From sxw@inf.ed.ac.uk Mon Aug 1 00:38:24 2011 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Mon, 1 Aug 2011 00:38:24 +0100 Subject: [OpenAFS] Unable to delete volume In-Reply-To: References: Message-ID: <3A66E280-03AE-4E36-B1C1-A23CC83BB327@inf.ed.ac.uk> --Apple-Mail-2-988340933 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii On 31 Jul 2011, at 21:15, Aaron Knister wrote: > Make that a vos syncserv not a syncvldb If I'm understanding your problem correctly you have a single entry for a RO= volume in your VLDB, with the corresponding server long since departed. In t= his case, vos delentry should let you remove it from the VLDB. S. --Apple-Mail-2-988340933 Content-Transfer-Encoding: 7bit Content-Type: text/html; charset=utf-8


On 31 Jul 2011, at 21:15, Aaron Knister <aaronk@umbc.edu> wrote:

Make that a vos syncserv not a syncvldb

If I'm understanding your problem correctly you have a single entry for a RO volume in your VLDB, with the corresponding server long since departed. In this case, vos delentry should let you remove it from the VLDB.

S.

--Apple-Mail-2-988340933-- From aaronk@umbc.edu Mon Aug 1 01:30:06 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Sun, 31 Jul 2011 20:30:06 -0400 Subject: [OpenAFS] Unable to delete volume In-Reply-To: <3A66E280-03AE-4E36-B1C1-A23CC83BB327@inf.ed.ac.uk> References: <3A66E280-03AE-4E36-B1C1-A23CC83BB327@inf.ed.ac.uk> Message-ID: <8FD9D8AE-97B9-4BB2-B076-4D2343F8B04A@umbc.edu> --Apple-Mail-7-991446018 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii That's what I thought, however delent doesn't seem to work on RO volumes IIR= C. Whatever the reason, delent fails.=20 Sent from my iPhone On Jul 31, 2011, at 7:38 PM, Simon Wilkinson wrote: >=20 >=20 > On 31 Jul 2011, at 21:15, Aaron Knister wrote: >=20 >> Make that a vos syncserv not a syncvldb >=20 > If I'm understanding your problem correctly you have a single entry for a R= O volume in your VLDB, with the corresponding server long since departed. In= this case, vos delentry should let you remove it from the VLDB. >=20 > S. >=20 --Apple-Mail-7-991446018 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=utf-8
That's what I thought, however delent d= oesn't seem to work on RO volumes IIRC. Whatever the reason, delent fails.&n= bsp;

Sent from my iPhone

On Jul 31, 2011, at 7:38 PM, S= imon Wilkinson <sxw@inf.ed.ac.uk&= gt; wrote:

=

On 31 Jul 2011, at 21:15, Aaron Knister <aaronk@umbc.edu> wrote:

Make that a vos syncserv not a syncvldb

If I'm understanding your problem co= rrectly you have a single entry for a RO volume in your VLDB, with the corre= sponding server long since departed. In this case, vos delentry should let y= ou remove it from the VLDB.

S.

=
= --Apple-Mail-7-991446018-- From adeason@sinenomine.net Mon Aug 1 01:54:36 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Sun, 31 Jul 2011 19:54:36 -0500 Subject: [OpenAFS] Re: Unable to delete volume References: Message-ID: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> On Sun, 31 Jul 2011 16:09:36 -0400 Aaron Knister wrote: > Something I did just now was change the address of the non-existent > server on the subnet we no-loner use to a temporary afs file server on > a subnet we are using: > > vos changeaddr -oldaddr 130.85.60.43 -newaddr 130.85.24.58 > > I did read the caveats in the man page before running the above > command. This at least points references to the non-existent > fileserver to an existing server. I was hoping I could do a syncvldb > and it would clean up the reference to the invalid volume but to no > avail. Please never do this. I don't know how to be more clear about it: do not use 'vos changeaddr' to change a fileserver address. There are _no_ situations where there is any reason to use it in that manner, unless you are dealing with very old fileservers that predate the multihoming mechanism. When you use it to deal with server addresses for servers that actually still exist (either new or old addr), you may (depending on the versions in play) duplicate server entries in the VLDB, making it somewhat difficult to get the VLDB back into its original state. I don't know if you've done that and I don't have time to look at it now, but if you don't notice any problems then you're probably fine. As Simon indicates, what you want 'vos delentry', for the entry that 'vos listvldb' showed you. -- Andrew Deason adeason@sinenomine.net From gsgatlin@ncsu.edu Mon Aug 1 02:09:52 2011 From: gsgatlin@ncsu.edu (Gary Gatling) Date: Sun, 31 Jul 2011 21:09:52 -0400 (EDT) Subject: [OpenAFS] Unable to delete volume In-Reply-To: References: 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. --26749252-410754994-1312160992=:14790 Content-Type: TEXT/PLAIN; charset=iso-8859-1; format=flowed Content-Transfer-Encoding: 8BIT What happens if you use the old hostname or the old ip address? Whatever the vldb gives indications that it "thinks" it is on? (Same thing with the partition) If it "thinks" it s a hostname, use that one. If it "thinks" its an IP address, use that one. I think vos remsite only works with the vldb. Maybe you might have to bring that old afs server back up? (I'm kind of thinking no on that though) Just a shot in the dark. Sorry if it doesn't work. Good luck with that. I would think one of vos remsite or vos remove or vos delentry should work though. (with whatver the vldb thinks its on) Gary Gatling | ITECS Systems On Sun, 31 Jul 2011, Aaron Knister wrote: > Gary, > > Thanks for the suggestion, unfortunately that didn't work: > > $ vos remsite -server 130.85.24.58 -partition /vicepa -id 536988828 >  Could not lock the VLDB entry for volume 536988828 > VLDB: no such entry > Error in vos remsite command. > VLDB: no such entry > > Something I did just now was change the address of the non-existent server > on the subnet we no-loner use to a temporary afs file server on a subnet we > are using: > > vos changeaddr -oldaddr 130.85.60.43 -newaddr 130.85.24.58 > > I did read the caveats in the man page before running the above command. > This at least points references to the non-existent fileserver to an > existing server. I was hoping I could do a syncvldb and it would clean up > the reference to the invalid volume but to no avail. > > -Aaron > > On Sun, Jul 31, 2011 at 3:38 PM, Gary Gatling wrote: > > On Sun, 31 Jul 2011, Aaron Knister wrote: > > There's a server that was decommissioned many moons ago > that I'm trying to > delete. The problem is there's a volume still associated > with it [1], a RO > volume at that. When I attempt to remove the volume usin > its volume id I get > this message: > > > Try vos remsite: > > http://www.inf.ed.ac.uk/systems/AFS/1/vos_remsite.html > > I think you will want to give it a volumeID intead of a volume name. > > Hope that helps, > > Gary Gatling      | ITECS Systems > > > --BEGIN-- > Could not delete entry for volume 536988828 > You must specify a RW volume name or ID (the entire VLDB entry > will be > deleted) > VLDB: no such entry > Deleted 0 VLDB entries > --END-- > > Here's the output of vos listvldb -server for the server in > question: > > --BEGIN-- > VLDB entries for server 130.85.60.43 > > g.ilin22.local >     ROnly: 536988828 >     number of sites -> 1 >        server 130.85.60.43 partition /vicepa RO Site > > Total entries: 1 > --END-- > > To make matters more interesting, there's another volume of the > same name > but with a different volume ID. > > Does anyone have any suggestions? > > -Aaron > > -- > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu > > > > > -- > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu > > --26749252-410754994-1312160992=:14790-- From shadow@dementia.org Mon Aug 1 02:23:18 2011 From: shadow@dementia.org (Derrick Brashear) Date: Sun, 31 Jul 2011 21:23:18 -0400 Subject: [OpenAFS] Unable to delete volume In-Reply-To: References: Message-ID: --Apple-Mail-5-994637277 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii On Jul 31, 2011, at 14:51, Aaron Knister wrote: > There's a server that was decommissioned many moons ago that I'm trying to= delete. The problem is there's a volume still associated with it [1], a RO v= olume at that. When I attempt to remove the volume usin its volume id I get t= his message: >=20 > --BEGIN-- > Could not delete entry for volume 536988828 > You must specify a RW volume name or ID (the entire VLDB entry will be del= eted) so, what happened when you tried that? > VLDB: no such entry > Deleted 0 VLDB entries > --END-- >=20 > Here's the output of vos listvldb -server for the server in question: >=20 > --BEGIN-- > VLDB entries for server 130.85.60.43=20 >=20 > g.ilin22.local=20 > ROnly: 536988828=20 > number of sites -> 1 > server 130.85.60.43 partition /vicepa RO Site=20 >=20 > Total entries: 1 > --END-- >=20 > To make matters more interesting, there's another volume of the same name b= ut with a different volume ID. >=20 > Does anyone have any suggestions? >=20 > -Aaron >=20 > --=20 > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu --Apple-Mail-5-994637277 Content-Transfer-Encoding: 7bit Content-Type: text/html; charset=utf-8




