AFS ID 1) tokens for afs@ardiden.plainspace.com [Expires Oct 4 03:15]
This server contains two ethernet-cards and I had to reconfigure the DNS server to
return only the ardiden.plainspace.com address and internal ip address and not the
external ip address and ISP given internethostname. Does this have to do anything
with it?
Another thing is that loading the openafs.o kernelmodule makes the kernel complaint
about 'tainting' the kernel. Trying to rmmod the openafs.o modules crashes the kernel.
Should I upgrade the kernel to 2.4.x and recompile the module?
Any help would be greatly appreciated,
Jasper den Hertog
From traxtopel@HotPOP.com Thu Oct 3 17:37:27 2002
From: traxtopel@HotPOP.com (trax)
Date: 03 Oct 2002 18:37:27 +0200
Subject: [OpenAFS] openafs 1.2.7 - mandrake9.0
In-Reply-To:
References: <1033647143.2400.91.camel@localhost.localdomain>
Message-ID: <1033663130.3795.5.camel@localhost.localdomain>
Derek,
Actually I shouted to soon using /usr/lib/libcom_err.so creates the
files as required. I am gonna try and build the whole package. I will
let you know if it works. If anyone wants the srpm, I can post it.
On Thu, 2002-10-03 at 16:55, Derek Atkins wrote:
> trax writes:
>
> > gcc -o aklog aklog.o aklog_main.o aklog_param.o krb_util.o
> > linked_list.o adderrtable.o -lkrb524
> > -L/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib -L/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs -lsys -lprot -lubik -lauth -lrxkad -lrx -llwp -ldes -lsys /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/util.a -lresolv -lkrb5 -lk5crypto -lcom_err
> > /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/libcom_err.a(error_msg.o): In function `add_to_error_table':
> > error_msg.o(.text+0x220): multiple definition of `add_to_error_table'
> > adderrtable.o:/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/adderrtable.c:31: first defined here
> > /usr/bin/ld: Warning: size of symbol `add_to_error_table' changed from
> > 16 to 49 in
> > /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/libcom_err.a(error_msg.o)
> > collect2: ld returned 1 exit status
> > make: *** [aklog] Error 1
>
> [configure snipped snipped]
>
> > I have tried recompiling with gcc2.96 aswell as 3.2, both give the same
> > result.
>
> The problem is clearly a redefinition of add_to_error_table(). The
> question is why this is happening. Your configure snippet did not
> show anything interesting.
>
> This doesn't happen to me on any of my RH builds. The reason is that
> on my RH system it's pulling in /usr/kerberos/lib/libcom_err.so
> instead of .../lib/afs/libcom_err.a! So the problem is how you've
> changed the build process due to kerberos living in /usr instead of
> /usr/kerberos. You need to keep the -L/usr (as opposed to
> -L/usr/kerberos) early in the link path so it pulls in the "right"
> com_err library.
>
> -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 openafs-info@openafs.org Thu Oct 3 17:49:42 2002
From: openafs-info@openafs.org (Derek Atkins)
Date: 03 Oct 2002 12:49:42 -0400
Subject: [OpenAFS] openafs 1.2.7 - mandrake9.0
In-Reply-To: <1033661905.3795.1.camel@localhost.localdomain>
References: <1033647143.2400.91.camel@localhost.localdomain>
<1033661905.3795.1.camel@localhost.localdomain>
Message-ID:
Please CC your responses to the list.
trax writes:
> I am running configure with
>
> ./configure --prefix=/usr --with-krb5=/usr \
> --with-afs=/usr/src/RPM/BUILD/openafs-1.2.7/i386_linux24/dest
That LOOKS reasonable, however...
> Which corresponds to what is shown in the rh7.3 spec.
> Any further ideas?
What's going on is that configure is looking in the default link-path
before it tries the --with-krb5= setting. Because kerberos is in the
default link-path, it is finding it before it puts in a -L%{krb5lib}.
This is an artifact of AC_FIND_LIB and I know of no "correct" workaround.
You can try running this command to effect a workaround in your case:
env LIBS=-L/usr ./configure ....
-derek
> On Thu, 2002-10-03 at 16:55, Derek Atkins wrote:
> > trax writes:
> >
> > > gcc -o aklog aklog.o aklog_main.o aklog_param.o krb_util.o
> > > linked_list.o adderrtable.o -lkrb524
> > > -L/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib -L/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs -lsys -lprot -lubik -lauth -lrxkad -lrx -llwp -ldes -lsys /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/util.a -lresolv -lkrb5 -lk5crypto -lcom_err
> > > /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/libcom_err.a(error_msg.o): In function `add_to_error_table':
> > > error_msg.o(.text+0x220): multiple definition of `add_to_error_table'
> > > adderrtable.o:/usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/adderrtable.c:31: first defined here
> > > /usr/bin/ld: Warning: size of symbol `add_to_error_table' changed from
> > > 16 to 49 in
> > > /usr/src/RPM/BUILD/openafs-1.2.7/openafs-krb5-1.3/src/../../i386_linux24/dest//lib/afs/libcom_err.a(error_msg.o)
> > > collect2: ld returned 1 exit status
> > > make: *** [aklog] Error 1
> >
> > [configure snipped snipped]
> >
> > > I have tried recompiling with gcc2.96 aswell as 3.2, both give the same
> > > result.
> >
> > The problem is clearly a redefinition of add_to_error_table(). The
> > question is why this is happening. Your configure snippet did not
> > show anything interesting.
> >
> > This doesn't happen to me on any of my RH builds. The reason is that
> > on my RH system it's pulling in /usr/kerberos/lib/libcom_err.so
> > instead of .../lib/afs/libcom_err.a! So the problem is how you've
> > changed the build process due to kerberos living in /usr instead of
> > /usr/kerberos. You need to keep the -L/usr (as opposed to
> > -L/usr/kerberos) early in the link path so it pulls in the "right"
> > com_err library.
> >
> > -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
>
>
>
--
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 traxtopel@HotPOP.com Thu Oct 3 19:35:18 2002
From: traxtopel@HotPOP.com (trax)
Date: 03 Oct 2002 20:35:18 +0200
Subject: [OpenAFS] mandrake 9.0 segmentation fault
Message-ID: <1033670164.3973.4.camel@localhost.localdomain>
Okay, I have managed to get mdk9.0 to compile, upoun starting afs, I see
[root@localhost root]# service afs start
Found libafs-2.4.19-16mdksecure-i686.o from SymTable... Loading...
Starting AFS services.....
/var/log/messages shows ...
Oct 3 20:33:28 localhost su(pam_unix)[4100]: session opened for user
root by (uid=500)
Oct 3 20:34:08 localhost kernel: sock_release(rx_socket) FIXME
Oct 3 20:34:08 localhost kernel: AFS: RX failed to initialize.
The it just sits there. Am I missing something here?
From David.Bear@asu.edu Thu Oct 3 20:01:05 2002
From: David.Bear@asu.edu (David Bear)
Date: Thu, 03 Oct 2002 12:01:05 -0700 (MST)
Subject: [OpenAFS] windows issues
Message-ID:
strange happenings with win2k and openafs 1.2.2b.
Worked fine yesterday.
today, try to get tokens.. always fails.
Then from cmd window issue:
net stop "IBM AFS Client"
net start "IBM AFS Client"
stops and starts successfully. Then klog ..
works!!! Why?
--
David Bear
College of Public Programs/ASU
480-965-8257
...the way is like water, going where nobody wants it to go
From David.Bear@asu.edu Thu Oct 3 20:04:50 2002
From: David.Bear@asu.edu (David Bear)
Date: Thu, 03 Oct 2002 12:04:50 -0700 (MST)
Subject: [OpenAFS] using openafs over wireless
Message-ID:
Anyone have experience running open-afs on win2k over a wireless nic? we
expected lan throughput to be 2-4 mbs.
--
David Bear
College of Public Programs/ASU
480-965-8257
...the way is like water, going where nobody wants it to go
From adler@bnl.gov Thu Oct 3 20:40:49 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 15:40:49 -0400
Subject: [OpenAFS] sys_call_table symbol
Message-ID: <1033674049.7281.25.camel@newadler.phy.bnl.gov>
How does the openafs module resolve the sys_call_table symbol? I've
built a version of the redhat 8.0 kernel which "exports" this
symbol so that it is now in /proc/ksyms. Yet when I went to load
the module, I got the unresolved symbol error again. Is there
like some kind of command I need to execute which updates some
symbol table file somewhere? Or did my attempts at building
a kernel which exports sys_call_table fail?
Steve.
[adler@newadler adler]$ cat /etc/redhat-release
Red Hat Linux release 8.0 (Psyche)
[adler@newadler adler]$ grep sys_call_table /proc/ksyms
c0320d74 sys_call_table_Rsmp_dfdb18bd
From slack@quackmaster.net Thu Oct 3 20:43:59 2002
From: slack@quackmaster.net (Jack Neely)
Date: Thu, 3 Oct 2002 15:43:59 -0400
Subject: [OpenAFS] locking in OpenAFS 1.2.6 and 1.2.7
Message-ID: <20021003154359.K1217@anduril.pams.ncsu.edu>
--NzB8fVQJ5HfG6fxh
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Hi,
I've noticed some differences is the locking behavior between 1.2.6 and
1.2.7 using simple fcntl() locking. 1.2.6 had a patch that seemed to
fix previous locking problems and it passed my locking tests. 1.2.7 has
the same patch but my locking tests do not aquire a lock.
My locking test is attached. It's pretty much verbatim out of the
openafs bug tracking system.
Does any one know what's up here?
Thanks
Jack Neely
--
Jack Neely
Linux Realm Kit Administration and Development
PAMS Computer Operations at NC State University
GPG Fingerprint: 1917 5AC1 E828 9337 7AA4 EA6B 213B 765F 3B6A 5B89
--NzB8fVQJ5HfG6fxh
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename="locking_test.c"
#include
#include
#include
#define LOCKTYPE(t) (((t) == F_RDLCK) ? "read" : "write")
main(argc, argv)
int argc;
char *argv[];
{
int fd;
int i;
printf("My pid=%d\n", getpid());
for (i=1; i
Message-ID:
On 3 Oct 2002, Adler, Stephen wrote:
> How does the openafs module resolve the sys_call_table symbol? I've
> built a version of the redhat 8.0 kernel which "exports" this
> symbol so that it is now in /proc/ksyms. Yet when I went to load
> the module, I got the unresolved symbol error again. Is there
did you rebuild the module against the new kernel headers?
> like some kind of command I need to execute which updates some
> symbol table file somewhere? Or did my attempts at building
> a kernel which exports sys_call_table fail?
From warlord@MIT.EDU Thu Oct 3 20:50:21 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 03 Oct 2002 15:50:21 -0400
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To: <1033674049.7281.25.camel@newadler.phy.bnl.gov>
References: <1033674049.7281.25.camel@newadler.phy.bnl.gov>
Message-ID:
Well, after you rebuilt the kernel did you rebuild the AFS Module?
-derek
"Adler, Stephen" writes:
> How does the openafs module resolve the sys_call_table symbol? I've
> built a version of the redhat 8.0 kernel which "exports" this
> symbol so that it is now in /proc/ksyms. Yet when I went to load
> the module, I got the unresolved symbol error again. Is there
> like some kind of command I need to execute which updates some
> symbol table file somewhere? Or did my attempts at building
> a kernel which exports sys_call_table fail?
>
> Steve.
>
>
>
> [adler@newadler adler]$ cat /etc/redhat-release
> Red Hat Linux release 8.0 (Psyche)
> [adler@newadler adler]$ grep sys_call_table /proc/ksyms
> c0320d74 sys_call_table_Rsmp_dfdb18bd
>
>
> _______________________________________________
> 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 adler@bnl.gov Thu Oct 3 21:02:19 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 16:02:19 -0400
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To:
References: <1033674049.7281.25.camel@newadler.phy.bnl.gov>
Message-ID: <1033675339.7281.27.camel@newadler.phy.bnl.gov>
The answer is yes. I did rebuild the afs modules.
On Thu, 2002-10-03 at 15:50, Derek Atkins wrote:
> Well, after you rebuilt the kernel did you rebuild the AFS Module?
>
> -derek
>
> "Adler, Stephen" writes:
>
> > How does the openafs module resolve the sys_call_table symbol? I've
> > built a version of the redhat 8.0 kernel which "exports" this
> > symbol so that it is now in /proc/ksyms. Yet when I went to load
> > the module, I got the unresolved symbol error again. Is there
> > like some kind of command I need to execute which updates some
> > symbol table file somewhere? Or did my attempts at building
> > a kernel which exports sys_call_table fail?
> >
> > Steve.
> >
> >
> >
> > [adler@newadler adler]$ cat /etc/redhat-release
> > Red Hat Linux release 8.0 (Psyche)
> > [adler@newadler adler]$ grep sys_call_table /proc/ksyms
> > c0320d74 sys_call_table_Rsmp_dfdb18bd
> >
> >
> > _______________________________________________
> > 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 warlord@MIT.EDU Thu Oct 3 21:03:59 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 03 Oct 2002 16:03:59 -0400
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To: <1033675339.7281.27.camel@newadler.phy.bnl.gov>
References: <1033674049.7281.25.camel@newadler.phy.bnl.gov>
<1033675339.7281.27.camel@newadler.phy.bnl.gov>
Message-ID:
grep sys_call_table /proc/ksyms
nm libafs.o | grep sys_call_table
-derek
"Adler, Stephen" writes:
> The answer is yes. I did rebuild the afs modules.
>
> On Thu, 2002-10-03 at 15:50, Derek Atkins wrote:
> > Well, after you rebuilt the kernel did you rebuild the AFS Module?
> >
> > -derek
> >
> > "Adler, Stephen" writes:
> >
> > > How does the openafs module resolve the sys_call_table symbol? I've
> > > built a version of the redhat 8.0 kernel which "exports" this
> > > symbol so that it is now in /proc/ksyms. Yet when I went to load
> > > the module, I got the unresolved symbol error again. Is there
> > > like some kind of command I need to execute which updates some
> > > symbol table file somewhere? Or did my attempts at building
> > > a kernel which exports sys_call_table fail?
> > >
> > > Steve.
> > >
> > >
> > >
> > > [adler@newadler adler]$ cat /etc/redhat-release
> > > Red Hat Linux release 8.0 (Psyche)
> > > [adler@newadler adler]$ grep sys_call_table /proc/ksyms
> > > c0320d74 sys_call_table_Rsmp_dfdb18bd
> > >
> > >
> > > _______________________________________________
> > > 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
>
--
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 adler@bnl.gov Thu Oct 3 21:07:31 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 16:07:31 -0400
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To:
References: <1033674049.7281.25.camel@newadler.phy.bnl.gov>
<1033675339.7281.27.camel@newadler.phy.bnl.gov>
Message-ID: <1033675651.7281.29.camel@newadler.phy.bnl.gov>
[root@newadler modload]# grep sys_call_table /proc/ksyms
c0320d74 sys_call_table_Rsmp_dfdb18bd
[root@newadler modload]# nm libafs-2.4.18-14-athlon.mp.o | grep
sys_call_table
U sys_call_table
is it becuase ksyms has the appended _Rsmp_dfdb18bd
that libafs module cannot find sys_call_table?
On Thu, 2002-10-03 at 16:03, Derek Atkins wrote:
> grep sys_call_table /proc/ksyms
> nm libafs.o | grep sys_call_table
>
> -derek
>
> "Adler, Stephen" writes:
>
> > The answer is yes. I did rebuild the afs modules.
> >
> > On Thu, 2002-10-03 at 15:50, Derek Atkins wrote:
> > > Well, after you rebuilt the kernel did you rebuild the AFS Module?
> > >
> > > -derek
> > >
> > > "Adler, Stephen" writes:
> > >
> > > > How does the openafs module resolve the sys_call_table symbol? I've
> > > > built a version of the redhat 8.0 kernel which "exports" this
> > > > symbol so that it is now in /proc/ksyms. Yet when I went to load
> > > > the module, I got the unresolved symbol error again. Is there
> > > > like some kind of command I need to execute which updates some
> > > > symbol table file somewhere? Or did my attempts at building
> > > > a kernel which exports sys_call_table fail?
> > > >
> > > > Steve.
> > > >
> > > >
> > > >
> > > > [adler@newadler adler]$ cat /etc/redhat-release
> > > > Red Hat Linux release 8.0 (Psyche)
> > > > [adler@newadler adler]$ grep sys_call_table /proc/ksyms
> > > > c0320d74 sys_call_table_Rsmp_dfdb18bd
> > > >
> > > >
> > > > _______________________________________________
> > > > 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
> >
>
> --
> 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 warlord@MIT.EDU Thu Oct 3 21:09:05 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 03 Oct 2002 16:09:05 -0400
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To: <1033675651.7281.29.camel@newadler.phy.bnl.gov>
References: <1033674049.7281.25.camel@newadler.phy.bnl.gov>
<1033675339.7281.27.camel@newadler.phy.bnl.gov>
<1033675651.7281.29.camel@newadler.phy.bnl.gov>
Message-ID:
Yea.
did you "make mrproper" before you recompiled your kernel?
-derek
"Adler, Stephen" writes:
> [root@newadler modload]# grep sys_call_table /proc/ksyms
> c0320d74 sys_call_table_Rsmp_dfdb18bd
> [root@newadler modload]# nm libafs-2.4.18-14-athlon.mp.o | grep
> sys_call_table
> U sys_call_table
>
> is it becuase ksyms has the appended _Rsmp_dfdb18bd
> that libafs module cannot find sys_call_table?
>
> On Thu, 2002-10-03 at 16:03, Derek Atkins wrote:
> > grep sys_call_table /proc/ksyms
> > nm libafs.o | grep sys_call_table
> >
> > -derek
> >
> > "Adler, Stephen" writes:
> >
> > > The answer is yes. I did rebuild the afs modules.
> > >
> > > On Thu, 2002-10-03 at 15:50, Derek Atkins wrote:
> > > > Well, after you rebuilt the kernel did you rebuild the AFS Module?
> > > >
> > > > -derek
> > > >
> > > > "Adler, Stephen" writes:
> > > >
> > > > > How does the openafs module resolve the sys_call_table symbol? I've
> > > > > built a version of the redhat 8.0 kernel which "exports" this
> > > > > symbol so that it is now in /proc/ksyms. Yet when I went to load
> > > > > the module, I got the unresolved symbol error again. Is there
> > > > > like some kind of command I need to execute which updates some
> > > > > symbol table file somewhere? Or did my attempts at building
> > > > > a kernel which exports sys_call_table fail?
> > > > >
> > > > > Steve.
> > > > >
> > > > >
> > > > >
> > > > > [adler@newadler adler]$ cat /etc/redhat-release
> > > > > Red Hat Linux release 8.0 (Psyche)
> > > > > [adler@newadler adler]$ grep sys_call_table /proc/ksyms
> > > > > c0320d74 sys_call_table_Rsmp_dfdb18bd
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > 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
> > >
> >
> > --
> > 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
>
--
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 shadow@dementia.org Thu Oct 3 21:09:02 2002
From: shadow@dementia.org (Derrick J Brashear)
Date: Thu, 3 Oct 2002 16:09:02 -0400 (EDT)
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To: <1033675651.7281.29.camel@newadler.phy.bnl.gov>
Message-ID:
On 3 Oct 2002, Adler, Stephen wrote:
> [root@newadler modload]# grep sys_call_table /proc/ksyms
> c0320d74 sys_call_table_Rsmp_dfdb18bd
> [root@newadler modload]# nm libafs-2.4.18-14-athlon.mp.o | grep
> sys_call_table
> U sys_call_table
>
> is it becuase ksyms has the appended _Rsmp_dfdb18bd
ksyms didn't; the kernel actually uses that name
> that libafs module cannot find sys_call_table?
Well, yes, but if you built against the right headers the module should
have the correct name, instead of the unadorned name
From williams@cs.ucr.edu Thu Oct 3 21:07:07 2002
From: williams@cs.ucr.edu (David Williams)
Date: Thu, 03 Oct 2002 13:07:07 -0700
Subject: [OpenAFS] mandrake 9.0 segmentation fault
References: <1033670164.3973.4.camel@localhost.localdomain>
Message-ID: <3D9CA36B.5020108@cs.ucr.edu>
I had a similar problem with mdk8.2 and afs 1.2.5. Replacing the
mdksecure kernel with a stock one(plus xfs) fixed it for me.
david
trax wrote:
> Okay, I have managed to get mdk9.0 to compile, upoun starting afs, I see
>
> [root@localhost root]# service afs start
> Found libafs-2.4.19-16mdksecure-i686.o from SymTable... Loading...
> Starting AFS services.....
>
> /var/log/messages shows ...
> Oct 3 20:33:28 localhost su(pam_unix)[4100]: session opened for user
> root by (uid=500)
> Oct 3 20:34:08 localhost kernel: sock_release(rx_socket) FIXME
> Oct 3 20:34:08 localhost kernel: AFS: RX failed to initialize.
>
> The it just sits there. Am I missing something here?
>
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
>
From adler@bnl.gov Thu Oct 3 21:14:38 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 16:14:38 -0400
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To:
References:
Message-ID: <1033676078.26364.32.camel@newadler.phy.bnl.gov>
sorry for the stupid questions guys!!! where do I find the
headers? I'm trying to figure out if I need to install a
separate headers rpm or the kernel source rpm.... Should
they be in /usr/src/linux-2.4?
On Thu, 2002-10-03 at 16:09, Derrick J Brashear wrote:
> On 3 Oct 2002, Adler, Stephen wrote:
>
> > [root@newadler modload]# grep sys_call_table /proc/ksyms
> > c0320d74 sys_call_table_Rsmp_dfdb18bd
> > [root@newadler modload]# nm libafs-2.4.18-14-athlon.mp.o | grep
> > sys_call_table
> > U sys_call_table
> >
> > is it becuase ksyms has the appended _Rsmp_dfdb18bd
>
> ksyms didn't; the kernel actually uses that name
>
> > that libafs module cannot find sys_call_table?
>
> Well, yes, but if you built against the right headers the module should
> have the correct name, instead of the unadorned name
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
From shadow@dementia.org Thu Oct 3 21:16:50 2002
From: shadow@dementia.org (Derrick J Brashear)
Date: Thu, 3 Oct 2002 16:16:50 -0400 (EDT)
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To: <1033676078.26364.32.camel@newadler.phy.bnl.gov>
Message-ID:
On 3 Oct 2002, Adler, Stephen wrote:
> sorry for the stupid questions guys!!! where do I find the
> headers? I'm trying to figure out if I need to install a
> separate headers rpm or the kernel source rpm.... Should
> they be in /usr/src/linux-2.4?
they're in the kernel you built. where did you build your kernel?
give that directory as an argument to configure's
--with-linux-kernel-headers=
From shadow@dementia.org Thu Oct 3 21:31:45 2002
From: shadow@dementia.org (Derrick J Brashear)
Date: Thu, 3 Oct 2002 16:31:45 -0400 (EDT)
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To: <1033676537.7281.35.camel@newadler.phy.bnl.gov>
Message-ID:
you should reply to the list, not me
On 3 Oct 2002, Adler, Stephen wrote:
> I believe you answered the question indirectly. I've installed
> the newly built kernel source rpm and rebuilding afs. Lets see
> if this fixes it... Is there a particular file within the
> source tree that I should look for which tells me if this build
> will pickup the full symbol for sys_call_table?
somewhere in include/linux/modules; there are lots of .ver files. grep for
sys_call_table in that directory
From kolya@MIT.EDU Thu Oct 3 21:30:27 2002
From: kolya@MIT.EDU (Nickolai Zeldovich)
Date: Thu, 03 Oct 2002 16:30:27 -0400
Subject: [OpenAFS] locking in OpenAFS 1.2.6 and 1.2.7
Message-ID: <200210032030.QAA03592@contents-vnder-pressvre.mit.edu>
> I've noticed some differences is the locking behavior between 1.2.6 and
> 1.2.7 using simple fcntl() locking. 1.2.6 had a patch that seemed to
> fix previous locking problems and it passed my locking tests. 1.2.7 has
> the same patch but my locking tests do not aquire a lock.
Can you be a little more verbose about your problem? The program you
attached simply displays the lock status of given files, and appears
to do so correctly for me on a sun4x_56 running OpenAFS 1.2.7, as well
as a Linux machine running OpenAFS cvs-head.
-- kolya
From adler@bnl.gov Thu Oct 3 21:50:32 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 16:50:32 -0400
Subject: [OpenAFS] OMG it works!!!! (i.e. openafs 1.2.7 on redhat 8.0)
Message-ID: <1033678232.8446.2.camel@newadler.phy.bnl.gov>
Well... after a lot of back and forth with the afs guys, its done!!! I
got a special version of the red hat 8.0 kernel built along with
open afs 1.2.7. I'll put my work on my ftp site soon so that others
can use this work around for now. (will follow another post to this
e-mail list when the rpms are available for upload.)
Cheers to openafs and open source!
Steve.
From shadow@dementia.org Thu Oct 3 21:57:29 2002
From: shadow@dementia.org (Derrick J Brashear)
Date: Thu, 3 Oct 2002 16:57:29 -0400 (EDT)
Subject: [OpenAFS] workaround for RedHat 8
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.
Send mail to mime@docserver.cac.washington.edu for more info.
--42002020-1449831134-1033678649=:2629
Content-Type: TEXT/PLAIN; charset=US-ASCII
Chaskiel Grundman wrote this patch, which basically figures out where the
sys_call_table is and uses it anyway.
It's evil. Use it if you like. I'm waiting for the first objector (to
this method) so I can laugh at them. If you provide your phone number I'll
be happy to actually call you and laugh at you for a while.
--42002020-1449831134-1033678649=:2629
Content-Type: TEXT/PLAIN; charset=US-ASCII; name="rh8-stable.diff"
Content-Transfer-Encoding: BASE64
Content-ID:
Content-Description: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--42002020-1449831134-1033678649=:2629--
From adler@bnl.gov Thu Oct 3 22:14:15 2002
From: adler@bnl.gov (Adler, Stephen)
Date: 03 Oct 2002 17:14:15 -0400
Subject: [OpenAFS] red hat 8.0/openafs ftp site open
Message-ID: <1033679655.8446.7.camel@newadler.phy.bnl.gov>
Guys,
I've uploaded my work to
ftp://ftp.phy.bnl.gov/pub/openafs-rh80
There you'll find a red hat 8.0 kernel with sys_call_table exported
as well as my build of openafs 1.2.7. Only the athlon kernel build
is there right now. I'm building the i586 and i686 soon followed
by the i386 kernel build. Those will be uploaded by tomorrow AM.
I hope this work is useful to others in the openafs community.
Cheers. Steve Adler.
From warlord@MIT.EDU Thu Oct 3 23:50:35 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 03 Oct 2002 18:50:35 -0400
Subject: [OpenAFS] mandrake 9.0 segmentation fault
In-Reply-To: <1033670164.3973.4.camel@localhost.localdomain>
References: <1033670164.3973.4.camel@localhost.localdomain>
Message-ID:
Sounds like mandrake is adding some new kernel item that we (openafs)
need to initialize..
-derek
trax writes:
> Okay, I have managed to get mdk9.0 to compile, upoun starting afs, I see
>
> [root@localhost root]# service afs start
> Found libafs-2.4.19-16mdksecure-i686.o from SymTable... Loading...
> Starting AFS services.....
>
> /var/log/messages shows ...
> Oct 3 20:33:28 localhost su(pam_unix)[4100]: session opened for user
> root by (uid=500)
> Oct 3 20:34:08 localhost kernel: sock_release(rx_socket) FIXME
> Oct 3 20:34:08 localhost kernel: AFS: RX failed to initialize.
>
> The it just sits there. Am I missing something here?
>
>
> _______________________________________________
> 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 excds@kth.se Fri Oct 4 00:13:17 2002
From: excds@kth.se (Daniel =?ISO-8859-1?Q?Sw=E4rd?=)
Date: 04 Oct 2002 01:13:17 +0200
Subject: [OpenAFS] sys_call_table symbol
In-Reply-To: <1033676078.26364.32.camel@newadler.phy.bnl.gov>
References:
<1033676078.26364.32.camel@newadler.phy.bnl.gov>
Message-ID: <1033686797.1138.2.camel@hybris>
On Thu, 2002-10-03 at 22:14, Adler, Stephen wrote:
> sorry for the stupid questions guys!!!
"Now remember Kyle, there are no stupid questions. Only stupid people."
Sorry... Just had to... ;-)
Anyway, I believe I'm in the lead of stupid questions here... ;-))
/Daniel, too tired to exist.
From jds@soltis.cc Fri Oct 4 02:52:00 2002
From: jds@soltis.cc (Jesus Delgado)
Date: Thu, 3 Oct 2002 19:52:00 -0600
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
In-Reply-To: <1033649961.15481.4.camel@newadler.phy.bnl.gov>
References: <20021002151128.M26033@soltis.cc> <20021002195332.M26176@soltis.cc> <1033649961.15481.4.camel@newadler.phy.bnl.gov>
Message-ID: <20021003195200.M44025@soltis.cc>
Thanks, perfect the build Ok Rethat 8.0 kernel 2.4.19 openafs-1.2.7 OK
Muchas Gracias
> You got this because of red hat going to autoconf 2.53. I aliased
> autoconf to autoconf-2.13 which fixed the build.
>
> Steve.
>
> On Wed, 2002-10-02 at 21:53, Jesus Delgado wrote:
> >
> > ---------- Forwarded Message -----------
> > Hi:
> >
> > Problems when try the build openafs source rpm in redhat 8.0
> > the errors is the same with openafs-1.2.6 y openafs-1.2.7:
> >
> > Errors in openafs-1.2.7
> >
> > .4.1eq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1
rpmlib(CompressedFileNames) <=
> > 3.0.4-1
> > Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> > rpmlib(CompressedFileNames) <= 3.0.4-1
> > Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
> > libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
> > Processing files: openafs-krb5-1.2.7-rh7.3.1
> > error: File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
> > error: File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
> > Requires: openafs = 1.2.7
> >
> > RPM build errors:
> > File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
> > File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
> >
> > Errors with openafs-1.2.6
> >
> > Requires: openafs = 1.2.6 openafs-client = 1.2.6
> > Obsoletes: openafs-client-compat
> > Processing files: openafs-kpasswd-1.2.6-rh7.3.1
> > Finding Provides: /usr/lib/rpm/find-provides
> > Finding Requires: /usr/lib/rpm/find-requires
> > PreReq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1 rpmlib(CompressedFileNames) <=
> > 3.0.4-1
> > Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> > rpmlib(CompressedFileNames) <= 3.0.4-1
> > Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
> > libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
> > Processing files: openafs-krb5-1.2.6-rh7.3.1
> > error: File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
> > error: File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
> > Requires: openafs = 1.2.6
> >
> > RPM build errors:
> > File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
> > File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
> >
> > Help me plase
> >
> > Regards.