On Jul 31, 2011, at 14:51, Aaron Knister <aaronk@umbc.edu> wrote:

There's a server that was decommissioned many moons ago that I'm trying to delete. The problem is there's a volume still associated with it [1], a RO volume at that. When I attempt to remove the volume usin its volume id I get this message:

--BEGIN--
Could not delete entry for volume 536988828
You must specify a RW volume name or ID (the entire VLDB entry will be deleted)

so, what happened when you tried that?


VLDB: no such entry
Deleted 0 VLDB entries
--END--

Here's the output of vos listvldb -server for the server in question:

--BEGIN--
VLDB entries for server 130.85.60.43

g.ilin22.local
    ROnly: 536988828
    number of sites -> 1
       server 130.85.60.43 partition /vicepa RO Site

Total entries: 1
--END--

To make matters more interesting, there's another volume of the same name but with a different volume ID.

Does anyone have any suggestions?

-Aaron

--
Aaron Knister
Systems Administrator
Division of Information Technology
University of Maryland, Baltimore County
aaronk@umbc.edu
--Apple-Mail-5-994637277-- From aaronk@umbc.edu Mon Aug 1 03:24:39 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Sun, 31 Jul 2011 22:24:39 -0400 Subject: [OpenAFS] Re: Unable to delete volume In-Reply-To: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> Message-ID: On Jul 31, 2011, at 8:54 PM, Andrew Deason wrote: > On Sun, 31 Jul 2011 16:09:36 -0400 > Aaron Knister wrote: >=20 >> Something I did just now was change the address of the non-existent >> server on the subnet we no-loner use to a temporary afs file server on >> a subnet we are using: >>=20 >> vos changeaddr -oldaddr 130.85.60.43 -newaddr 130.85.24.58 >>=20 >> I did read the caveats in the man page before running the above >> command. This at least points references to the non-existent >> fileserver to an existing server. I was hoping I could do a syncvldb >> and it would clean up the reference to the invalid volume but to no >> avail. >=20 > Please never do this. I don't know how to be more clear about it: do not > use 'vos changeaddr' to change a fileserver address. There are _no_ > situations where there is any reason to use it in that manner, unless > you are dealing with very old fileservers that predate the multihoming > mechanism. My thought process was this: there was one volume (the un deletable volume) a= ssociated with the oldaddr and no volumes with the newaddr. There were also n= o protection groups containing either machine address. The oldaddr is on a s= ubnet that I can't get access to so by changing oldaddr to newaddr which us o= n a reachable subnet I was hoping to do a syncserv to clear out the phantom e= ntry. > When you use it to deal with server addresses for servers > that actually still exist (either new or old addr), you may (depending > on the versions in play) duplicate server entries in the VLDB, making it > somewhat difficult to get the VLDB back into its original state. This wasn't one of the caveats mentioned in the man page. Had I known that I= likely wouldn't have proceeded.=20 > I don't > know if you've done that and I don't have time to look at it now, but if > you don't notice any problems then you're probably fine. >=20 > As Simon indicates, what you want 'vos delentry', for the entry that > 'vos listvldb' showed you. vos delentry doesn't work-- I tried that a bunch of times before I posted to= the list. Heres the error I get: Could not delete entry for volume 536988828 You must specify a RW volume name or ID (the entire VLDB entry will be delet= ed) VLDB: no such entry Deleted 0 VLDB entries Thanks! >=20 > --=20 > Andrew Deason > adeason@sinenomine.net >=20 > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info From adeason@sinenomine.net Mon Aug 1 15:18:31 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 1 Aug 2011 09:18:31 -0500 Subject: [OpenAFS] Re: Unable to delete volume References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> Message-ID: <20110801091831.3539844a.adeason@sinenomine.net> On Sun, 31 Jul 2011 22:24:39 -0400 Aaron Knister wrote: > > As Simon indicates, what you want 'vos delentry', for the entry that > > 'vos listvldb' showed you. > > vos delentry doesn't work-- I tried that a bunch of times before I > posted to the list. Heres the error I get: > > Could not delete entry for volume 536988828 > You must specify a RW volume name or ID (the entire VLDB entry will be deleted) > VLDB: no such entry > Deleted 0 VLDB entries You can't delete by non-RW id. Just give the volume name like the message says, and it will delete the entry. -- Andrew Deason adeason@sinenomine.net From aaronk@umbc.edu Mon Aug 1 15:40:48 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Mon, 1 Aug 2011 10:40:48 -0400 Subject: [OpenAFS] Re: Unable to delete volume In-Reply-To: <20110801091831.3539844a.adeason@sinenomine.net> References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> <20110801091831.3539844a.adeason@sinenomine.net> Message-ID: --20cf306845e528fd4104a9729f97 Content-Type: text/plain; charset=ISO-8859-1 On Mon, Aug 1, 2011 at 10:18 AM, Andrew Deason wrote: > On Sun, 31 Jul 2011 22:24:39 -0400 > Aaron Knister wrote: > > > > As Simon indicates, what you want 'vos delentry', for the entry that > > > 'vos listvldb' showed you. > > > > vos delentry doesn't work-- I tried that a bunch of times before I > > posted to the list. Heres the error I get: > > > > Could not delete entry for volume 536988828 > > You must specify a RW volume name or ID (the entire VLDB entry will be > deleted) > > VLDB: no such entry > > Deleted 0 VLDB entries > > You can't delete by non-RW id. Just give the volume name like the > message says, and it will delete the entry. > > I can't, because as mentioned the volume name is also associated with another volume id that I definitely don't want to delete. To illustrate what I mean, 130.85.24.58 is the server I want to delete and the RO volume associated with it that I want to delete is named g.ilin22.local: ------- $ vos listvldb -server 130.85.24.58 VLDB entries for server 130.85.24.58 g.ilin22.local ROnly: 536988828 number of sites -> 1 server solaris10-test1.core.umbc.edu partition /vicepa RO Site Total entries: 1 ------- However, if I query the vldb for the volume g.ilin22.local you'll see (below) that it and its RO replicas exist with a different set of volume IDs. As I mentioned, I can't delete this volume by name. Doing so would be very bad for me. ------- vos listvldb g.ilin22.local g.ilin22.local RWrite: 537195552 ROnly: 537195559 Backup: 537195686 number of sites -> 3 server ifs1.afs.umbc.edu partition /vicepd RW Site server ifs1.afs.umbc.edu partition /vicepd RO Site server ifs2.afs.umbc.edu partition /vicepd RO Site ------- -- > Andrew Deason > adeason@sinenomine.net > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --20cf306845e528fd4104a9729f97 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
On Mon, Aug 1, 2011 at 10:18 AM, Andrew Deason <= span dir=3D"ltr"><adeason@sine= nomine.net> wrote:
On Sun, 31 Jul 2011 22:24:39 -0400
Aaron Knister <aaro= nk@umbc.edu> wrote:

> > As Simon indicates, what you want 'vo= s delentry', for the entry that
> > 'vos listvldb' showed you.
>
> vos delentry doesn't work-- I tried that a bunch of times before I=
> posted to the list. Heres the error I get:
>
> Could not delete entry for volume 536988828
> You must specify a RW volume name or ID (the entire VLDB entry will be= deleted)
> VLDB: no such entry
> Deleted 0 VLDB entries

You can't delete by non-RW id. Just give the volume name like the=
message says, and it will delete the entry.


I can't, because as mentioned the volume name is also associate= d with another volume id that I definitely don't want to delete. To ill= ustrate what I mean, 130.85.24.58 is the server I want to delete and the RO= volume associated with it that I want to delete is named g.ilin22.local:

-------
$=A0vos listvldb -server 130.85.24.58=
VLDB entries for server 130.85.24.58=A0

g.ilin22.local=A0
=A0 =A0 ROnly: 536988828=A0
=A0 =A0 = number of sites -> 1
=A0 =A0 =A0 =A0server solaris10-test1.core.umbc.edu partition /vicepa RO Site=A0
<= br>
Total entries: 1
-------

H= owever, if I query the vldb for the volume g.ilin22.local you'll see (b= elow) that it and its RO replicas exist with a different set of volume IDs.= As I mentioned, I can't delete this volume by name. Doing so would be = very bad for me.

-------
vos listvldb g.ilin22.local

g.ilin22.local=A0
=A0 =A0 RWrite: 537195552 =A0 = =A0 ROnly: 537195559 =A0 =A0 Backup: 537195686=A0
=A0 =A0 number = of sites -> 3
=A0 =A0 =A0 =A0server ifs1.afs.um= bc.edu partition /vicepd RW Site=A0
=A0 =A0 =A0 =A0server ifs1.afs.umbc.edu partition /vicepd R= O Site=A0
=A0 =A0 =A0 =A0server ifs2.afs.umbc.ed= u partition /vicepd RO Site=A0
-------=A0

--
Andrew Deason
adeason@sinenomine.net

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



--
Aaron Knist= er
Systems Administrator
Division of Information Technology
Univer= sity of Maryland, Baltimore County
aaronk@umbc.edu
--20cf306845e528fd4104a9729f97-- From shadow@gmail.com Mon Aug 1 15:45:42 2011 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 1 Aug 2011 10:45:42 -0400 Subject: [OpenAFS] Re: Unable to delete volume In-Reply-To: References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> <20110801091831.3539844a.adeason@sinenomine.net> Message-ID: On Mon, Aug 1, 2011 at 10:40 AM, Aaron Knister wrote: > On Mon, Aug 1, 2011 at 10:18 AM, Andrew Deason > wrote: >> >> On Sun, 31 Jul 2011 22:24:39 -0400 >> Aaron Knister wrote: >> >> > > As Simon indicates, what you want 'vos delentry', for the entry that >> > > 'vos listvldb' showed you. >> > >> > vos delentry doesn't work-- I tried that a bunch of times before I >> > posted to the list. Heres the error I get: >> > >> > Could not delete entry for volume 536988828 >> > You must specify a RW volume name or ID (the entire VLDB entry will be >> > deleted) >> > VLDB: no such entry >> > Deleted 0 VLDB entries >> >> You can't delete by non-RW id. Just give the volume name like the >> message says, and it will delete the entry. >> > > I can't, because as mentioned the volume name is also associated with > another volume id that I definitely don't want to delete. To illustrate w= hat > I mean, 130.85.24.58 is the server I want to delete and the RO volume > associated with it that I want to delete is named g.ilin22.local: > ------- > $=A0vos listvldb -server 130.85.24.58 > VLDB entries for server 130.85.24.58 > g.ilin22.local > =A0 =A0 ROnly: 536988828 > =A0 =A0 number of sites -> 1 > =A0 =A0 =A0 =A0server solaris10-test1.core.umbc.edu partition /vicepa RO = Site > Total entries: 1 ah, so, your problem is this id isn't the RO id associated with the same name below. wild guess: vos ex 536988827 and see what it is. if nothing, try vos delentry 536988827 and see if that's more helpful. would be sad if it is.... > ------- > However, if I query the vldb for the volume g.ilin22.local you'll see > (below) that it and its RO replicas exist with a different set of volume > IDs. As I mentioned, I can't delete this volume by name. Doing so would b= e > very bad for me. > ------- > vos listvldb g.ilin22.local > g.ilin22.local > =A0 =A0 RWrite: 537195552 =A0 =A0 ROnly: 537195559 =A0 =A0 Backup: 537195= 686 > =A0 =A0 number of sites -> 3 > =A0 =A0 =A0 =A0server ifs1.afs.umbc.edu partition /vicepd RW Site > =A0 =A0 =A0 =A0server ifs1.afs.umbc.edu partition /vicepd RO Site > =A0 =A0 =A0 =A0server ifs2.afs.umbc.edu partition /vicepd RO Site > ------- >> >> -- >> Andrew Deason >> adeason@sinenomine.net >> >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info > > > > -- > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu > --=20 Derrick From aaronk@umbc.edu Mon Aug 1 15:53:09 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Mon, 1 Aug 2011 10:53:09 -0400 Subject: [OpenAFS] Re: Unable to delete volume In-Reply-To: References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> <20110801091831.3539844a.adeason@sinenomine.net> Message-ID: --bcaec501625b5049de04a972cb5a Content-Type: text/plain; charset=ISO-8859-1 Hi Derrick, Thanks for the suggestion. I tried that yesterday but to no avail: [aaronk@linux2:~ $ ] vos examine 536988827 Volume 536988827 does not exist in VLDB Dump only information from VLDB g.ilin22.local ROnly: 536988828 number of sites -> 1 server solaris10-test1.core.umbc.edu partition /vicepa RO Site [aaronk@linux2:~ $ ] vos delentry 536988827 Could not delete entry for volume 536988827 You must specify a RW volume name or ID (the entire VLDB entry will be deleted) VLDB: no such entry Deleted 0 VLDB entries I also tried Andrew's suggestion after realizing I could specify a -server argument so it wouldn't clobber the other volume by the same name but that didn't work either: [aaronk@linux2:~ $ ] vos delent -prefix g.ilin22.local -server 130.85.24.58 -partition /vicepa Deleting VLDB entries for server 130.85.24.58 partition /vicepa which are prefixed with g.ilin22.local Could not delete VDLB entry for g.ilin22.local VLDB: no such entry ---------------------- Total VLDB entries deleted: 0; failed to delete: 1 -Aaron On Mon, Aug 1, 2011 at 10:45 AM, Derrick Brashear wrote: > On Mon, Aug 1, 2011 at 10:40 AM, Aaron Knister wrote: > > On Mon, Aug 1, 2011 at 10:18 AM, Andrew Deason > > wrote: > >> > >> On Sun, 31 Jul 2011 22:24:39 -0400 > >> Aaron Knister wrote: > >> > >> > > As Simon indicates, what you want 'vos delentry', for the entry that > >> > > 'vos listvldb' showed you. > >> > > >> > vos delentry doesn't work-- I tried that a bunch of times before I > >> > posted to the list. Heres the error I get: > >> > > >> > Could not delete entry for volume 536988828 > >> > You must specify a RW volume name or ID (the entire VLDB entry will be > >> > deleted) > >> > VLDB: no such entry > >> > Deleted 0 VLDB entries > >> > >> You can't delete by non-RW id. Just give the volume name like the > >> message says, and it will delete the entry. > >> > > > > I can't, because as mentioned the volume name is also associated with > > another volume id that I definitely don't want to delete. To illustrate > what > > I mean, 130.85.24.58 is the server I want to delete and the RO volume > > associated with it that I want to delete is named g.ilin22.local: > > ------- > > $ vos listvldb -server 130.85.24.58 > > VLDB entries for server 130.85.24.58 > > g.ilin22.local > > ROnly: 536988828 > > number of sites -> 1 > > server solaris10-test1.core.umbc.edu partition /vicepa RO Site > > Total entries: 1 > > ah, so, your problem is this id isn't the RO id associated with the > same name below. > wild guess: > vos ex 536988827 and see what it is. if nothing, try vos delentry > 536988827 and see > if that's more helpful. would be sad if it is.... > > > ------- > > However, if I query the vldb for the volume g.ilin22.local you'll see > > (below) that it and its RO replicas exist with a different set of volume > > IDs. As I mentioned, I can't delete this volume by name. Doing so would > be > > very bad for me. > > ------- > > vos listvldb g.ilin22.local > > g.ilin22.local > > RWrite: 537195552 ROnly: 537195559 Backup: 537195686 > > number of sites -> 3 > > server ifs1.afs.umbc.edu partition /vicepd RW Site > > server ifs1.afs.umbc.edu partition /vicepd RO Site > > server ifs2.afs.umbc.edu partition /vicepd RO Site > > ------- > >> > >> -- > >> Andrew Deason > >> adeason@sinenomine.net > >> > >> _______________________________________________ > >> OpenAFS-info mailing list > >> OpenAFS-info@openafs.org > >> https://lists.openafs.org/mailman/listinfo/openafs-info > > > > > > > > -- > > Aaron Knister > > Systems Administrator > > Division of Information Technology > > University of Maryland, Baltimore County > > aaronk@umbc.edu > > > > > > -- > Derrick > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --bcaec501625b5049de04a972cb5a Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi Derrick,