> > ------- End of Forwarded Message -------
> >
> >
> >
> > _______________________________________________
> > OpenAFS-info mailing list
> > OpenAFS-info@openafs.org
> > https://lists.openafs.org/mailman/listinfo/openafs-info
From elizabeth@linuxbox.nu Fri Oct 4 02:46:12 2002
From: elizabeth@linuxbox.nu (Elizabeth Ziph)
Date: 03 Oct 2002 21:46:12 -0400
Subject: [OpenAFS] AFS Training ?
In-Reply-To:
References:
Message-ID: <1033695973.25528.528.camel@apollo>
if the class is on, we might send one person.
elizabeth
On Thu, 2002-10-03 at 01:41, Mitch Collinsworth wrote:
>
> Following up to my own followup: There is currently only one person
> registered for the December class. One of my group would also like
> to attend, making a total of 2 so far. We couldn't get an exact
> answer from IBM as to what constitutes sufficient critical mass to
> be sure the class won't be cancelled, but it seems to be in the
> neighborhood of 4 or 5. Are there others here who might be interested
> in this? Write to me off-list and I can give you the address of the
> teacher if you want more direct info. My selfish motivation is in
> recruiting enough additional attendees to see the class not get
> canceled again. If you'd like to go but the December dates are bad,
> when is better? Maybe we can work with them in scheduling it for a
> time when we can succeed in achieving critical mass.
>
> It's been over a decade since I took this class, but it was a good
> class then and would be quite useful today even if little has changed
> in the presentation since then.
>
> -Mitch
>
>
> On Sun, 29 Sep 2002, Mitch Collinsworth wrote:
>
> > External web site is:
> > http://www7b.software.ibm.com/wsdd/education/enablement/curriculum/sw800.html
> >
> > A word of caution: Just because it's listed here doesn't mean it will
> > actually be held. The last class that was scheduled for Pittsburgh was
> > cancelled at the last minute due to insufficient number of registrants.
> >
> > There was supposed to be another one this fall in California (San Jose?).
> > It no longer seems to be on the schedule. I take it it was cancelled, too.
> >
> > -Mitch
> >
> >
> > On Sun, 29 Sep 2002, Daniel Clark/Cambridge/IBM wrote:
> >
> > > Paul Blackburn wrote:
> > > > Transarc used to run some excellent AFS administrator courses.
> > > > Does anyone know of AFS training available today?
> > >
> > > According to w3.education.ibm.com (internal IBM site, there is some
> > > external equivalent but I don't know what it is) Transarc - now "IBM
> > > Pittsburgh Lab" - still runs the "AFS Administration" course. It's course
> > > code SW800 and will next occur in Pittsburgh, PA 2002-12-10 to 2002-12-13.
> > > Below are the full details, which include a contact number/email to get
> > > more info and future course dates. I have taken this class and can vouch
> > > that it is excellent. My class was taught by someone who had been
> > > administering AFS since it was a research project at CMU, so in addition to
> > > access to deep technical knowledge there were also interesting historical
> > > asides.
> > >
> > > Course Details: AFS Administration
> > >
> > > 2002-12-10 to 2002-12-13
> > >
> > > Course Code:SW800
> > > Section:K8II
> > >
> > > Course Location:
> > > Room: TBA 2
> > > IBM Pittsburgh Lab 9Fl
> > > 11 Stanwix Street
> > > Pittsburgh
> > > United States
> > > 15222
> > >
> > > Course Contact:
> > > Tracy Linza
> > > linza at IBMUS
> > > 412-667-4477
> > > TL 989-4477
> > >
> > > Course Schedule:
> > > 2002-12-10 to 2002-12-13
> > > 09:00:00 to 18:00:00
> > > Last Day End Time: 17:30:00
> > >
> > > Last Cancellation Date without penalty:
> > > 2002-11-25
> > >
> > >
> > > Equipment Required:
> > > None
> > >
> > > Course Enrollments:
> > > Total Students Confirmed: 1 of 14
> > > Total on Standby: 0
> > >
> > > Additional Course Information:
> > > None
> > >
> > > Comments:
> > > None
> > >
> > > Location Specific Info:
> > >
> > > LOCATION:
> > > IBM Pittsburgh Lab
> > > 11 Stanwix Street
> > > 9th FL
> > > Pittsburgh, PA 15222
> > >
> > > CONTACT:
> > > 412-667-4433
> > > TL 989-4433
> > >
> > > BADGES:
> > > See lobby receptionist on 9th Floor
> > > All visitors are required to sign-in
> > > (8:00am - 6:00pm)
> > >
> > > LOCAL HOTEL INFORMATION
> > >
> > > Pittsburgh Hilton
> > > 600 Commonwealth Place
> > > Gateway Center
> > > Pittsburgh, PA 15222
> > > 412-391-4600
> > > Ask for IBM/Transarc rate of $92 per night
> > > (Subject to change without notice)
> > > 5 minute walk to Lab
> > >
> > > Sheraton Hotel
> > > Carson & SMithfield Streets
> > > Pittsburgh, PA 15219
> > > 412-261-2000
> > > 5-10 minute subway ride to Lab
> > > Ask for IBM/Transarc rate of $116 per night
> > > (Subject to change without notice)
> > >
> > >
> > > DIRECTIONS
> > >
> > > FROM AIRPORT:
> > > * From Airport follow the signs toward Pittsburgh
> > > (Approx 15 miles) to the Fort Pitt Tunnel
> > > * Stay in the left lane.
> > > * After the tunnel, you will be on the Fort Pitt Bridge
> > > * Take the Boulevard of the Allies Exit
> > > * Proceed to Stanwix St and turn right.
> > > * Proceed one block to 11 Stanwix
> > > (Look for Blattner Bruner revolving sign)
> > > * Parking garage is in the building
> > > * Parking in the garage is $14 per day - student expense.
> > >
> > > --
> > > Daniel Clark # Sys Admin & Release Engineer
> > > IBM > Lotus > Messaging Technology Group
> > >
> > > _______________________________________________
> > > 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
> >
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
--
elizabeth ziph
734.761.4689
www.linuxbox.nu
From security@xauth.net Fri Oct 4 05:33:16 2002
From: security@xauth.net (Charles Clancy)
Date: Thu, 3 Oct 2002 23:33:16 -0500 (CDT)
Subject: [OpenAFS] using openafs over wireless
In-Reply-To:
Message-ID:
> Anyone have experience running open-afs on win2k over a wireless nic? we
> expected lan throughput to be 2-4 mbs.
I've done it with RH71, kernel 2.4.9, with a Prism2-based card. Wireless
is no different than wireline -- just slower. I wouldn't want to run my
homedir from AFS over 802.11, but for casual access it's just fine.
For that matter, I use OpenAFS over my 768/128K ADSL connection from home.
Again, it's fine for casual access, but I wouldn't want to run
applications from it.
[ t charles clancy ]--[ tclancy@uiuc.edu ]--[ www.uiuc.edu/~tclancy ]
From abuechle@fhzh.ch Fri Oct 4 08:21:43 2002
From: abuechle@fhzh.ch (Andreas Buechler)
Date: Fri, 4 Oct 2002 09:21:43 +0200
Subject: [OpenAFS] pam and openafs 1.2.7 for RH 7.2
In-Reply-To:
References: <15770.49260.4762.561962@hszpc38.isz.ch>
Message-ID: <15773.16775.746446.362012@hszpc38.isz.ch>
> Try removing that option, and see if it works then.
Unfortunately that doesn't have any impact, error messages in
/var/log/messages still look the same after an attempt to login in.
I really don't have a clou what's going on. Because I tried the same
pam.d configuration on an Intel machine (with SuSE Linux 7.3,
openafs1.2.3) and it works as expected! Do you have any ideas what
else I could try out?
Andi
--
From turbo@bayour.com Fri Oct 4 09:07:46 2002
From: turbo@bayour.com (Turbo Fredriksson)
Date: 04 Oct 2002 10:07:46 +0200
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
Message-ID: <87ofaaeiel.fsf@papadoc.bayour.com>
I have a number of machines at home which I can log in to.
If I login on multiple machines, the previous host looses
connection to the AFS file server.
Example:
1. Login in on host1. Homedir on AFS.
Works fine.
2. Login in on host2. Homedir on AFS.
Works fine - BUT: host1 looses the connection
to the file server!
The firewall is a Linux machine, doing NAT (masquerading).
Is this supposed to happen?
From tino.schwarze@informatik.tu-chemnitz.de Fri Oct 4 10:04:30 2002
From: tino.schwarze@informatik.tu-chemnitz.de (Tino Schwarze)
Date: Fri, 4 Oct 2002 11:04:30 +0200
Subject: [OpenAFS] using openafs over wireless
In-Reply-To: ; from security@xauth.net on Thu, Oct 03, 2002 at 11:33:16PM -0500
References:
Message-ID: <20021004110430.A23960@informatik.tu-chemnitz.de>
On Thu, Oct 03, 2002 at 11:33:16PM -0500, Charles Clancy wrote:
> > Anyone have experience running open-afs on win2k over a wireless nic? we
> > expected lan throughput to be 2-4 mbs.
>
> I've done it with RH71, kernel 2.4.9, with a Prism2-based card. Wireless
> is no different than wireline -- just slower. I wouldn't want to run my
> homedir from AFS over 802.11, but for casual access it's just fine.
>
> For that matter, I use OpenAFS over my 768/128K ADSL connection from home.
> Again, it's fine for casual access, but I wouldn't want to run
> applications from it.
Some time ago, I've been developing a little app and half the libraries
(GTK2, Pango et al.) were built in AFS space. Since I was too lazy to
recompile at home, I just used them where they were. The AFS caching and
on demand paging kicked ass.
All went over a single ISDN line and I was able to observe that running
a program accesses different parts of a library than debugging. B-)
Bye, Tino.
--
* LINUX - Where do you want to be tomorrow? *
http://www.tu-chemnitz.de/linux/tag/
From andrei@caspur.it Fri Oct 4 12:55:59 2002
From: andrei@caspur.it (Andrei Maslennikov)
Date: Fri, 4 Oct 2002 11:55:59 +0000 ( )
Subject: [OpenAFS] rh8.0 kernel RPMs
In-Reply-To: <20021004110430.A23960@informatik.tu-chemnitz.de>
Message-ID:
If somebody wants to give them a try, there's a set of kernel RPMs with
sys_call_table re-exported (I've built them yesterday; libafs 1.2.7 does
load with these):
# ls /afs/italia/project/afs/rh80
kernel-2.4.18-14.athlon.rpm kernel-doc-2.4.18-14.i386.rpm
kernel-2.4.18-14.i586.rpm kernel-smp-2.4.18-14.athlon.rpm
kernel-2.4.18-14.i686.rpm kernel-smp-2.4.18-14.i686.rpm
kernel-BOOT-2.4.18-14.i386.rpm kernel-source-2.4.18-14.i386.rpm
kernel-bigmem-2.4.18-14.i686.rpm kernel-uml-2.4.18-14.i686.rpm
kernel-debug-2.4.18-14.i686.rpm
Or, alternatively (same thing):
http://afs.caspur.it/afs/italia/project/afs/rh80/
Andrei.
----------------------------------
>italia # Italian Public AFS Cell
193.204.5.9 #afs.caspur.it
From adler@bnl.gov Fri Oct 4 13:04:18 2002
From: adler@bnl.gov (Stephen Adler)
Date: 04 Oct 2002 08:04:18 -0400
Subject: [OpenAFS] rh8.0 kernel RPMs
In-Reply-To:
References:
Message-ID: <1033733059.12281.1.camel@dhcp-21-1>
And here is another source which I built.... oh well, I guess
its better to have 2 independent builds rather than 0...
http://ssadler.phy.bnl.gov/adler/openafs-rh80/
Cheers. Steve.
On Fri, 2002-10-04 at 07:55, Andrei Maslennikov wrote:
>
> If somebody wants to give them a try, there's a set of kernel RPMs with
> sys_call_table re-exported (I've built them yesterday; libafs 1.2.7 does
> load with these):
>
> # ls /afs/italia/project/afs/rh80
> kernel-2.4.18-14.athlon.rpm kernel-doc-2.4.18-14.i386.rpm
> kernel-2.4.18-14.i586.rpm kernel-smp-2.4.18-14.athlon.rpm
> kernel-2.4.18-14.i686.rpm kernel-smp-2.4.18-14.i686.rpm
> kernel-BOOT-2.4.18-14.i386.rpm kernel-source-2.4.18-14.i386.rpm
> kernel-bigmem-2.4.18-14.i686.rpm kernel-uml-2.4.18-14.i686.rpm
> kernel-debug-2.4.18-14.i686.rpm
>
> Or, alternatively (same thing):
> http://afs.caspur.it/afs/italia/project/afs/rh80/
>
> Andrei.
>
> ----------------------------------
> >italia # Italian Public AFS Cell
> 193.204.5.9 #afs.caspur.it
>
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
From warlord@MIT.EDU Fri Oct 4 14:43:13 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 04 Oct 2002 09:43:13 -0400
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To: <87ofaaeiel.fsf@papadoc.bayour.com>
References: <87ofaaeiel.fsf@papadoc.bayour.com>
Message-ID:
How long are your UDP timeouts?
-derek
Turbo Fredriksson writes:
> I have a number of machines at home which I can log in to.
>
> If I login on multiple machines, the previous host looses
> connection to the AFS file server.
>
> Example:
> 1. Login in on host1. Homedir on AFS.
> Works fine.
> 2. Login in on host2. Homedir on AFS.
> Works fine - BUT: host1 looses the connection
> to the file server!
>
>
> The firewall is a Linux machine, doing NAT (masquerading).
>
> Is this supposed to happen?
> _______________________________________________
> 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 mitch@ccmr.cornell.edu Fri Oct 4 14:47:21 2002
From: mitch@ccmr.cornell.edu (Mitch Collinsworth)
Date: Fri, 4 Oct 2002 09:47:21 -0400 (EDT)
Subject: [OpenAFS] AFS Training ?
In-Reply-To: <1033695973.25528.528.camel@apollo>
Message-ID:
This is exactly why I wrote the previous message. The class _will_ be
on if enough people sign up, but if everyone waits to register until
they know the class is on, the class won't happen. If you have someone
who wants to take the class, please have them sign up and be counted.
If the critical mass is "4 or 5", and we're at 2 today, your person
plus the other one who contacted me yesterday off-list gets us to 4,
then we're either there now, or in need of just one more.
-Mitch
On 3 Oct 2002, Elizabeth Ziph wrote:
> if the class is on, we might send one person.
>
> elizabeth
>
>
> On Thu, 2002-10-03 at 01:41, Mitch Collinsworth wrote:
> >
> > Following up to my own followup: There is currently only one person
> > registered for the December class. One of my group would also like
> > to attend, making a total of 2 so far. We couldn't get an exact
> > answer from IBM as to what constitutes sufficient critical mass to
> > be sure the class won't be cancelled, but it seems to be in the
> > neighborhood of 4 or 5. Are there others here who might be interested
> > in this? Write to me off-list and I can give you the address of the
> > teacher if you want more direct info. My selfish motivation is in
> > recruiting enough additional attendees to see the class not get
> > canceled again. If you'd like to go but the December dates are bad,
> > when is better? Maybe we can work with them in scheduling it for a
> > time when we can succeed in achieving critical mass.
> >
> > It's been over a decade since I took this class, but it was a good
> > class then and would be quite useful today even if little has changed
> > in the presentation since then.
> >
> > -Mitch
> >
> >
> > On Sun, 29 Sep 2002, Mitch Collinsworth wrote:
> >
> > > External web site is:
> > > http://www7b.software.ibm.com/wsdd/education/enablement/curriculum/sw800.html
> > >
> > > A word of caution: Just because it's listed here doesn't mean it will
> > > actually be held. The last class that was scheduled for Pittsburgh was
> > > cancelled at the last minute due to insufficient number of registrants.
> > >
> > > There was supposed to be another one this fall in California (San Jose?).
> > > It no longer seems to be on the schedule. I take it it was cancelled, too.
> > >
> > > -Mitch
> > >
> > >
> > > On Sun, 29 Sep 2002, Daniel Clark/Cambridge/IBM wrote:
> > >
> > > > Paul Blackburn wrote:
> > > > > Transarc used to run some excellent AFS administrator courses.
> > > > > Does anyone know of AFS training available today?
> > > >
> > > > According to w3.education.ibm.com (internal IBM site, there is some
> > > > external equivalent but I don't know what it is) Transarc - now "IBM
> > > > Pittsburgh Lab" - still runs the "AFS Administration" course. It's course
> > > > code SW800 and will next occur in Pittsburgh, PA 2002-12-10 to 2002-12-13.
> > > > Below are the full details, which include a contact number/email to get
> > > > more info and future course dates. I have taken this class and can vouch
> > > > that it is excellent. My class was taught by someone who had been
> > > > administering AFS since it was a research project at CMU, so in addition to
> > > > access to deep technical knowledge there were also interesting historical
> > > > asides.
> > > >
> > > > Course Details: AFS Administration
> > > >
> > > > 2002-12-10 to 2002-12-13
> > > >
> > > > Course Code:SW800
> > > > Section:K8II
> > > >
> > > > Course Location:
> > > > Room: TBA 2
> > > > IBM Pittsburgh Lab 9Fl
> > > > 11 Stanwix Street
> > > > Pittsburgh
> > > > United States
> > > > 15222
> > > >
> > > > Course Contact:
> > > > Tracy Linza
> > > > linza at IBMUS
> > > > 412-667-4477
> > > > TL 989-4477
> > > >
> > > > Course Schedule:
> > > > 2002-12-10 to 2002-12-13
> > > > 09:00:00 to 18:00:00
> > > > Last Day End Time: 17:30:00
> > > >
> > > > Last Cancellation Date without penalty:
> > > > 2002-11-25
> > > >
> > > >
> > > > Equipment Required:
> > > > None
> > > >
> > > > Course Enrollments:
> > > > Total Students Confirmed: 1 of 14
> > > > Total on Standby: 0
> > > >
> > > > Additional Course Information:
> > > > None
> > > >
> > > > Comments:
> > > > None
> > > >
> > > > Location Specific Info:
> > > >
> > > > LOCATION:
> > > > IBM Pittsburgh Lab
> > > > 11 Stanwix Street
> > > > 9th FL
> > > > Pittsburgh, PA 15222
> > > >
> > > > CONTACT:
> > > > 412-667-4433
> > > > TL 989-4433
> > > >
> > > > BADGES:
> > > > See lobby receptionist on 9th Floor
> > > > All visitors are required to sign-in
> > > > (8:00am - 6:00pm)
> > > >
> > > > LOCAL HOTEL INFORMATION
> > > >
> > > > Pittsburgh Hilton
> > > > 600 Commonwealth Place
> > > > Gateway Center
> > > > Pittsburgh, PA 15222
> > > > 412-391-4600
> > > > Ask for IBM/Transarc rate of $92 per night
> > > > (Subject to change without notice)
> > > > 5 minute walk to Lab
> > > >
> > > > Sheraton Hotel
> > > > Carson & SMithfield Streets
> > > > Pittsburgh, PA 15219
> > > > 412-261-2000
> > > > 5-10 minute subway ride to Lab
> > > > Ask for IBM/Transarc rate of $116 per night
> > > > (Subject to change without notice)
> > > >
> > > >
> > > > DIRECTIONS
> > > >
> > > > FROM AIRPORT:
> > > > * From Airport follow the signs toward Pittsburgh
> > > > (Approx 15 miles) to the Fort Pitt Tunnel
> > > > * Stay in the left lane.
> > > > * After the tunnel, you will be on the Fort Pitt Bridge
> > > > * Take the Boulevard of the Allies Exit
> > > > * Proceed to Stanwix St and turn right.
> > > > * Proceed one block to 11 Stanwix
> > > > (Look for Blattner Bruner revolving sign)
> > > > * Parking garage is in the building
> > > > * Parking in the garage is $14 per day - student expense.
> > > >
> > > > --
> > > > Daniel Clark # Sys Admin & Release Engineer
> > > > IBM > Lotus > Messaging Technology Group
> > > >
> > > > _______________________________________________
> > > > OpenAFS-info mailing list
> > > > OpenAFS-info@openafs.org
> > > > https://lists.openafs.org/mailman/listinfo/openafs-info
> > > >
> > >
> > > _______________________________________________
> > > OpenAFS-info mailing list
> > > OpenAFS-info@openafs.org
> > > https://lists.openafs.org/mailman/listinfo/openafs-info
> > >
> >
> > _______________________________________________
> > OpenAFS-info mailing list
> > OpenAFS-info@openafs.org
> > https://lists.openafs.org/mailman/listinfo/openafs-info
> --
> elizabeth ziph
> 734.761.4689
> www.linuxbox.nu
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
>
From klaas@northsailor.de Wed Oct 2 15:34:10 2002
From: klaas@northsailor.de (klaas hagemann)
Date: Wed, 2 Oct 2002 16:34:10 +0200
Subject: [OpenAFS] scripts for automatic installation + client cache questions
References: <003a01c26a0d$e1dd6fe0$2b03a8c0@mummert.priv>
Message-ID: <007301c26a20$c6774c40$2b03a8c0@mummert.priv>
----- Original Message -----
From: "Derek Atkins"
To: "Klaas Hagemann"
Cc:
Sent: Wednesday, October 02, 2002 3:17 PM
Subject: Re: [OpenAFS] scripts for automatic installation + client cache
questions
> "Klaas Hagemann" writes:
>
> > Hello,
> >
> > i am currently trying to write scripts for an automatic setup for
> > openafs-file-servers and dbms-servers.
> > Therefor i use the bos exec command quite a lot.
> > But when adding a new server, i have to restart all servers.
> > So i want a "bos exec host "/etc/init.d/afs restart"
> > But it ends up with a communication failure. Does someone have an idea
for a
> > workaround for this?
>
> No, because the bosserver get's killed before the command completes.
> Is there any particular reason you don't use "bos
restart -all -bosserver"?
Thankys, i have not known this option yet. Thanks a lott, i think that is
exactly for what i was looking for.
Yes, RTFM sometimes help.....
>
> > Then i have got another question:
> > what are the advances for an own partition for the client on
> > /usr/vice/cache?
> > Currently, i do not have a seperate partition for it and it works good.
>
> Basically, if your cache partition fills, AFS is screwed. The benefit
> of having your own AFS cache partition is that if only AFS uses the
> partition you are guaranteed that it will never fill to capacity. If
> you share the cache partition, then you have to always worry about how
> much space is left, because AFS will crash ungracefully in the face of
> a full cache (or worse, it could destroy data).
Ok, i thought of something like that. But there is no performance increase?
Thanks, Klaas
>
> > Thanks for your answers
> > Klaas
>
> -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
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
From jds@soltis.cc Wed Oct 2 22:11:28 2002
From: jds@soltis.cc (Jesus Delgado)
Date: Wed, 2 Oct 2002 15:11:28 -0600
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
Message-ID: <20021002151128.M26033@soltis.cc>
Hi:
Problems when try the build openafs source rpm in redhat 8.0
the errors is the same with openafs-1.2.6 y openafs-1.2.7:
Errors in openafs-1.2.7
.4.1eq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1 rpmlib(CompressedFileNames) <=
3.0.4-1
Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
rpmlib(CompressedFileNames) <= 3.0.4-1
Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
Processing files: openafs-krb5-1.2.7-rh7.3.1
error: File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
error: File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
Requires: openafs = 1.2.7
RPM build errors:
File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
Errors with openafs-1.2.6
Requires: openafs = 1.2.6 openafs-client = 1.2.6
Obsoletes: openafs-client-compat
Processing files: openafs-kpasswd-1.2.6-rh7.3.1
Finding Provides: /usr/lib/rpm/find-provides
Finding Requires: /usr/lib/rpm/find-requires
PreReq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1 rpmlib(CompressedFileNames) <=
3.0.4-1
Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
rpmlib(CompressedFileNames) <= 3.0.4-1
Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
Processing files: openafs-krb5-1.2.6-rh7.3.1
error: File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
error: File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
Requires: openafs = 1.2.6
RPM build errors:
File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
Help me plase
Regards.
From warlord@MIT.EDU Fri Oct 4 15:29:17 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 04 Oct 2002 10:29:17 -0400
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
In-Reply-To: <20021002151128.M26033@soltis.cc>
References: <20021002151128.M26033@soltis.cc>
Message-ID:
RH8 is not supported by the RPMs yet.
I need to find a RH Mirror to install RH8 so I can build RPMS and
get the SPEC to work right.
This particular problem that you get looks like it's not building
aklog and asetkey properly. Most likely the issue is 'autoconf'
being 2.53 instead of 2.13.
-derek
"Jesus Delgado" writes:
> Hi:
>
> Problems when try the build openafs source rpm in redhat 8.0
> the errors is the same with openafs-1.2.6 y openafs-1.2.7:
>
> Errors in openafs-1.2.7
>
> .4.1eq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1 rpmlib(CompressedFileNames) <=
> 3.0.4-1
> Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> rpmlib(CompressedFileNames) <= 3.0.4-1
> Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
> libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
> Processing files: openafs-krb5-1.2.7-rh7.3.1
> error: File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
> error: File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
> Requires: openafs = 1.2.7
>
>
> RPM build errors:
> File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
> File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
>
> Errors with openafs-1.2.6
>
> Requires: openafs = 1.2.6 openafs-client = 1.2.6
> Obsoletes: openafs-client-compat
> Processing files: openafs-kpasswd-1.2.6-rh7.3.1
> Finding Provides: /usr/lib/rpm/find-provides
> Finding Requires: /usr/lib/rpm/find-requires
> PreReq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1 rpmlib(CompressedFileNames) <=
> 3.0.4-1
> Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> rpmlib(CompressedFileNames) <= 3.0.4-1
> Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
> libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
> Processing files: openafs-krb5-1.2.6-rh7.3.1
> error: File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
> error: File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
> Requires: openafs = 1.2.6
>
>
> RPM build errors:
> File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
> File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
>
> Help me plase
>
> Regards.
>
>
>
> _______________________________________________
> 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 dawson@fnal.gov Fri Oct 4 15:31:11 2002
From: dawson@fnal.gov (Troy Dawson)
Date: Fri, 04 Oct 2002 09:31:11 -0500
Subject: [OpenAFS] Re: mandrake 9.0 segmentation fault
Message-ID: <3D9DA62F.4030806@fnal.gov>
This is just throwing my two cents along with my experience.
I was able to recompile the 1.2.6 rpm's for Mandrake 9.0, and 8.2 without any
problem other than turning off the kerberos, and taking out the SuidCells.
(We already have a kerberos package with aklog in it, so we didn't need it,
and I have no idea what the SuidCells does) I had started with the RedHat 7.3
srpm.
I found that the first time I ran AFS, and the first time I went into an AFS
area I had very long lags. I suppose this was it loading everything into the
cache, but then after I was there, everything was as quick as afs normally is.
The machines I had these run on were
Mandrake 8.2 - Pentium 4 - 1.6 Ghz
Mandrake 9.0 - Pentium III - 450 Mhz - two processors
Both have stock kernels.
Though I can't say everything has been perfect. Both of them seem to loose
contact with the servers periodically, causing either long lags or a complete
inability to use AFS. One of them (the Mandrake 9.0) will loose contact with
the AFS server even though my redhat machine, on same switch, in same home
area, doesn't.
Troy Dawson
p.s. If anyone wants more details, I can send them.
--
__________________________________________________
Troy Dawson dawson@fnal.gov (630)840-6468
Fermilab ComputingDivision/OSS CSI Group
__________________________________________________
From yumiceva@sc.edu Fri Oct 4 17:12:12 2002
From: yumiceva@sc.edu (Francisco Yumiceva)
Date: Fri, 04 Oct 2002 12:12:12 -0400
Subject: [OpenAFS] volume disappeared
Message-ID: <3D9DBDDC.2090202@sc.edu>
Hi,
One of my volumes disappeared today. I tried "bos salvage -sever
myserver -all" but it didn't work.
Here are the symptoms:
vos listvol myserver
...
user.white 536870924 RW 57 K On-line
user.white.backup 536870926 BK 57 K On-line
user.yumiceva 536870921 RW 1954701 K On-line
**** Could not attach volume 536870923 ****
Total volumes onLine 17 ; Total volumes offLine 1 ; Total busy 0
When I do "bos status myserver -long" I see this error:
Instance backupusers, (type is cron) currently running normally.
Auxiliary status is: run next at Sat Oct 5 01:00:00 2002.
Process last started at Fri Oct 4 01:00:01 2002 (5 proc starts)
Last exit at Fri Oct 4 01:00:05 2002
Last error exit at Fri Oct 4 01:00:05 2002, due to signal 13
Command 1 is '/usr/sbin/vos backupsys -prefix user -localauth'
Command 2 is '1:00'
I cannot acces the volumen user.yumiceva may be because something is
wrong with the backup.
Please, I need help with this.
Thanks,
Francisco
--
_________________________________________________
Francisco Yumiceva
High Energy Group - Dept. of Physics & Astronomy
University of South Carolina
phone: 803.7771438
http://www.slac.stanford.edu/~yumiceva
_________________________________________________
From matt@slackers.net Fri Oct 4 17:20:24 2002
From: matt@slackers.net (Matthew N. Andrews)
Date: Fri, 04 Oct 2002 09:20:24 -0700
Subject: [OpenAFS] Re: mandrake 9.0 segmentation fault
References: <3D9DA62F.4030806@fnal.gov>
Message-ID: <3D9DBFC8.1010304@slackers.net>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Troy Dawson wrote:
| This is just throwing my two cents along with my experience.
| I was able to recompile the 1.2.6 rpm's for Mandrake 9.0, and 8.2
| without any problem other than turning off the kerberos, and taking out
| the SuidCells. (We already have a kerberos package with aklog in it, so
| we didn't need it, and I have no idea what the SuidCells does)
this is a list of cells from which you are willing to treat programs with the
set uid bit as actualy being set uid. In particular this is essentially saying
that you trust the administrators of that cell to be able to run things as
root on your machine if they can run things at all. you should probably only
place the cell(s) you administer(or those administered by the same people who
administer your machine) in this file.