Thanks for the suggestion. I tried that yest= erday but to no avail:

[aaronk@linux2:~ $ ] v= os examine 536988827
Volume 536988827 does not exist in VLDB

Dump only information from VLDB

g.ilin22.local=A0
=A0 =A0 ROnly: 536988828=A0
=A0 =A0= number of sites -> 1
=A0 =A0 =A0 =A0server solaris10-test1.core.umbc.edu partition = /vicepa RO Site=A0

[aaronk@linux2:~ $ ] vos delentry 536988827
C= ould not delete entry for volume 536988827
You must specify a RW = volume name or ID (the entire VLDB entry will be deleted)
VLDB: n= o such entry
Deleted 0 VLDB entries

I also tried Andrew= 9;s suggestion after realizing I could specify a -server argument so it wou= ldn't clobber the other volume by the same name but that didn't wor= k either:

[aaronk@linux2:~ $ ] vos delent -prefix g.ilin22.l= ocal -server 130.85.24.58 -partition /vicepa
Deleting VLDB entrie= s for server 130.85.24.58 partition /vicepa which are prefixed with g.ilin2= 2.local=A0
Could not delete VDLB entry for =A0g.ilin22.local
VLDB: no s= uch entry
----------------------
Total VLDB entries del= eted: 0; failed to delete: 1

-Aaron

On Mon, Aug 1, 2011 at 10:45 AM, Derrick Brashea= r <shadow@gmail.co= m> wrote:
On Mon, Aug 1, 2011 at 10:40 AM, Aaron Kn= ister <aaronk@umbc.edu> wrote:=
> On Mon, Aug 1, 2011 at 10:18 AM, Andrew Deason <adeason@sinenomine.net>
> wrote:
>>
>> On Sun, 31 Jul 2011 22:24:39 -0400
>> Aaron Knister <aaronk@umbc.e= du> wrote:
>>
>> > > As Simon indicates, what you want 'vos delentry'= , for the entry that
>> > > 'vos listvldb' showed you.
>> >
>> > vos delentry doesn't work-- I tried that a bunch of times= before I
>> > posted to the list. Heres the error I get:
>> >
>> > Could not delete entry for volume 536988828
>> > You must specify a RW volume name or ID (the entire VLDB entr= y will be
>> > deleted)
>> > VLDB: no such entry
>> > Deleted 0 VLDB entries
>>
>> You can't delete by non-RW id. Just give the volume name like = the
>> message says, and it will delete the entry.
>>
>
> I can't, because as mentioned the volume name is also associated w= ith
> another volume id that I definitely don't want to delete. To illus= trate what
> I mean, 130.85.24.58 is the server I want to delete and the RO volume<= br> > associated with it that I want to delete is named g.ilin22.local:
> -------
> $=A0vos listvldb -server 130.85.24.58
> VLDB entries for server 130.85.24.58
> g.ilin22.local
> =A0 =A0 ROnly: 536988828
> =A0 =A0 number of sites -> 1
> =A0 =A0 =A0 =A0server solaris10-test1.core.umbc.edu partition /vicepa RO S= ite
> Total entries: 1

ah, so, your problem is this id isn't the RO id associated = with the
same name below.
wild guess:
vos ex 536988827 and see what it is. if nothing, try vos delentry
536988827 and see
if that's more helpful. would be sad if it is....

> -------
> However, if I query the vldb for the volume g.ilin22.local you'll = see
> (below) that it and its RO replicas exist with a different set of volu= me
> IDs. As I mentioned, I can't delete this volume by name. Doing so = would be
> very bad for me.
> -------
> vos listvldb g.ilin22.local
> g.ilin22.local
> =A0 =A0 RWrite: 537195552 =A0 =A0 ROnly: 537195559 =A0 =A0 Backup: 537= 195686
> =A0 =A0 number of sites -> 3
> =A0 =A0 =A0 =A0server ifs1.afs.umbc.edu partition /vicepd RW Site
> =A0 =A0 =A0 =A0server ifs1.afs.umbc.edu partition /vicepd RO Site
> =A0 =A0 =A0 =A0server ifs2.afs.umbc.edu partition /vicepd RO Site
> -------
>>
>> --
>> Andrew Deason
>> adeason@sinenomine.net
>>
>> _______________________________________________
>> OpenAFS-info mailing list
>> OpenAFS-info@openafs.o= rg
>> https://lists.openafs.org/mailman/listinfo/openafs-info=
>
>
>
> --
> Aaron Knister
> Systems Administrator
> Division of Information Technology
> University of Maryland, Baltimore County
> aaronk@umbc.edu
>



--
Derrick



--
Aaron KnisterSystems Administrator
Division of Information Technology
University = of Maryland, Baltimore County
aaronk@umbc.edu
--bcaec501625b5049de04a972cb5a-- From adeason@sinenomine.net Mon Aug 1 15:59:19 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 1 Aug 2011 09:59:19 -0500 Subject: [OpenAFS] Re: Unable to delete volume References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> <20110801091831.3539844a.adeason@sinenomine.net> Message-ID: <20110801095919.c2a07878.adeason@sinenomine.net> On Mon, 1 Aug 2011 10:45:42 -0400 Derrick Brashear wrote: > > I can't, because as mentioned the volume name is also associated > > with another volume id that I definitely don't want to delete. To > > illustrate what I mean, 130.85.24.58 is the server I want to delete > > and the RO volume associated with it that I want to delete is named > > g.ilin22.local: Well, your vldb is mildly broken, then. Do you know how the other volume got created? Our volume creation tools should prevent you from duplicating a volume name. And 'vos delentry' doesn't delete volumes. You could just 'vos delentry' the volume twice, and presumably that will get rid of the 'bad' one. Afterwards, you can run syncserv/syncvldb/'addsite -valid' to get the entry back for the 'good' one. But what Derrick suggests is what you should try first: > > ------- > > $ vos listvldb -server 130.85.24.58 > > VLDB entries for server 130.85.24.58 > > g.ilin22.local > >     ROnly: 536988828 > >     number of sites -> 1 > >        server solaris10-test1.core.umbc.edu partition /vicepa RO Site > > Total entries: 1 > > ah, so, your problem is this id isn't the RO id associated with the > same name below. > wild guess: > vos ex 536988827 and see what it is. if nothing, try vos delentry > 536988827 and see vos listvl 536988827 Since the server it's on doesn't exist, or the reported server doesn't have that volume, so 'examine' is going to hang / yell at you. The 'listvl' will show you the same entry for 536988828 if 536988827 is the RW you want, so just 'delentry' it if so. -- Andrew Deason adeason@sinenomine.net From shadow@gmail.com Mon Aug 1 16:03:01 2011 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 1 Aug 2011 11:03:01 -0400 Subject: [OpenAFS] Re: Unable to delete volume In-Reply-To: <20110801095919.c2a07878.adeason@sinenomine.net> References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> <20110801091831.3539844a.adeason@sinenomine.net> <20110801095919.c2a07878.adeason@sinenomine.net> Message-ID: On Mon, Aug 1, 2011 at 10:59 AM, Andrew Deason wrote: > On Mon, 1 Aug 2011 10:45:42 -0400 > Derrick Brashear wrote: > >> > I can't, because as mentioned the volume name is also associated >> > with another volume id that I definitely don't want to delete. To >> > illustrate what I mean, 130.85.24.58 is the server I want to delete >> > and the RO volume associated with it that I want to delete is named >> > g.ilin22.local: > > Well, your vldb is mildly broken, then. Do you know how the other volume > got created? Our volume creation tools should prevent you from > duplicating a volume name. wild guess: restored ro to alternate id while the rw had no clone; failed removal, later clone of rw. alternately: that was the old id for the ro and a restore later was done to that id because that ued to be the id before deletion. but yes, we should save you from that. > And 'vos delentry' doesn't delete volumes. You could just 'vos delentry' > the volume twice, and presumably that will get rid of the 'bad' one. > Afterwards, you can run syncserv/syncvldb/'addsite -valid' to get the > entry back for the 'good' one. and that's probably the simplest answer. -- Derrick From adeason@sinenomine.net Mon Aug 1 16:21:01 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 1 Aug 2011 10:21:01 -0500 Subject: [OpenAFS] Re: Unable to delete volume References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> <20110801091831.3539844a.adeason@sinenomine.net> Message-ID: <20110801102101.0d8b7fa2.adeason@sinenomine.net> On Mon, 1 Aug 2011 10:53:09 -0400 Aaron Knister wrote: > I also tried Andrew's suggestion after realizing I could specify a > -server argument so it wouldn't clobber the other volume by the same > name but that didn't work either: Well, that wasn't my suggestion, and yeah, that's not going to work. Using those flags is just a convenience thing and isn't going to let you do anything a regular 'delent' can't do; they're just a shortcut from grepping 'vos listvl' and running 'vos delent' on the parsed names. If you give derrick or me or someone a copy of your vldb.DB0, we might be able to tell you what you can delete... but really, the easiest way is just to delete both vldb entries. If you really really don't want to temporarily delete the other one, you could just rename it ('vos rename') out of the way, and then vos delentry, and rename it back. -- Andrew Deason adeason@sinenomine.net From aaronk@umbc.edu Mon Aug 1 16:33:49 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Mon, 1 Aug 2011 11:33:49 -0400 Subject: [OpenAFS] Re: Unable to delete volume In-Reply-To: <20110801102101.0d8b7fa2.adeason@sinenomine.net> References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> <20110801091831.3539844a.adeason@sinenomine.net> <20110801102101.0d8b7fa2.adeason@sinenomine.net> Message-ID: --bcaec51d2552bd679c04a9735cec Content-Type: text/plain; charset=ISO-8859-1 On Mon, Aug 1, 2011 at 11:21 AM, Andrew Deason wrote: > On Mon, 1 Aug 2011 10:53:09 -0400 > Aaron Knister wrote: > > > I also tried Andrew's suggestion after realizing I could specify a > > -server argument so it wouldn't clobber the other volume by the same > > name but that didn't work either: > > Well, that wasn't my suggestion, and yeah, that's not going to work. > Using those flags is just a convenience thing and isn't going to let you > do anything a regular 'delent' can't do; they're just a shortcut from > grepping 'vos listvl' and running 'vos delent' on the parsed names. > > If you give derrick or me or someone a copy of your vldb.DB0, we might > be able to tell you what you can delete... but really, the easiest way > is just to delete both vldb entries. If you really really don't want to > temporarily delete the other one, you could just rename it ('vos > rename') out of the way, and then vos delentry, and rename it back. > [aaronk@linux2:~ $ ] vos rename g.ilin22.local g.ilin22.local.tmp Renamed volume g.ilin22.local to g.ilin22.local.tmp [aaronk@linux2:~ $ ] vos delent g.ilin22.local VLDB: no such entry Deleted 0 VLDB entries [aaronk@linux2:~ $ ] vos rename g.ilin22.local.tmp g.ilin22.local Renamed volume g.ilin22.local.tmp to g.ilin22.local sigh...the vldb clearly is unhappy. I'll send you a copy of the vldb.DB0 Thanks for your help > > -- > Andrew Deason > adeason@sinenomine.net > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --bcaec51d2552bd679c04a9735cec Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
On Mon, Aug 1, 2011 at 11:21 AM, Andrew Dea= son <adeason= @sinenomine.net> wrote:
On Mon, 1 Aug 2011 10:53:09 -0400
Aaron Knister <aaro= nk@umbc.edu> wrote:

> I also tried Andrew's suggestion after rea= lizing I could specify a
> -server argument so it wouldn't clobber the other volume by the sa= me
> name but that didn't work either:

Well, that wasn't my suggestion, and yeah, that's not going t= o work.
Using those flags is just a convenience thing and isn't going to let yo= u
do anything a regular 'delent' can't do; they're just a sho= rtcut from
grepping 'vos listvl' and running 'vos delent' on the parse= d names.

If you give derrick or me or someone a copy of your vldb.DB0, we might
be able to tell you what you can delete... but really, the easiest way
is just to delete both vldb entries. If you really really don't want to=
temporarily delete the other one, you could just rename it ('vos
rename') out of the way, and then vos delentry, and rename it back.
=

[aaronk@linux2:~ $ ] vos rename g.ili= n22.local g.ilin22.local.tmp
Renamed volume g.ilin22.local to g.i= lin22.local.tmp
[aaronk@linux2:~ $ ] vos delent g.ilin22.local
VLDB: no such= entry
Deleted 0 VLDB entries
[aaronk@linux2:~ $ ] vos = rename g.ilin22.local.tmp g.ilin22.local
Renamed volume g.ilin22.= local.tmp to g.ilin22.local

sigh...the vldb clearly is unhappy. I'll send= you a copy of the vldb.DB0

Thanks for your help
=A0

--
Andrew Deason
adeason@sinenomine.net

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



--
Aaron Knist= er
Systems Administrator
Division of Information Technology
Univer= sity of Maryland, Baltimore County
aaronk@umbc.edu
--bcaec51d2552bd679c04a9735cec-- From tmaher@watson.org Mon Aug 1 22:05:33 2011 From: tmaher@watson.org (Tom Maher) Date: Mon, 1 Aug 2011 14:05:33 -0700 Subject: [OpenAFS] Plans for IPv6 In-Reply-To: <59582E0B-9562-4AE3-9D31-957CD83076AE@inf.ed.ac.uk> References: <20100722022208.o4fvhtirhr6okkgk@msi.umrk.nl> <4C4898A8.1070709@your-file-system.com> <20110723050104.GB12228@excalibur.hozed.org> <59582E0B-9562-4AE3-9D31-957CD83076AE@inf.ed.ac.uk> Message-ID: Is the vlserver DB format documented anywhere? On Mon, Jul 25, 2011 at 1:28 PM, Simon Wilkinson wrote: > > I saw a mention of IPv6 support sometime in 2011 in my old email.. > > It's a work in progress. > There are actually multiple components to achieving IPv6 support in AFS-3= . > Different amounts of progress have been made on each of these components.= .. > 1/ Refactor the OpenAFS RX API so that IPv6 addresses can be used as > endpoints. There are two sub tasks here. Firstly we need additional > functions that can support opening connections to =C2=A0v6 endpoints, and > internal changes to support accepting connections over IPv6. Secondly, we > want to rework the API so that the internals of RX structures such as > connections and calls aren't exposed to library users. This work is also = a > prerequisite for rx/tcp support, and as such is on my list. > 2/ Create a generic XDR representation for an endpoint. The work in > afs3-stds to produce an extensible union actually grew from this - once t= his > work is complete (it pretty much is) we can move forwards to create an > address type > 3/ Create new versions of all AFS-3 RPCs which take IPv4 addresses, and h= ave > these new versions use the extensible address type. The plan is to do thi= s > work as part of RPC refresh. Chaz Chandler has already made great progres= s > on the new definitions - they're in a comparable form in gerrit 4573. The= se > new RPCs will require standardisation. > 4/ Extend the vlserver database so that it can store IPv6 addresses. This= is > potentially the trickiest part of the puzzle, as we would like to do this > and maintain a clear mechanism for backing out server upgrades. Ubik > database formats are private to OpenAFS, so this doesn't require > standardisation work. Some scoping work has been done here, but no > implementation of =C2=A0which I am aware. > 5/ Rework all of OpenAFS's internal data structures and APIs to store > addresses in a new extensible format, rather than just using int32s. > There's enough work here that could be easily accomplished in parallel if > anyone is interested in contributing to it. > Cheers, > Simon. > --=20 Tom Maher From adeason@sinenomine.net Mon Aug 1 23:21:20 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 1 Aug 2011 17:21:20 -0500 Subject: [OpenAFS] Re: Plans for IPv6 References: <20100722022208.o4fvhtirhr6okkgk@msi.umrk.nl> <4C4898A8.1070709@your-file-system.com> <20110723050104.GB12228@excalibur.hozed.org> <59582E0B-9562-4AE3-9D31-957CD83076AE@inf.ed.ac.uk> Message-ID: <20110801172120.3ebde35e.adeason@sinenomine.net> On Mon, 1 Aug 2011 14:05:33 -0700 Tom Maher wrote: > Is the vlserver DB format documented anywhere? I don't think so. At least, I don't think the current form of storing fileserver addresses is documented in any clear fashion; the old formats might be, but they're not as useful. If anyone wants to look at it, some basics: in the data for each vldb entry, the "server" is represented by a "server number", which is an index into a global table called 'hostaddress' in the in-memory representation, and 'IpMappedAddr' in the structures for the on-disk representation. That global table is just an array of 4-byte integers, and historically that was the IPv4 address for the server. There are only 256 entries in the table, which is the cause for the per-cell fileserver limit in AFS. A modern extension to the database is to allow fileservers to have more than one IP in the database. So, instead of hostaddress[X] being an IP address, if the first byte of hostaddress[X] if 0xff, it's treated as a special multihomed entry. The second byte is the 'base' of the entry (indicating with MH block we look at), and the last 2 bytes are the 'index' (indicating which entry in the MH block to look at). An "MH block" is just a structure used to allocate several addresses at once, so we can allocate in chunks, but we don't need to allocate all possible MH structures all the time. The first MH block is pointed to by the SIT member in the vldb header structure ('struct vlheader'), iirc. Anyway, the 'struct extentaddr' structure is the current modern structure used for storing addresses, which is in src/vlserver/vlserver.p.h. Looking at the FindExtentBlock function in vlutils.c I think helps understand the MH structures a bit, though that doesn't really deal with the on-disk layout. For that, the readExtents function may help, since it basically converts the on-disk MH format to the in-memory ones. Each MH block has a 'flags' field, so this may involve just defining a new 'type' of MH block (aside from the 'header' and 'continuation' ones right now), which may be able to be of any size and layout that we want (except for the 'flags' field which needs to be in the same position). Or it may be more complicated than that; I don't think too much thought has been put into the details of this yet, since, while it is important, it is one of the later steps of getting IPv6 to be useful, since the lower levels like Rx need to work first. -- Andrew Deason adeason@sinenomine.net From shadow@gmail.com Tue Aug 2 15:09:41 2011 From: shadow@gmail.com (Derrick Brashear) Date: Tue, 2 Aug 2011 10:09:41 -0400 Subject: [OpenAFS] issue with OSx Lion In-Reply-To: <03F599DE-9A60-4186-B433-6D83C0362628@gmail.com> References: <03F599DE-9A60-4186-B433-6D83C0362628@gmail.com> Message-ID: #define KRB5_PROG_ETYPE_NOSUPP (-1765328234L) set the allow_weak_crypto =3D yes option in /Library/Preferences/edu.mit.Kerberos On Wed, Jul 20, 2011 at 4:03 PM, H. Farahani wr= ote: > Hi, > > =A0I installed OpenAFS 1.6.0pre7 for MacOS 10.7 (Lion). When I try "aklog= " I received this error message: > > "aklog: Couldn't get pdc.kth.se AFS tickets: > aklog: unknown RPC error (-1765328234) while getting AFS tickets" > > =A0I appreciate your help. > > Best regards, > Shahrab > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > --=20 Derrick From sxw@inf.ed.ac.uk Tue Aug 2 15:36:13 2011 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Tue, 2 Aug 2011 15:36:13 +0100 Subject: [OpenAFS] issue with OSx Lion In-Reply-To: References: <03F599DE-9A60-4186-B433-6D83C0362628@gmail.com> Message-ID: On 2 Aug 2011, at 15:09, Derrick Brashear wrote: > #define KRB5_PROG_ETYPE_NOSUPP (-1765328234L) >=20 > set the > allow_weak_crypto =3D yes > option in /Library/Preferences/edu.mit.Kerberos Firstly, I thought that the 1.6 aklog converted both Kerberos and AFS errors= into text. Secondly, aklog should be telling Kerberos to allow the use of s= ingle DES enctypes, without requiring allow_weak_crypto. Is this all broken b= y Lion, or do you have an old aklog binary in your path? Cheers, Simon >=20 From shadow@gmail.com Tue Aug 2 15:49:48 2011 From: shadow@gmail.com (Derrick Brashear) Date: Tue, 2 Aug 2011 10:49:48 -0400 Subject: [OpenAFS] issue with OSx Lion In-Reply-To: References: <03F599DE-9A60-4186-B433-6D83C0362628@gmail.com> Message-ID: On Tue, Aug 2, 2011 at 10:36 AM, Simon Wilkinson wrote: > On 2 Aug 2011, at 15:09, Derrick Brashear wrote: > >> #define KRB5_PROG_ETYPE_NOSUPP =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 (-176= 5328234L) >> >> set the >> allow_weak_crypto =3D yes >> option in /Library/Preferences/edu.mit.Kerberos > > Firstly, I thought that the 1.6 aklog converted both Kerberos and AFS err= ors into text. Secondly, aklog should be telling Kerberos to allow the use = of single DES enctypes, without requiring allow_weak_crypto. Is this all br= oken by Lion, or do you have an old aklog binary in your path? > That would presume the APIs needed to do both those things worked. [scully:openafs/src/aklog] shadow% nm /System//Library/Frameworks/Kerberos.framework/Kerberos|egrep -i 'krb5_enctype_enable|krb5_allow_weak_crypto' [scully:openafs/src/aklog] shadow% So, you can't have that. The error_table call should be supported but I haven't looked yet to see why it fails. It's not a dummy call in MITKerberosShim. --=20 Derrick From ktdreyer@ktdreyer.com Fri Aug 5 19:08:46 2011 From: ktdreyer@ktdreyer.com (Ken Dreyer) Date: Fri, 5 Aug 2011 12:08:46 -0600 Subject: [OpenAFS] -backuptree and defaults Message-ID: I'm curious why -backuptree / FollowBackupPath is not enabled by default for clients? To me it makes more sense to have backup volume traversals follow the same paradigm as RW traversals. - Ken From ktdreyer@ktdreyer.com Fri Aug 5 19:18:26 2011 From: ktdreyer@ktdreyer.com (Ken Dreyer) Date: Fri, 5 Aug 2011 12:18:26 -0600 Subject: [OpenAFS] Re: -backuptree and defaults In-Reply-To: References: Message-ID: On Fri, Aug 5, 2011 at 12:08 PM, Ken Dreyer wrote: > I'm curious why -backuptree / FollowBackupPath is not enabled by > default for clients? To me it makes more sense to have backup volume > traversals follow the same paradigm as RW traversals. Looks like this was discussed about four years ago: https://lists.openafs.org/pipermail/openafs-info/2007-December/028090.html Could this be changed before 1.6 is finalized? - Ken From jaltman@secure-endpoints.com Fri Aug 5 19:43:55 2011 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Fri, 05 Aug 2011 14:43:55 -0400 Subject: [OpenAFS] Re: -backuptree and defaults In-Reply-To: References: Message-ID: <4E3C39EB.3040606@secure-endpoints.com> This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig7D5CA68274006392B143F127 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On 8/5/2011 2:18 PM, Ken Dreyer wrote: > On Fri, Aug 5, 2011 at 12:08 PM, Ken Dreyer wro= te: >> I'm curious why -backuptree / FollowBackupPath is not enabled by >> default for clients? To me it makes more sense to have backup volume >> traversals follow the same paradigm as RW traversals. >=20 > Looks like this was discussed about four years ago: > https://lists.openafs.org/pipermail/openafs-info/2007-December/028090.h= tml >=20 > Could this be changed before 1.6 is finalized? In my opinion, its too late for 1.6 since pre-release candidates have already shipped. The time to have made the change would have been last Summer before the 1.6 branch was created. In any case, nothing has changed with regards to the need for a consistent .backup policy across all volumes. As indicated in that thread, individual sites that know they have consistent backup policies can distribute their own packing or configuration to clients that sets the defaults they want. At the moment there is no ability for a "cell" to notify the client via a policy what the follow backup rule should be. The correct place may be a per volume policy bit. Jeffrey Altman --------------enig7D5CA68274006392B143F127 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (MingW32) iQEcBAEBAgAGBQJOPDnxAAoJENxm1CNJffh4nNwIAKWW6ZG+yXBjqqdY4jddwNwg TSGxkSy/bZl91pbApwt2llrqcZSO5Q8KiWwC9jlUJfKdVDzyOFWOPy30wJeL5UqQ qsH+j0vfIZQ40DX2H4bFXD0wrytLc8H2KZry2AEGCVmjEyseuWCV34Re9CIOTAgL ial2pnSIUAlN8428RnVVWf1Ak2cBZok0emkT3ce6hT+4voVutwo6htrhhvcnNwci Xy+ZkU0NmKEfigYdzHXtqrlREp1HH4sbVXKpJHW15ZZSqcLRwLkSaDY4XhFj6B4m 7OFkEDyGIFnQJ4FHW23Kh5nb3YKnpW3ETAeDOEuKVLdLCTXzjJjodMOofjF9G9I= =CvgV -----END PGP SIGNATURE----- --------------enig7D5CA68274006392B143F127-- From kwebb@teradactyl.com Mon Aug 8 17:15:24 2011 From: kwebb@teradactyl.com (Kristen J. Webb) Date: Mon, 08 Aug 2011 10:15:24 -0600 Subject: [OpenAFS] European AFS & Kerberos Conference 2011? Message-ID: <4E400B9C.7010905@teradactyl.com> Does anyone know the details for this event? Is there a web site? We are interested in attending, but having trouble finding any updated information on the conference for this year. Many thanks in advance, Kris -- Mr. Kristen J. Webb Teradactyl LLC. PHONE: 1-505-242-1091 EMAIL: kwebb@teradactyl.com VISIT: http://www.teradactyl.com Home of the True incremental Backup System From kallbac@indiana.edu Mon Aug 8 18:56:58 2011 From: kallbac@indiana.edu (Kallback-Rose, Kristy A) Date: Mon, 8 Aug 2011 13:56:58 -0400 Subject: [OpenAFS] OpenAFS 1.6.0pre7 and Mac OS 10.7 (Lion) cannot aklog Message-ID: <941254EC-1056-4D5F-AD68-CB80C2E7F990@indiana.edu> Hello, I am trying to get OpenAFS client working under Lion. I have tried the = tricks listed in other posts such as allow_weak_crypto =3D true and = directly specifying encryption types in the Kerberos config file. = However, I still cannot get AFS to startup or aklog to work. Kinit works = fine. I should note that a co-worker upgraded to Lion using this version + the = weak crypto option and all worked fine. I am trying this on 2 test = MacBooks with the same results --so I don't think it's on the OpenAFS = server side. Let me know what other details I should provide that might be useful in = troubleshooting this. When I try to start the client I see in the console: 8/4/11 3:36:37.178 PM System Preferences: Starting up afs 8/4/11 3:36:41.284 PM authexec: executing = /Library/OpenAFS/Tools/root.client/usr/vice/etc/afs.rc 8/4/11 3:36:41.359 PM com.apple.kextd: Failed to load = /var/db/openafs/etc/afs.kext - (libkern/kext) kext (kmod) start/stop = routine failed. I feel confident that the allow_weak_crypto =3D true is recognized = because when I aklog without that in the file I get: ~ kallbac$ aklog -d Authenticating to cell iu.edu (server rfs1). Trying to authenticate to user's realm ADS.IU.EDU. Getting tickets: afs/iu.edu@ADS.IU.EDU Kerberos error code returned by get_cred : -1765328234 aklog: Couldn't get iu.edu AFS tickets: aklog: unknown RPC error (-1765328234) while getting AFS tickets --where -1765328234 is -1765328234 KRB5_PROG_ETYPE_NOSUPP Program = lacks support for encryption type while with allow_weak_crypto I get (the rather strange looking) ~ kallbac$ aklog -d Authenticating to cell iu.edu (server rfs1). Trying to authenticate to user's realm ADS.IU.EDU. Getting tickets: afs/iu.edu@ADS.IU.EDU Using Kerberos V5 ticket natively About to resolve name ??d?. ??d? to id in cell iu.edu. Id 32766 Set username to ??d?. ??d? Setting tokens. ??d?. ??d? @ iu.edu aklog: a pioctl failed while obtaining tokens for cell iu.edu By running kextutil -v 6 /var/db/openafs/etc/afs.kext as root I get Kext user-space log filter changed from 0xff2 to 0xfff. Kext kernel-space log filter changed from 0xff2 to 0xfff. Kext library architecture set to i386. Kext library recording diagnostics for: validation authentication = dependencies warnings. /var/db/openafs/etc/afs.kext appears to be loadable (not including = linkage for on-disk libraries). Loading /var/db/openafs/etc/afs.kext. Reading load info for 4 kexts. Reading loaded kext info from kernel. (kernel) User-space log flags changed from 0x0 to 0xfff. (kernel) Received 'Get Loaded Kext Info' request from user space. (kernel) Returning loaded kext info. Loaded kext info: { "com.apple.kpi.mach" =3D { "OSBundleWiredSize" =3D 2032 "CFBundleIdentifier" =3D "com.apple.kpi.mach" "OSBundleUUID" =3D "OSBundleLoadAddress" =3D 9646080 "OSBundleLoadTag" =3D 5 "OSBundleStarted" =3D true "OSBundleCompatibleVersion" =3D "8.0.0d0" "OSKernelResource" =3D true "CFBundleVersion" =3D "11.0.0" "OSBundlePath" =3D "//System/Library/Extensions/System.kext/PlugIns/Mach.kext" "OSBundleRetainCount" =3D 86 "OSBundlePrelinked" =3D true "OSBundleIsInterface" =3D true "OSBundleLoadSize" =3D 2032 } "com.apple.kpi.bsd" =3D { "OSBundleWiredSize" =3D 23880 "CFBundleIdentifier" =3D "com.apple.kpi.bsd" "OSBundleUUID" =3D <2c050b25 bab14483 92f4bf45 e21665c5> "OSBundleLoadAddress" =3D 9719808 "OSBundleLoadTag" =3D 1 "OSBundleStarted" =3D true "OSBundleCompatibleVersion" =3D "8.0.0b1" "OSKernelResource" =3D true "CFBundleVersion" =3D "11.0.0" "OSBundlePath" =3D "//System/Library/Extensions/System.kext/PlugIns/BSDKernel.kext" "OSBundleRetainCount" =3D 69 "OSBundlePrelinked" =3D true "OSBundleIsInterface" =3D true "OSBundleLoadSize" =3D 23880 } "com.apple.kpi.libkern" =3D { "OSBundleWiredSize" =3D 36448 "CFBundleIdentifier" =3D "com.apple.kpi.libkern" "OSBundleUUID" =3D "OSBundleLoadAddress" =3D 9682944 "OSBundleLoadTag" =3D 4 "OSBundleStarted" =3D true "OSBundleCompatibleVersion" =3D "8.0.0d0" "OSKernelResource" =3D true "CFBundleVersion" =3D "11.0.0" "OSBundlePath" =3D "//System/Library/Extensions/System.kext/PlugIns/Libkern.kext" "OSBundleRetainCount" =3D 95 "OSBundlePrelinked" =3D true "OSBundleIsInterface" =3D true "OSBundleLoadSize" =3D 36448 } } /System/Library/Extensions/System.kext/PlugIns/Mach.kext (version 11.0, = UUID BAC1677F-9CAD-49E7-ABEF-4D6E6D42C02E) is loaded. /System/Library/Extensions/System.kext/PlugIns/Mach.kext (version 11.0): = is started. setting load address of = /System/Library/Extensions/System.kext/PlugIns/Mach.kext to 0x933000 /System/Library/Extensions/System.kext/PlugIns/BSDKernel.kext (version = 11.0, UUID 2C050B25-BAB1-4483-92F4-BF45E21665C5) is loaded. /System/Library/Extensions/System.kext/PlugIns/BSDKernel.kext (version = 11.0): is started. setting load address of = /System/Library/Extensions/System.kext/PlugIns/BSDKernel.kext to = 0x945000 /System/Library/Extensions/System.kext/PlugIns/Libkern.kext (version = 11.0, UUID A9EE657F-1F7B-45E9-8A18-53A0C2D00BAD) is loaded. /System/Library/Extensions/System.kext/PlugIns/Libkern.kext (version = 11.0): is started. setting load address of = /System/Library/Extensions/System.kext/PlugIns/Libkern.kext to 0x93c000 Omitting loaded kext = /System/Library/Extensions/System.kext/PlugIns/Mach.kext from mkext for = kernel load. Omitting loaded kext = /System/Library/Extensions/System.kext/PlugIns/BSDKernel.kext from mkext = for kernel load. Omitting loaded kext = /System/Library/Extensions/System.kext/PlugIns/Libkern.kext from mkext = for kernel load. Adding /var/db/openafs/etc/afs.kext to mkext. /var/db/openafs/etc/afs.kext added 1054644-byte noncompressed executable = to mkext. Created mkext for architecture i386 containing 1 kexts. Loading /var/db/openafs/etc/afs.kext. (kernel) User-space log flags changed from 0x0 to 0xfff. (kernel) Received kext load request from user space. (kernel) Received request from user space to load kext = org.openafs.filesystems.afs. (kernel) Loading kext org.openafs.filesystems.afs. (kernel) Kext com.apple.kpi.mach is already loaded. (kernel) Kext com.apple.kpi.bsd is already loaded. (kernel) Kext com.apple.kpi.libkern is already loaded. (kernel) Allocated link buffer for kext org.openafs.filesystems.afs at = 0x1562000 (851968 bytes). (kernel) Kext org.openafs.filesystems.afs executable loaded; 208 pages = at 0x1562000 (load tag 117). (kernel) Kext org.openafs.filesystems.afs calling module start function. (kernel) Kext org.openafs.filesystems.afs did not start (return code = 0x5). (kernel) Kext org.openafs.filesystems.afs start failed (result 0x5). (kernel) Kext org.openafs.filesystems.afs failed to load (0xdc008017). (kernel) Failed to load kext org.openafs.filesystems.afs (error = 0xdc008017). (kernel) Kext org.openafs.filesystems.afs unloading. (kernel) Kext org.openafs.filesystems.afs unwiring and unmapping linked = executable. (kernel) Kext org.openafs.filesystems.afs unloaded. (kernel) Kext org.openafs.filesystems.afs removing all personalities = naming it from the IOCatalogue. Kernel error handling kext request - (libkern/kext) kext (kmod) = start/stop routine failed. Failed to load /var/db/openafs/etc/afs.kext - (libkern/kext) kext (kmod) = start/stop routine failed. Failed to load /var/db/openafs/etc/afs.kext - (libkern/kext) kext (kmod) = start/stop routine failed. From shadow@gmail.com Mon Aug 8 19:03:18 2011