~ -Matt Andrews
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQE9nb/IpLF3UzlwZVgRAte4AKDdczND9i2p5J90se55kFhdvDBwewCg3iq0
+GbxFKnIF/MrD61rycs28NI=
=MByZ
-----END PGP SIGNATURE-----
From security@xauth.net Fri Oct 4 18:06:29 2002
From: security@xauth.net (Charles Clancy)
Date: Fri, 4 Oct 2002 12:06:29 -0500 (CDT)
Subject: [OpenAFS] pam and openafs 1.2.7 for RH 7.2
In-Reply-To: <15773.16775.746446.362012@hszpc38.isz.ch>
Message-ID:
> > Try removing that option, and see if it works then.
>
> Unfortunately that doesn't have any impact, error messages in
> /var/log/messages still look the same after an attempt to login in.
>
> I really don't have a clou what's going on. Because I tried the same
> pam.d configuration on an Intel machine (with SuSE Linux 7.3,
> openafs1.2.3) and it works as expected! Do you have any ideas what
> else I could try out?
Perhaps it's SSHd related rather than PAM related.
I suggest running sshd in debug mode:
# sshd -d -d -d -D
Then, connect with a verbose client:
$ ssh -v -v -v user@host
Check the massive amounts of debug info for things that look
inappropriate.
I don't recall if you said so earlier -- but does your pam_afs.so work
with other applications? Is your SuSE box running the same version of
sshd?
[ t charles clancy ]--[ tclancy@uiuc.edu ]--[ www.uiuc.edu/~tclancy ]
From security@xauth.net Fri Oct 4 18:10:06 2002
From: security@xauth.net (Charles Clancy)
Date: Fri, 4 Oct 2002 12:10:06 -0500 (CDT)
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To:
Message-ID:
> > I have a number of machines at home which I can log in to.
> >
> > If I login on multiple machines, the previous host looses
> > connection to the AFS file server.
> >
> > Example:
> > 1. Login in on host1. Homedir on AFS.
> > Works fine.
> > 2. Login in on host2. Homedir on AFS.
> > Works fine - BUT: host1 looses the connection
> > to the file server!
> >
> >
> > The firewall is a Linux machine, doing NAT (masquerading).
>
> How long are your UDP timeouts?
In my experience, I can get one client to work fine from behind a NAT by
using long UDP timeouts. However, dispite what others have reported, I've
never been able to get multiple clients to work from behind a NAT (using
both IPF on Solaris and Win2K Server's built-in NAT router).
Of course, you'll have even more trouble with your krb5 TGTs.
[ t charles clancy ]--[ tclancy@uiuc.edu ]--[ www.uiuc.edu/~tclancy ]
From warlord@MIT.EDU Fri Oct 4 18:15:54 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 04 Oct 2002 13:15:54 -0400
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To:
References:
Message-ID:
Charles Clancy writes:
> Of course, you'll have even more trouble with your krb5 TGTs.
Turn off addressing in krb5 -- works great (kinit -A)
> [ t charles clancy ]--[ tclancy@uiuc.edu ]--[ www.uiuc.edu/~tclancy ]
-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 tim@umbc.edu Fri Oct 4 18:48:19 2002
From: tim@umbc.edu (Tim C.)
Date: Fri, 4 Oct 2002 13:48:19 -0400 (EDT)
Subject: [OpenAFS] volume disappeared
In-Reply-To: <3D9DBDDC.2090202@sc.edu>
Message-ID:
> One of my volumes disappeared today. I tried "bos salvage -sever
> myserver -all" but it didn't work.
>
Well hopefully that is supposed to be "-server". ;) But I would try this
command:
bos salvage -server myserver -partition /vicepa -volume user.yumiceva -salvagedirs -orphans remove
That's the command I normally use. BUT...looking at what you've shown, it
looks like the volume is online, so it's probably not bad. It will normally
take a volume off-line if it needs salvaged. I'm guessing that VI 536870923 is
the backup volume of the user. That should not affect accessing the actual
volume. What error messages are you getting when you try to access the volume?
I'm assuming it's mounted and you're trying to cd into it when you get these
errors?
As for the backup, if you'd like to fix that, here's the way I always use,
which works as well as any other. Just log into myserver as root, and cd to
the partition that the volume is on, and do an "rm V0536870923.vol". Make sure
that volume's there first of course(and make sure you have the right volumeid
;). Once you've done that, then you can just do a vos backup user.yumiceva.
I hope this helps you some, if not, try answering my above questions and
sending back to the list.
Thanks,
Tim
> Here are the symptoms:
> vos listvol myserver
> ...
> user.white 536870924 RW 57 K On-line
> user.white.backup 536870926 BK 57 K On-line
> user.yumiceva 536870921 RW 1954701 K On-line
> **** Could not attach volume 536870923 ****
>
> Total volumes onLine 17 ; Total volumes offLine 1 ; Total busy 0
>
> When I do "bos status myserver -long" I see this error:
> Instance backupusers, (type is cron) currently running normally.
> Auxiliary status is: run next at Sat Oct 5 01:00:00 2002.
> Process last started at Fri Oct 4 01:00:01 2002 (5 proc starts)
> Last exit at Fri Oct 4 01:00:05 2002
> Last error exit at Fri Oct 4 01:00:05 2002, due to signal 13
> Command 1 is '/usr/sbin/vos backupsys -prefix user -localauth'
> Command 2 is '1:00'
>
> I cannot acces the volumen user.yumiceva may be because something is
> wrong with the backup.
>
> Please, I need help with this.
>
> Thanks,
> Francisco
> --
> _________________________________________________
> Francisco Yumiceva
> High Energy Group - Dept. of Physics & Astronomy
> University of South Carolina
> phone: 803.7771438
>
> http://www.slac.stanford.edu/~yumiceva
> _________________________________________________
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
>
-----------------------------------------------------------------------
Tim Craig These are my opinions and not my employers. :)
OIT-Systems & Imaging Research Center
tim@umbc.edu It's hard to be serious when you're
naked. - Garfield
-----------------------------------------------------------------------
From ckovacs@DEPAUW.EDU Fri Oct 4 21:55:58 2002
From: ckovacs@DEPAUW.EDU (Corey Kovacs)
Date: Fri, 04 Oct 2002 15:55:58 -0500
Subject: [OpenAFS] butc error 22?
Message-ID: <200210041555.59095.ckovacs@depauw.edu>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
I've seen mention of a butc error around but I cannot find anything describing
what it means or a way to fix it. I've got a new tape unit, new tape and an
otherwise clean backup but at the end of the backup I get an error message
stating that butc can't close the tape etc. Anyone have any insight?
Corey
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.0 (GNU/Linux)
iD8DBQE9ngBeglw65kKkYY4RAiVWAJ4oc/74TRVEH4MKKaJnoUH2s3HKAACfVFKn
tmR+G4+wG2xNwDwJ9vjeCiI=
=XdrC
-----END PGP SIGNATURE-----
From andrei@caspur.it Fri Oct 4 23:40:34 2002
From: andrei@caspur.it (Andrei Maslennikov)
Date: Fri, 4 Oct 2002 22:40:34 +0000 ( )
Subject: [OpenAFS] rh8.0: follow-up
Message-ID:
http://afs.caspur.it now contains a working "grand-unified"
1.2.7 client RPM. Supported are: rh72,73,80(NB).
From warlord@MIT.EDU Fri Oct 4 23:43:08 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 04 Oct 2002 18:43:08 -0400
Subject: [OpenAFS] rh8.0: follow-up
In-Reply-To:
References:
Message-ID:
Well, www.openafs.org will hopefully, soon, have "official" RH8.0
RPMS.
-derek
Andrei Maslennikov writes:
> http://afs.caspur.it now contains a working "grand-unified"
> 1.2.7 client RPM. Supported are: rh72,73,80(NB).
>
>
> _______________________________________________
> 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 yumiceva@sc.edu Sat Oct 5 00:45:21 2002
From: yumiceva@sc.edu (Francisco Yumiceva)
Date: Fri, 04 Oct 2002 19:45:21 -0400
Subject: [OpenAFS] volume disappeared
References:
Message-ID: <3D9E2811.4070103@sc.edu>
Hi,
I removed the volume and I ran the salvage. Now, I got
access to the volume but I lost many files. I guess the
backup volume is gone after "rm V0536870923.vol" so bad :(
because the nightly backup also failed:
Last error exit at Fri Oct 4 01:00:05 2002, due to signal 13
I wonder if I would have been able to recover that volume or
the only solution was to remove it?. No idea what caused the
error in the backup.
Thanks,
Francisco
>
> Well hopefully that is supposed to be "-server". ;) But I would try this
> command:
>
> bos salvage -server myserver -partition /vicepa -volume user.yumiceva -salvagedirs -orphans remove
>
> That's the command I normally use. BUT...looking at what you've shown, it
> looks like the volume is online, so it's probably not bad. It will normally
> take a volume off-line if it needs salvaged. I'm guessing that VI 536870923 is
> the backup volume of the user. That should not affect accessing the actual
> volume. What error messages are you getting when you try to access the volume?
> I'm assuming it's mounted and you're trying to cd into it when you get these
> errors?
>
> As for the backup, if you'd like to fix that, here's the way I always use,
> which works as well as any other. Just log into myserver as root, and cd to
> the partition that the volume is on, and do an "rm V0536870923.vol". Make sure
> that volume's there first of course(and make sure you have the right volumeid
> ;). Once you've done that, then you can just do a vos backup user.yumiceva.
>
> I hope this helps you some, if not, try answering my above questions and
> sending back to the list.
>
> Thanks,
> Tim
>
>
>>Here are the symptoms:
>>vos listvol myserver
>>...
>>user.white 536870924 RW 57 K On-line
>>user.white.backup 536870926 BK 57 K On-line
>>user.yumiceva 536870921 RW 1954701 K On-line
>>**** Could not attach volume 536870923 ****
>>
>>Total volumes onLine 17 ; Total volumes offLine 1 ; Total busy 0
>>
>>When I do "bos status myserver -long" I see this error:
>>Instance backupusers, (type is cron) currently running normally.
>> Auxiliary status is: run next at Sat Oct 5 01:00:00 2002.
>> Process last started at Fri Oct 4 01:00:01 2002 (5 proc starts)
>> Last exit at Fri Oct 4 01:00:05 2002
>> Last error exit at Fri Oct 4 01:00:05 2002, due to signal 13
>> Command 1 is '/usr/sbin/vos backupsys -prefix user -localauth'
>> Command 2 is '1:00'
>>
>>I cannot acces the volumen user.yumiceva may be because something is
>>wrong with the backup.
>>
>>Please, I need help with this.
>>
>>Thanks,
>>Francisco
>>--
--
_________________________________________________
Francisco Yumiceva
High Energy Group - Dept. of Physics & Astronomy
University of South Carolina
phone: 803.7771438
http://www.slac.stanford.edu/~yumiceva
_________________________________________________
From 6delgado@informatik.uni-hamburg.de Sat Oct 5 01:01:45 2002
From: 6delgado@informatik.uni-hamburg.de (Friedrich Delgado Friedrichs)
Date: Sat, 5 Oct 2002 02:01:45 +0200
Subject: [OpenAFS] volume disappeared
In-Reply-To: <3D9E2811.4070103@sc.edu>
References: <3D9E2811.4070103@sc.edu>
Message-ID: <20021005000145.GA14081@taupan.ath.cx>
--6c2NcOVqGQ03X4Wi
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
Hi!
Francisco Yumiceva schrieb:
> I removed the volume and I ran the salvage. Now, I got=20
> access to the volume but I lost many files. I guess the=20
> backup volume is gone after "rm V0536870923.vol" so bad :(=20
> because the nightly backup also failed:
> Last error exit at Fri Oct 4 01:00:05 2002, due to signal 13
Yeah, with "-orphans attach" instead of "-orphans remove" you would
have kept your data, but the directory structure was lost forever
(i.e. the orphaned files and directories would have been attached to
the mountpoint of your volume and named AFSORPHAN-1232423 or some such
(i can't remember it exactly, had to do this once before))
> I wonder if I would have been able to recover that volume or=20
> the only solution was to remove it?. No idea what caused the=20
> error in the backup.
I have a guess what might have caused the error.
May i ask what version of OpenAFS you are using on which Operating
System?
> >bos salvage -server myserver -partition /vicepa -volume user.yumiceva=20
> >-salvagedirs -orphans remove
Bad, bad, bad... I wonder why Tim C. didn't ask which OpenAFS Version
and OS you were using...
Well, a little consolation: Moving the attached orphans around is no
fun at all and takes a lot of time... So restoring the tapes from
yesterday is probably the better solution anyways.
Ciao
Friedel
--=20
Friedrich Delgado Friedrichs
Laziness led to the invention of the most useful tools.
--6c2NcOVqGQ03X4Wi
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)
iEYEARECAAYFAj2eK+kACgkQCTmCEtF2zECAYACdGeVDF8Yi9lxq+PzwWH+JtVBi
4wEAn2nMwmZdgO27lvdqpaqdf/BbdBF6
=vvgg
-----END PGP SIGNATURE-----
--6c2NcOVqGQ03X4Wi--
From turbo@bayour.com Sat Oct 5 09:48:22 2002
From: turbo@bayour.com (Turbo Fredriksson)
Date: 05 Oct 2002 10:48:22 +0200
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To:
References:
Message-ID: <87d6qpuv8p.fsf@papadoc.bayour.com>
>>>>> "Derek" == Derek Atkins writes:
Derek> Charles Clancy writes:
>> Of course, you'll have even more trouble with your krb5 TGTs.
Derek> Turn off addressing in krb5 -- works great (kinit -A)
I have the same problem (using krb5 TGTs), and I _DO_ have address less
tickets.. Still no go.
From warlord@MIT.EDU Sat Oct 5 14:52:25 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 05 Oct 2002 09:52:25 -0400
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To: <87d6qpuv8p.fsf@papadoc.bayour.com>
References:
<87d6qpuv8p.fsf@papadoc.bayour.com>
Message-ID:
Turbo Fredriksson writes:
> Derek> Turn off addressing in krb5 -- works great (kinit -A)
>
> I have the same problem (using krb5 TGTs), and I _DO_ have address less
> tickets.. Still no go.
Eh? aklog works fine from behind NAT using addressless tickets...
Not _all_ apps will work, however. What in particular fails for
you?
-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 turbo@bayour.com Sat Oct 5 14:59:32 2002
From: turbo@bayour.com (Turbo Fredriksson)
Date: 05 Oct 2002 15:59:32 +0200
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To:
References:
<87d6qpuv8p.fsf@papadoc.bayour.com>
Message-ID: <878z1dt29n.fsf@papadoc.bayour.com>
Quoting Derek Atkins :
> Turbo Fredriksson writes:
>
> > Derek> Turn off addressing in krb5 -- works great (kinit -A)
> >
> > I have the same problem (using krb5 TGTs), and I _DO_ have address less
> > tickets.. Still no go.
>
> Eh? aklog works fine from behind NAT using addressless tickets...
> Not _all_ apps will work, however. What in particular fails for
> you?
I never had any problems with aklog. I have problem with the second login
on the second host disrupting the AFS traffic on the first host!
From yumiceva@sc.edu Sat Oct 5 15:02:14 2002
From: yumiceva@sc.edu (Francisco Yumiceva)
Date: Sat, 05 Oct 2002 10:02:14 -0400
Subject: [OpenAFS] volume disappeared
References: <3D9E2811.4070103@sc.edu> <20021005000145.GA14081@taupan.ath.cx>
Message-ID: <3D9EF0E6.5010206@sc.edu>
Hi,
>>I wonder if I would have been able to recover that volume or
>>the only solution was to remove it?. No idea what caused the
>>error in the backup.
>
> I have a guess what might have caused the error.
>
> May i ask what version of OpenAFS you are using on which Operating
> System?
The server where the problems appeared was running an old
openafs (1.0.4) and it is a RH 6.2, the rest of servers and
clients are RH7.2 with openafs 1.2.6. I have updated that
server to 1.2.7. I hope this will fix the backup error.
> Well, a little consolation: Moving the attached orphans around is no
> fun at all and takes a lot of time... So restoring the tapes from
> yesterday is probably the better solution anyways.
:( yeah.
Francisco
--
_________________________________________________
Francisco Yumiceva
High Energy Group - Dept. of Physics & Astronomy
University of South Carolina
phone: 803.7771438
http://www.slac.stanford.edu/~yumiceva
_________________________________________________
From warlord@MIT.EDU Sat Oct 5 15:48:50 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 05 Oct 2002 10:48:50 -0400
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To: <878z1dt29n.fsf@papadoc.bayour.com>
References:
<87d6qpuv8p.fsf@papadoc.bayour.com>
<878z1dt29n.fsf@papadoc.bayour.com>
Message-ID:
Turbo Fredriksson writes:
> I never had any problems with aklog. I have problem with the second login
> on the second host disrupting the AFS traffic on the first host!
What version of AFS are you running on your servers? I suppose I can
test this using VMware.. I will at some point and see what I can see,
but honestly it SHOULD work -- the servers are supposed to use client
uuid to identify, and use the source ip:port to respond.
-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 warlord@MIT.EDU Sat Oct 5 15:52:19 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 05 Oct 2002 10:52:19 -0400
Subject: [OpenAFS] volume disappeared
In-Reply-To: <3D9EF0E6.5010206@sc.edu>
References:
<3D9E2811.4070103@sc.edu> <20021005000145.GA14081@taupan.ath.cx>
<3D9EF0E6.5010206@sc.edu>
Message-ID:
Francisco Yumiceva writes:
> The server where the problems appeared was running an old openafs
> (1.0.4) and it is a RH 6.2, the rest of servers and clients are RH7.2
> with openafs 1.2.6. I have updated that server to 1.2.7. I hope this
> will fix the backup error.
Ahh, this might have been the old CopyOnWrite bug.. If so, then moving
to 1.2.7 _should_ fix it...
-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 turbo@bayour.com Sat Oct 5 16:07:38 2002
From: turbo@bayour.com (Turbo Fredriksson)
Date: 05 Oct 2002 17:07:38 +0200
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To:
References:
<87d6qpuv8p.fsf@papadoc.bayour.com>
<878z1dt29n.fsf@papadoc.bayour.com>
Message-ID: <874rc0udol.fsf@papadoc.bayour.com>
Quoting Derek Atkins :
> Turbo Fredriksson writes:
>
> > I never had any problems with aklog. I have problem with the second login
> > on the second host disrupting the AFS traffic on the first host!
>
> What version of AFS are you running on your servers? I suppose I can
> test this using VMware.. I will at some point and see what I can see,
> but honestly it SHOULD work -- the servers are supposed to use client
> uuid to identify, and use the source ip:port to respond.
1.2.6 on both the server and client (even the win clients have 1.2.6).
Since it's NAT, _ALL_ machines have the same IP/port (from the server's
stand point any way).
From warlord@MIT.EDU Sat Oct 5 16:30:57 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 05 Oct 2002 11:30:57 -0400
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To: <874rc0udol.fsf@papadoc.bayour.com>
References:
<87d6qpuv8p.fsf@papadoc.bayour.com>
<878z1dt29n.fsf@papadoc.bayour.com>
<874rc0udol.fsf@papadoc.bayour.com>
Message-ID:
Turbo Fredriksson writes:
> Since it's NAT, _ALL_ machines have the same IP/port (from the server's
> stand point any way).
No, all clients have the same IP. They do NOT have the same port
(unless you misconfigred your NAT box).... If you specifically set up
port-forwarding for port 7001 in your NAT box, then yes, you have shot
yourself in the foot. However, if you let the ports float (i.e., if
you let the NAT box choose an arbitrary port for every "connection")
then it should work just fine.
Assume you have a picture like this:
A--\ ______
B---\___| NAT |-- AFS Server
... / ------
X--/
The AFS server will see 'A' as 'NAT:12345', 'B" as 'NAT:23456' and 'X'
as 'NAT:23489' (just to throw out random numbers).
FTR, I'm running two VMware guest machines behind NAT on this host and
both can access AFS just fine.
If you are having NAT problems it is almost assuredly a NAT
configuration bug on your part. Note that Linux-2.4's NAT has
hard-coded UDP timeouts -- you need to recompile your kernel to fix
that.
-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 turbo@bayour.com Sat Oct 5 16:52:22 2002
From: turbo@bayour.com (Turbo Fredriksson)
Date: 05 Oct 2002 17:52:22 +0200
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To:
References:
<87d6qpuv8p.fsf@papadoc.bayour.com>
<878z1dt29n.fsf@papadoc.bayour.com>
<874rc0udol.fsf@papadoc.bayour.com>
Message-ID: <87zntssx1l.fsf@papadoc.bayour.com>
Quoting Derek Atkins :
> Turbo Fredriksson writes:
>
> > Since it's NAT, _ALL_ machines have the same IP/port (from the server's
> > stand point any way).
>
> No, all clients have the same IP. They do NOT have the same port
> (unless you misconfigred your NAT box)....
Right, sorry.
> If you specifically set up
> port-forwarding for port 7001 in your NAT box, then yes, you have shot
> yourself in the foot. However, if you let the ports float (i.e., if
> you let the NAT box choose an arbitrary port for every "connection")
> then it should work just fine.
No portforwarding, no specific port configuration so they 'float'.
> If you are having NAT problems it is almost assuredly a NAT
> configuration bug on your part. Note that Linux-2.4's NAT has
> hard-coded UDP timeouts -- you need to recompile your kernel to fix
> that.
UDP timeouts... Haven't changed that, so I guess I'm running default
values. Is this an issue on the firewall, the server or the client
(how to change that on win?)?
From warlord@MIT.EDU Sat Oct 5 17:03:37 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 05 Oct 2002 12:03:37 -0400
Subject: [OpenAFS] Multiple hosts behind firewall and AFS cell
In-Reply-To: <87zntssx1l.fsf@papadoc.bayour.com>
References:
<87d6qpuv8p.fsf@papadoc.bayour.com>
<878z1dt29n.fsf@papadoc.bayour.com>
<874rc0udol.fsf@papadoc.bayour.com>
<87zntssx1l.fsf@papadoc.bayour.com>
Message-ID:
Turbo Fredriksson writes:
> > If you are having NAT problems it is almost assuredly a NAT
> > configuration bug on your part. Note that Linux-2.4's NAT has
> > hard-coded UDP timeouts -- you need to recompile your kernel to fix
> > that.
>
> UDP timeouts... Haven't changed that, so I guess I'm running default
> values. Is this an issue on the firewall, the server or the client
> (how to change that on win?)?
it is an issue on the NAT-firewall. How you change it depends on
your particular firewall.
-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 schmitt@inf.ethz.ch Sat Oct 5 19:19:44 2002
From: schmitt@inf.ethz.ch (Marc Schmitt)
Date: Sat, 05 Oct 2002 20:19:44 +0200
Subject: [OpenAFS] pam and openafs 1.2.7 for RH 7.2
References: <15770.43001.206989.710003@hszpc38.isz.ch>
Message-ID: <3D9F2D40.4050704@inf.ethz.ch>
Hi Andi,
Was the sshd version on the alpha machine built --with-afs?
I`m seeing the problem you describe under RedHat 7.3 with
openafs-1.2.7-rh7.3.1 and openssh-3.4p1-3 (what I changed between
3.4p1-2 and 3.4p1-3 is adding "--with-afs=/usr
--with-kerberos4=/usr/athena" to the configure line, krb4 is version 1.2).
If I use openssh-3.4p1-2, I get:
Oct 5 19:35:14 otherhost sshd(pam_unix)[8281]: session opened for user
testuser by (uid=0)
If I use openssh-3.4p1-3, I get:
Oct 5 19:47:42 otherhost pam_afs[15855]: AFS Authentication failed for
user testuser. password was incorrect
Oct 5 19:47:42 otherhost sshd(pam_unix)[15851]: check pass; user unknown
Oct 5 19:47:42 otherhost sshd(pam_unix)[15851]: authentication failure;
logname= uid=0 euid=0 tty=NODEVssh ruser= rhost=somehost
Looking at the debug output of sshd:
With openssh-3.4p1-2:
Oct 5 20:03:53 otherhost sshd[23253]: Failed none for testuser from
129.132.10.58 port 35551
Oct 5 20:03:53 otherhost sshd[23253]: debug3: mm_request_receive entering
Oct 5 20:03:56 otherhost sshd[23253]: debug3: monitor_read: checking
request 10
Oct 5 20:03:56 otherhost sshd[23253]: debug1: PAM Password
authentication accepted for user "testuser"
Oct 5 20:03:56 otherhost sshd[23253]: debug3: mm_answer_authpassword:
sending result 1
Oct 5 20:03:56 otherhost sshd[23253]: debug3: mm_request_send entering:
type 11
Oct 5 20:03:56 otherhost sshd[23253]: debug2: pam_acct_mgmt() = 0
Oct 5 20:03:56 otherhost sshd[23253]: Accepted password for testuser
from 129.132.10.58 port 35551
Oct 5 20:03:56 otherhost sshd[23253]: debug1: monitor_child_preauth:
testuser has been authenticated by privileged process
and openssh-3.4p1-3:
Oct 5 19:47:39 otherhost sshd[15851]: Failed none for testuser from
129.132.10.58 port 35528
Oct 5 19:47:39 otherhost sshd[15851]: debug3: mm_request_receive entering
Oct 5 19:47:42 otherhost sshd[15851]: debug3: monitor_read: checking
request 10
Oct 5 19:47:44 otherhost sshd[15851]: debug1: PAM Password
authentication for "testuser" failed[7]: Authentication failure
Oct 5 19:47:44 otherhost sshd[15851]: debug3: mm_answer_authpassword:
sending result 0
Oct 5 19:47:44 otherhost sshd[15851]: debug3: mm_request_send entering:
type 11
Oct 5 19:47:44 otherhost sshd[15851]: Failed password for testuser from
129.132.10.58 port 35528
PAM authentication fails... but why? Nothing has changed in
/etc/pam.d/system-auth nor /etc/pam.d/sshd between the two tests.
Looks like AFS support in OpenSSH bites pam AFS authentication...
Regards,
Marc
Andreas Buechler wrote:
> Hello,
>
> I just installed openafs 1.2.7 on a alpha machine. Everything worked fine
> (rebuilding and installing the rpm's) and at the end I was told to change
> the files cacheinfo and ThisCell. I changed both files, now I am able to
> get tokens etc as root for any afs-user. To be able to login and get a
> token automatically I changed /etc/pam.d/system-auth as discribed at the end of the
> installation.
> Does anybody have an idea why I still cant login via ssh as an afs-user?
> I posted my sshd and system-auth pam-files at the end of this mail.
>
> Thanks for any help and sorry if this message was posted twice!
>
> Andi
From 6delgado@informatik.uni-hamburg.de Sat Oct 5 22:34:37 2002
From: 6delgado@informatik.uni-hamburg.de (Friedrich Delgado Friedrichs)
Date: Sat, 5 Oct 2002 23:34:37 +0200
Subject: [OpenAFS] volume disappeared
In-Reply-To:
References: <3D9E2811.4070103@sc.edu> <20021005000145.GA14081@taupan.ath.cx> <3D9EF0E6.5010206@sc.edu>
Message-ID: <20021005213437.GA13527@taupan.ath.cx>
--azLHFNyN32YCQGCU
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
Hiho!
Derek Atkins schrieb:
> Francisco Yumiceva writes:
> > The server where the problems appeared was running an old openafs
> > (1.0.4) and it is a RH 6.2, the rest of servers and clients are RH7.2
> > with openafs 1.2.6. I have updated that server to 1.2.7. I hope this
> > will fix the backup error.
> Ahh, this might have been the old CopyOnWrite bug.. If so, then moving
> to 1.2.7 _should_ fix it...
The description sounded very much like my Experiences with the
CopyOnWrite bug on linux. And that's what my guess was.
Upgrading to 1.2.5 or higher has seemingly fixed it for quite some
time now...
At least i have started to feel a lot more comfortable with OpenAFS on
Linux. :)
Cheerio
Friedel
--=20
Friedrich Delgado Friedrichs
Laziness led to the invention of the most useful tools.
--azLHFNyN32YCQGCU
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)
iEYEARECAAYFAj2fWu0ACgkQCTmCEtF2zEAaVACeIJ4SDYLIYnzDE31MvHOncc3Y
bmwAn08E6DIm6Lo/bj1PZo3urHuIAoke
=NQGN
-----END PGP SIGNATURE-----
--azLHFNyN32YCQGCU--
From fw@fwsystems.com Sun Oct 6 05:14:08 2002
From: fw@fwsystems.com (forrest whitcher)
Date: Sun, 6 Oct 2002 00:14:08 -0400
Subject: [OpenAFS] Problem authenticating against (aix) Kerberos v 1.2.6
Message-ID: <20021006001408.73238c2e.fw@fwsystems.com>
My kerberos admin server is an aix box, on updating it I found that
(reportedly) valid afs tokens would still be issued as had worked
before, however they didn't work for actual access.
On reverting to kerberos 1.2.5 w/ xdr patch everything now works
ok again.
The other kdc's run Linux and there seem to be no problems there,
both are updated to 1.2.6.
This behavior was true for clients running both openafs 1.2.7
and the early release 1.0.3.
I can't give much better diagnostics right now, as it's not easy
to make changes to the kadmind service and still keep tings running
smoothly
forrest
From fbo2@gmx.net Sun Oct 6 08:16:17 2002
From: fbo2@gmx.net (FBO)
Date: Sun, 6 Oct 2002 09:16:17 +0200
Subject: [OpenAFS] Unknown processes after update to openafs 1.2.7
Message-ID: <20021006071617.GA6094@fbo.no-ip.org>
Hi,
after updating to openafs 1.2.7 i found
some processes never seen before:
# pstree -p
[snip]
|-keventd(2)-+-afs_background(402)
| |-afs_background(403)
| |-afs_background(404)
| |-afs_cachetrim(407)
| |-afs_callback(393)
| |-afs_checkserver(398)
| |-afs_rxevent(392)
| |-afs_rxlistener(391)
| -afsd(397)
[snip]
Can someone tell what they do or
what I can do with them?
FBO
From kolya@MIT.EDU Sun Oct 6 08:34:03 2002
From: kolya@MIT.EDU (Nickolai Zeldovich)
Date: Sun, 06 Oct 2002 03:34:03 -0400
Subject: [OpenAFS] Unknown processes after update to openafs 1.2.7
Message-ID: <200210060734.DAA23103@contents-vnder-pressvre.mit.edu>
> |-keventd(2)-+-afs_background(402)
> [...]
>
> Can someone tell what they do or
> what I can do with them?
These are basically the same processes that you used to see as "afsd"
before. I'm not sure why they're all displayed as children of keventd
(maybe some function of the way Linux does kernel threads), if that's
what you're asking. In short, they make AFS work -- process requests
in the background, receive packets, and perform other miscellaneous
tasks. There isn't much you can do with them, kind-of in the same
way there isn't much you can do with, say, kflushd.
-- kolya
From nneul@umr.edu Sun Oct 6 15:39:29 2002
From: nneul@umr.edu (Nathan Neulinger)
Date: 06 Oct 2002 09:39:29 -0500
Subject: [OpenAFS] Unknown processes after update to openafs 1.2.7
In-Reply-To: <200210060734.DAA23103@contents-vnder-pressvre.mit.edu>
References: <200210060734.DAA23103@contents-vnder-pressvre.mit.edu>
Message-ID: <1033915168.3299.1.camel@cessna.rollanet.org>
This surprised me as well... I think sometime recently, they changed
from being listed as afsd to being listed as specific names. They've
always been there, just never dipsplayed that way before.
On Sun, 2002-10-06 at 02:34, Nickolai Zeldovich wrote:
> > |-keventd(2)-+-afs_background(402)
> > [...]
> >
> > Can someone tell what they do or
> > what I can do with them?
>
> These are basically the same processes that you used to see as "afsd"
> before. I'm not sure why they're all displayed as children of keventd
> (maybe some function of the way Linux does kernel threads), if that's
> what you're asking. In short, they make AFS work -- process requests
> in the background, receive packets, and perform other miscellaneous
> tasks. There isn't much you can do with them, kind-of in the same
> way there isn't much you can do with, say, kflushd.
>
> -- kolya
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
--
------------------------------------------------------------
Nathan Neulinger EMail: nneul@umr.edu
University of Missouri - Rolla Phone: (573) 341-4841
Computing Services Fax: (573) 341-4216
From cg2v@andrew.cmu.edu Sun Oct 6 15:42:06 2002
From: cg2v@andrew.cmu.edu (Chaskiel M Grundman)
Date: Sun, 06 Oct 2002 10:42:06 -0400
Subject: [OpenAFS] Unknown processes after update to openafs 1.2.7
In-Reply-To: <200210060734.DAA23103@contents-vnder-pressvre.mit.edu>
References: <200210060734.DAA23103@contents-vnder-pressvre.mit.edu>
Message-ID: <17670000.1033915326@endicott>
--On Sunday, October 06, 2002 03:34:03 -0400 Nickolai Zeldovich
wrote:
> I'm not sure why they're all displayed as children of keventd
The almost-consensus of the various mailing list posts I read about how to
properly create a truly user-process independent kernel thread was that
'daemonize()' doesn't really completely sever a kernel thread from it's
user-process parent, and the only way to be sure was to use schedule_task()
to make sure the thread's parent was keventd.
From warlord@MIT.EDU Sun Oct 6 15:45:42 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 06 Oct 2002 10:45:42 -0400
Subject: [OpenAFS] Unknown processes after update to openafs 1.2.7
In-Reply-To: <1033915168.3299.1.camel@cessna.rollanet.org>
References: <200210060734.DAA23103@contents-vnder-pressvre.mit.edu>
<1033915168.3299.1.camel@cessna.rollanet.org>
Message-ID:
It surprised me, but it does sort of change the FAQ from:
Q: What are all these afsd processes (and how do I get rid of them)
to
Q: What are all these afs_* processes (and how do I get rid of them)
;)
-derek
Nathan Neulinger writes:
> This surprised me as well... I think sometime recently, they changed
> from being listed as afsd to being listed as specific names. They've
> always been there, just never dipsplayed that way before.
>
> On Sun, 2002-10-06 at 02:34, Nickolai Zeldovich wrote:
> > > |-keventd(2)-+-afs_background(402)
> > > [...]
> > >
> > > Can someone tell what they do or
> > > what I can do with them?
> >
> > These are basically the same processes that you used to see as "afsd"
> > before. I'm not sure why they're all displayed as children of keventd
> > (maybe some function of the way Linux does kernel threads), if that's
> > what you're asking. In short, they make AFS work -- process requests
> > in the background, receive packets, and perform other miscellaneous
> > tasks. There isn't much you can do with them, kind-of in the same
> > way there isn't much you can do with, say, kflushd.
> >
> > -- kolya
> > _______________________________________________
> > OpenAFS-info mailing list
> > OpenAFS-info@openafs.org
> > https://lists.openafs.org/mailman/listinfo/openafs-info
> --
>
> ------------------------------------------------------------
> Nathan Neulinger EMail: nneul@umr.edu
> University of Missouri - Rolla Phone: (573) 341-4841
> Computing Services Fax: (573) 341-4216
>
> _______________________________________________
> 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 nneul@umr.edu Sun Oct 6 16:22:19 2002
From: nneul@umr.edu (Nathan Neulinger)
Date: 06 Oct 2002 10:22:19 -0500
Subject: [OpenAFS] Unknown processes after update to openafs 1.2.7
In-Reply-To:
References: <200210060734.DAA23103@contents-vnder-pressvre.mit.edu>
<1033915168.3299.1.camel@cessna.rollanet.org>
Message-ID: <1033917739.1617.5.camel@cessna.rollanet.org>
We might consider changing them from:
527 ? 00:00:01 afs_rxlistener
528 ? 00:00:00 afs_callback
529 ? 00:00:16 afs_rxevent
559 ? 00:00:02 afs_checkserver
560 ? 00:00:00 afs_background
561 ? 00:00:00 afs_background
565 ? 00:00:00 afs_background
566 ? 00:00:00 afs_background
567 ? 00:00:05 afs_background
569 ? 00:00:01 afs_cachetrim
To the style more commonly used by ftpd, rsync, etc.
afsd [rxlistener]
afsd [callback]
afsd [checkserver]
afsd [background]
....
That may be less confusing, and clearly indicates that they are spawned
from afsd.
Doesn't matter much to me though...
-- Nathan
On Sun, 2002-10-06 at 09:45, Derek Atkins wrote:
> It surprised me, but it does sort of change the FAQ from:
>
> Q: What are all these afsd processes (and how do I get rid of them)
>
> to
>
> Q: What are all these afs_* processes (and how do I get rid of them)
>
> ;)
>
> -derek
>
> Nathan Neulinger writes:
>
> > This surprised me as well... I think sometime recently, they changed
> > from being listed as afsd to being listed as specific names. They've
> > always been there, just never dipsplayed that way before.
> >
> > On Sun, 2002-10-06 at 02:34, Nickolai Zeldovich wrote:
> > > > |-keventd(2)-+-afs_background(402)
> > > > [...]
> > > >
> > > > Can someone tell what they do or
> > > > what I can do with them?
> > >
> > > These are basically the same processes that you used to see as "afsd"
> > > before. I'm not sure why they're all displayed as children of keventd
> > > (maybe some function of the way Linux does kernel threads), if that's
> > > what you're asking. In short, they make AFS work -- process requests
> > > in the background, receive packets, and perform other miscellaneous
> > > tasks. There isn't much you can do with them, kind-of in the same
> > > way there isn't much you can do with, say, kflushd.
> > >
> > > -- kolya
> > > _______________________________________________
> > > OpenAFS-info mailing list
> > > OpenAFS-info@openafs.org
> > > https://lists.openafs.org/mailman/listinfo/openafs-info
> > --
> >
> > ------------------------------------------------------------
> > Nathan Neulinger EMail: nneul@umr.edu
> > University of Missouri - Rolla Phone: (573) 341-4841
> > Computing Services Fax: (573) 341-4216
> >
> > _______________________________________________
> > 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
--
------------------------------------------------------------
Nathan Neulinger EMail: nneul@umr.edu
University of Missouri - Rolla Phone: (573) 341-4841
Computing Services Fax: (573) 341-4216
From ian@assv.net Sun Oct 6 23:34:04 2002
From: ian@assv.net (Ian Delahorne)
Date: 07 Oct 2002 00:34:04 +0200
Subject: [OpenAFS] Problems releasing root.cell
Message-ID:
Why am I getting this?
ian@dominatrix .assv.net $ fs mkm /afs/.assv.net/cvs cvs
ian@dominatrix .assv.net $ vos release root.cell
Failed to reclone the RW volume 536870915
: Invalid cross-device link
Error in vos release command.
: Invalid cross-device link
ian@dominatrix .assv.net $ vos exa 536870915
root.cell 536870915 RW 10 K On-line
dominatrix.lackflicka.nu /vicepa
RWrite 536870915 ROnly 0 Backup 0
MaxQuota 5000 K
Creation Tue Sep 24 22:57:32 2002
Last Update Mon Oct 7 00:29:31 2002
3 accesses in the past day (i.e., vnode references)
RWrite: 536870915 ROnly: 536870916 Backup: 536870917
number of sites -> 2
server dominatrix.lackflicka.nu partition /vicepa RW Site
server dominatrix.lackflicka.nu partition /vicepa RO Site
ian@dominatrix .assv.net $
vos status says that there are no active transactions.
--
/Ian D
ian@assv.net - www.assv.net
From enrico@it.kth.se Mon Oct 7 11:41:38 2002
From: enrico@it.kth.se (Enrico Pelletta)
Date: Mon, 07 Oct 2002 12:41:38 +0200
Subject: [OpenAFS] FreeBSD & OpenAFS
Message-ID: <3DA164E2.2F781226@it.kth.se>
Hi!
I would like to create an AFS cell using OpenAFS and FreeBSD for the
servers. By now, I'm just making some tests for learning, but I have not
found much documentation/information about OpenAFS using FreeBSD, so I
wonder if it's a good combination. Anybody know where I could get some
more information about?
Thanks,
Enrico.
--
Enrico Pelletta
IT Universitetet System Group
KTH (Royal Institute of Technology) Stockholm (S).
From dfs17@cornell.edu Mon Oct 7 07:43:18 2002
From: dfs17@cornell.edu (Daniel Schmidt)
Date: Mon, 7 Oct 2002 02:43:18 -0400
Subject: [OpenAFS] NOTE: Installer Bug OpenAFS for Windows 1.2.2 b
Message-ID:
Hi folks.
I recently installed OpenAFS for Windows 1.2.2b (as DLed from
www.openafs.org) on my Windows 2000 SP3 machine (hence I got the NT release;
dunno if the 9X release has the same issues). I chose the default
installation, Client, Control Center, and Docs, but no Server, and found an
_extremely_ obnoxious bug:
When this program installs itself, it edits the PATH environment variable
and makes multiple registry entries, including some uninstall entries.
_Every single entry_ referring to a needed DLL (including the uninstall
DLLs) was miswritten - Not to mention the additions to the PATH statement.
They _all_ pointed to "subdirectories of C:\Program Files\IBM\..." or
something like that, a directory which OpenAFS did not create, much less
install anything into - OpenAFS installed all files in %WINNT%\, but did not
know they were there.
The results were predictable. Immediate errors on startup about being
unable to load the appropriate library, and my complete inability to
uninstall any part of the software.
Having figured this out after a fair amount of swearing, I edited the
appropriate registry entries and verified that it was indeed functional. As
a result I'm now somewhat less unhappy - But this _really_ needs to be
fixed. It looks like, between the docs (IBM logos everywhere, does IBM
mind?) and the registry entries, a lot of artifacts from the days when AFS
was an IBM product still exist.
Funny thing is, the 1.2.6 "unofficial" client-only install the kind folks at
Cornell sent me a link to works just fine, and doesn't do this - Too bad
it's not available at www.openafs.org :( The link is here BTW, in case you
want to put it on the OpenAFS site:
http://www.unc.edu/~sdw/OpenAFS_Client_126.exe
I know that no one is really actively maintaining the OpenAFS distribution
for Windows, and I don't expect that to change just 'cause of me - But since
this is so extremely annoying (and since it should be so easy to correct),
if someone who has access to these files could fix this, you'd be doing a
big favor to anyone else who might download it.
Thanks!!
PS - I'm not subscribed to the OpenAFS info mailing list, and I swear I will
trouble you no longer; I just figured someone might benefit from this
knowledge. If you need to get in touch, feel free to drop me an e-mail.
PPS - Derrick J Brashear - Do you remember me? CMU, '94, I lived on your
hall! 1st floor 'Schlag, what a hole that was... Small world... (Hi
Mitch! :)
--
Daniel Schmidt
dfs17@cornell.edu
--------------------------------------------------
Department of Materials Science and Engineering
Cornell University
--------------------------------------------------
[Opinions expressed are my own exclusively]
From mtawafig@optonline.com Mon Oct 7 12:46:11 2002
From: mtawafig@optonline.com (Muhsin Tawafig)
Date: Mon, 7 Oct 2002 07:46:11 -0400
Subject: [OpenAFS] REDHAT 7.3 & openafs 1.2.7 rpms
Message-ID:
hi
When seting up initial security the following message is recieved.
openafs-1.2.7 from rpms.
[root@h2 bin]# kas bizittech.com -noauth
ka> list
list: server or network not responding calling KAM_ListEntry
any hints
thanks
From mike@bizittech.com Mon Oct 7 16:02:47 2002
From: mike@bizittech.com (mike@bizittech.com)
Date: Mon, 7 Oct 2002 11:02:47 -0400
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
References: <20021002151128.M26033@soltis.cc> <20021002195332.M26176@soltis.cc>
Message-ID: <014101c26e12$99441dc0$01000001@micron>
hi
I had the same problem .Use the source code and compile .I got better
results .
Thanks
----- Original Message -----
From: "Jesus Delgado"
To:
Sent: Wednesday, October 02, 2002 9:53 PM
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
>
> ---------- Forwarded Message -----------
> Hi:
>
> Problems when try the build openafs source rpm in redhat 8.0
> the errors is the same with openafs-1.2.6 y openafs-1.2.7:
>
> Errors in openafs-1.2.7
>
> .4.1eq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1
rpmlib(CompressedFileNames) <=
> 3.0.4-1
> Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> rpmlib(CompressedFileNames) <= 3.0.4-1
> Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
> libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
> Processing files: openafs-krb5-1.2.7-rh7.3.1
> error: File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
> error: File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
> Requires: openafs = 1.2.7
>
> RPM build errors:
> File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
> File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
>
> Errors with openafs-1.2.6
>
> Requires: openafs = 1.2.6 openafs-client = 1.2.6
> Obsoletes: openafs-client-compat
> Processing files: openafs-kpasswd-1.2.6-rh7.3.1
> Finding Provides: /usr/lib/rpm/find-provides
> Finding Requires: /usr/lib/rpm/find-requires
> PreReq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1
rpmlib(CompressedFileNames) <=
> 3.0.4-1
> Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> rpmlib(CompressedFileNames) <= 3.0.4-1
> Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
> libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
> Processing files: openafs-krb5-1.2.6-rh7.3.1
> error: File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
> error: File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
> Requires: openafs = 1.2.6
>
> RPM build errors:
> File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
> File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
>
> Help me plase
>
> Regards.
> ------- End of Forwarded Message -------
>
>
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
>
From warlord@MIT.EDU Mon Oct 7 17:43:35 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 07 Oct 2002 12:43:35 -0400
Subject: [OpenAFS] REDHAT 7.3 & openafs 1.2.7 rpms
In-Reply-To:
References:
Message-ID:
Did you add your cell information to /usr/vice/etc/CellServDB?
-derek
"Muhsin Tawafig" writes:
> hi
> When seting up initial security the following message is recieved.
> openafs-1.2.7 from rpms.
> [root@h2 bin]# kas bizittech.com -noauth
> ka> list
> list: server or network not responding calling KAM_ListEntry
>
> any hints
>
> thanks
>
> _______________________________________________
> 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 warlord@MIT.EDU Mon Oct 7 17:44:07 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 07 Oct 2002 12:44:07 -0400
Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
In-Reply-To: <014101c26e12$99441dc0$01000001@micron>
References: <20021002151128.M26033@soltis.cc>
<20021002195332.M26176@soltis.cc>
<014101c26e12$99441dc0$01000001@micron>
Message-ID:
This is fixed in the "official" RH8.0 RPMS
-derek
writes:
> hi
> I had the same problem .Use the source code and compile .I got better
> results .
> Thanks
> ----- Original Message -----
> From: "Jesus Delgado"
> To:
> Sent: Wednesday, October 02, 2002 9:53 PM
> Subject: [OpenAFS] Problems in RedHat 8.0 openafs 1.2.6 y 1.2.7
>
>
> >
> > ---------- Forwarded Message -----------
> > Hi:
> >
> > Problems when try the build openafs source rpm in redhat 8.0
> > the errors is the same with openafs-1.2.6 y openafs-1.2.7:
> >
> > Errors in openafs-1.2.7
> >
> > .4.1eq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> rpmlib(CompressedFileNames) <=
> > 3.0.4-1
> > Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> > rpmlib(CompressedFileNames) <= 3.0.4-1
> > Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
> > libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
> > Processing files: openafs-krb5-1.2.7-rh7.3.1
> > error: File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
> > error: File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
> > Requires: openafs = 1.2.7
> >
> > RPM build errors:
> > File not found: /var/tmp/openafs-1.2.7-root/usr/bin/aklog
> > File not found: /var/tmp/openafs-1.2.7-root/usr/sbin/asetkey
> >
> > Errors with openafs-1.2.6
> >
> > Requires: openafs = 1.2.6 openafs-client = 1.2.6
> > Obsoletes: openafs-client-compat
> > Processing files: openafs-kpasswd-1.2.6-rh7.3.1
> > Finding Provides: /usr/lib/rpm/find-provides
> > Finding Requires: /usr/lib/rpm/find-requires
> > PreReq: rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> rpmlib(CompressedFileNames) <=
> > 3.0.4-1
> > Requires(rpmlib): rpmlib(PayloadFilesHavePrefix) <= 4.0-1
> > rpmlib(CompressedFileNames) <= 3.0.4-1
> > Requires: openafs libc.so.6 libc.so.6(GLIBC_2.0) libc.so.6(GLIBC_2.1)
> > libc.so.6(GLIBC_2.3) libresolv.so.2 libresolv.so.2(GLIBC_2.2)
> > Processing files: openafs-krb5-1.2.6-rh7.3.1
> > error: File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
> > error: File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
> > Requires: openafs = 1.2.6
> >
> > RPM build errors:
> > File not found: /var/tmp/openafs-1.2.6-root/usr/bin/aklog
> > File not found: /var/tmp/openafs-1.2.6-root/usr/sbin/asetkey
> >
> > Help me plase
> >
> > Regards.
> > ------- End of Forwarded Message -------
> >
> >
> >
> > _______________________________________________
> > 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
--
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 security@xauth.net Mon Oct 7 22:24:50 2002
From: security@xauth.net (Charles Clancy)
Date: Mon, 7 Oct 2002 16:24:50 -0500 (CDT)
Subject: [OpenAFS] pam and openafs 1.2.7 for RH 7.2
In-Reply-To: <3D9F2D40.4050704@inf.ethz.ch>
Message-ID:
If you compile --with-kerberos4 and --with-afs, OpenSSH should accept krb4
TGTs, krb4 passwords, or AFS tokens for authentication. OpenSSH will also
grab a PAG for you and run krb_afslog() when you log in. They still need
enabled in your sshd_config.
PAM authentication and krb4 authentication in OpenSSH are completely
independent. There's no reason why adding those options to ./configure
would inhibit PAM from working. In fact, I've compiled and used both
under the same sshd (though on Solaris).
The only think I can possibly think of is that somehow there are library
conflicts between the AFS and krb4 libraries statically linked into your
OpenSSH, versus those statically linked to pam_afs.so, and things go wrong
when OpenSSH (via libpam.so) dlopen's pam_afs.so.
Do other modules work with your kerberos-enabled version of OpenSSH?
In theory, if you have kerberos4 authentication with AFS support, you
don't need PAM. They shouldn't be mutually exclusive, but if they are, it
shouldn't matter, because you only need one or the other.
[ t charles clancy ]--[ tclancy@uiuc.edu ]--[ www.uiuc.edu/~tclancy ]
On Sat, 5 Oct 2002, Marc Schmitt wrote:
> Hi Andi,
>
> Was the sshd version on the alpha machine built --with-afs?
>
> I`m seeing the problem you describe under RedHat 7.3 with
> openafs-1.2.7-rh7.3.1 and openssh-3.4p1-3 (what I changed between
> 3.4p1-2 and 3.4p1-3 is adding "--with-afs=/usr
> --with-kerberos4=/usr/athena" to the configure line, krb4 is version 1.2).
>
> If I use openssh-3.4p1-2, I get:
>
> Oct 5 19:35:14 otherhost sshd(pam_unix)[8281]: session opened for user
> testuser by (uid=0)
>
> If I use openssh-3.4p1-3, I get:
>
> Oct 5 19:47:42 otherhost pam_afs[15855]: AFS Authentication failed for
> user testuser. password was incorrect
> Oct 5 19:47:42 otherhost sshd(pam_unix)[15851]: check pass; user unknown
> Oct 5 19:47:42 otherhost sshd(pam_unix)[15851]: authentication failure;
> logname= uid=0 euid=0 tty=NODEVssh ruser= rhost=somehost
>
> Looking at the debug output of sshd:
>
> With openssh-3.4p1-2:
>
> Oct 5 20:03:53 otherhost sshd[23253]: Failed none for testuser from
> 129.132.10.58 port 35551
> Oct 5 20:03:53 otherhost sshd[23253]: debug3: mm_request_receive entering
> Oct 5 20:03:56 otherhost sshd[23253]: debug3: monitor_read: checking
> request 10
> Oct 5 20:03:56 otherhost sshd[23253]: debug1: PAM Password
> authentication accepted for user "testuser"
> Oct 5 20:03:56 otherhost sshd[23253]: debug3: mm_answer_authpassword:
> sending result 1
> Oct 5 20:03:56 otherhost sshd[23253]: debug3: mm_request_send entering:
> type 11
> Oct 5 20:03:56 otherhost sshd[23253]: debug2: pam_acct_mgmt() = 0
> Oct 5 20:03:56 otherhost sshd[23253]: Accepted password for testuser
> from 129.132.10.58 port 35551
> Oct 5 20:03:56 otherhost sshd[23253]: debug1: monitor_child_preauth:
> testuser has been authenticated by privileged process
>
> and openssh-3.4p1-3:
>
> Oct 5 19:47:39 otherhost sshd[15851]: Failed none for testuser from
> 129.132.10.58 port 35528
> Oct 5 19:47:39 otherhost sshd[15851]: debug3: mm_request_receive entering
> Oct 5 19:47:42 otherhost sshd[15851]: debug3: monitor_read: checking
> request 10
> Oct 5 19:47:44 otherhost sshd[15851]: debug1: PAM Password
> authentication for "testuser" failed[7]: Authentication failure
> Oct 5 19:47:44 otherhost sshd[15851]: debug3: mm_answer_authpassword:
> sending result 0
> Oct 5 19:47:44 otherhost sshd[15851]: debug3: mm_request_send entering:
> type 11
> Oct 5 19:47:44 otherhost sshd[15851]: Failed password for testuser from
> 129.132.10.58 port 35528
>
> PAM authentication fails... but why? Nothing has changed in
> /etc/pam.d/system-auth nor /etc/pam.d/sshd between the two tests.
> Looks like AFS support in OpenSSH bites pam AFS authentication...
>
> Regards,
> Marc
>
>
> Andreas Buechler wrote:
> > Hello,
> >
> > I just installed openafs 1.2.7 on a alpha machine. Everything worked fine
> > (rebuilding and installing the rpm's) and at the end I was told to change
> > the files cacheinfo and ThisCell. I changed both files, now I am able to
> > get tokens etc as root for any afs-user. To be able to login and get a
> > token automatically I changed /etc/pam.d/system-auth as discribed at the end of the
> > installation.
> > Does anybody have an idea why I still cant login via ssh as an afs-user?
> > I posted my sshd and system-auth pam-files at the end of this mail.
> >
> > Thanks for any help and sorry if this message was posted twice!
> >
> > Andi
>
>
> _______________________________________________
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info
>
From cameron@ctc.com Mon Oct 7 23:41:59 2002
From: cameron@ctc.com (Frank J. Cameron)
Date: 07 Oct 2002 18:41:59 -0400
Subject: [OpenAFS] RE: windows issues
In-Reply-To: <20021003201102.B08239DC1@grand.central.org>
References: <20021003201102.B08239DC1@grand.central.org>
Message-ID: <1034030518.32300.16.camel@jsta6243.nsc.ctc.com>
Take a look at the event log (Application log) for any AFS-related
errors. Also, what error code did you get when klog failed?
> Date: Thu, 03 Oct 2002 12:01:05 -0700 (MST)
> From: David Bear
> To: openafs-info@openafs.org
> Subject: [OpenAFS] windows issues
>
>
> strange happenings with win2k and openafs 1.2.2b.
>
> Worked fine yesterday.
>
> today, try to get tokens.. always fails.
>
> Then from cmd window issue:
>
> net stop "IBM AFS Client"
> net start "IBM AFS Client"
>
> stops and starts successfully. Then klog ..
>
> works!!! Why?
>
> --
> David Bear
> College of Public Programs/ASU
> 480-965-8257
> ...the way is like water, going where nobody wants it to go
From chris@chrisdos.com Tue Oct 8 04:05:29 2002
From: chris@chrisdos.com (Chris Dos)
Date: Mon, 07 Oct 2002 21:05:29 -0600
Subject: [OpenAFS] Holy Grail of High Availability
Message-ID: <3DA24B79.60102@chrisdos.com>
I'm been looking at distributed files systems lately and I may have been
under the wrong impression that OpenAFS or Coda could be the Holy Grail
for High Availability for the hosting company that I work for. The part
that worries me right now is the server replication. According to the
documentation that I've been reading, the server replication is only
good for volumes that don't see many files changing. So, replicating a
database such as Oracle or MySQL that changes data often would be a bad
idea, but web sites or mail might be good? Would it replcate the entire
changed file, or just the pieces of the file that has changed. I'm
looking at putting three high end terabyte servers next to each via
Gigabit Ethernet, and having replication take place. Would the
replicants be read only? So all the write changes still have to take
place to one server. If that server goes down, will write changes go to
a server that is still up and running. Also, how does the client know
which server go to if there are three servers with identical data on the
same subnet? Is there any type of load balancing going on to help
distribute the load?
Am I totally off my rocker in thinking AFS might be able to provide all
these things? And if I am loony in thinking AFS or another distributed
file system might be my holy grail, are there any other alternatives I
should be looking at?
Thank you for any insight you might be able to provide. I sincerely
appreciate it.
Chris Dos
From security@xauth.net Tue Oct 8 05:08:04 2002
From: security@xauth.net (Charles Clancy)
Date: Mon, 7 Oct 2002 23:08:04 -0500 (CDT)
Subject: [OpenAFS] Holy Grail of High Availability
In-Reply-To: <3DA24B79.60102@chrisdos.com>
Message-ID:
> So, replicating a database such as Oracle or MySQL that changes data
> often would be a bad idea,
Correct.
> but web sites or mail might be good?
Websites: probably.
Mail spools: probably not.
Websites are semi-static, but mail spools change way too frequently (see
info on how replication works below).
> Would it replcate the entire changed file, or just the pieces of the
> file that has changed.
In AFS, collections of directories are stored in abstract units called
volumes. You can replicate volumes. Volume replicas are updated when the
"vos release volume-name" command is run.
An example scenario would be to have a volume for /afs/@cell/www/htdocs,
and then make several replicas. If you change any files to your site,
you'd need to release that volume, which would consequently release all
changes to your site.
> I'm looking at putting three high end terabyte servers next to each via
> Gigabit Ethernet, and having replication take place. Would the
> replicants be read only?
There is a single RW volume, and its RO replicas are updated, usually by
the person who just made changes to the RW volume, with the "vos release"
command.
> So all the write changes still have to take place to one server.
The writes are made to the RW volume, and that volume can live on any of
the servers.
> If that server goes down, will write changes go to a server that is
> still up and running.
If that server goes down, you won't be able to write. It may take 30
seconds, or so (these values can be changed at compile time), for clients
using the now down server to switch to one of the replicas.
> Also, how does the client know which server go to if there are three
> servers with identical data on the same subnet? Is there any type of
> load balancing going on to help distribute the load?
The clients generally pick one at random. I believe there is a hack that
has them first look for servers on their own subnet first.
> Am I totally off my rocker in thinking AFS might be able to provide all
> these things? And if I am loony in thinking AFS or another distributed
> file system might be my holy grail, are there any other alternatives I
> should be looking at?
AFS is a distributed file system. It works great for sharing files to
user workstations, especially those which may have slow links. Generally
the file system consists of a static set of shared application binaries,
and users' home directories (which aren't replicated for various
implementation reasons I can describe if you're interested).
I don't think of AFS as a very good method of providing HA between
high-speed servers. Its replication model just doesn't fit very well in
that environment. It's much better at replicating [mostly] static data.
[ t charles clancy ]--[ tclancy@uiuc.edu ]--[ www.uiuc.edu/~tclancy ]
From warlord@MIT.EDU Tue Oct 8 05:32:07 2002
From: warlord@MIT.EDU (Derek Atkins)
Date: 08 Oct 2002 00:32:07 -0400
Subject: [OpenAFS] Holy Grail of High Availability
In-Reply-To: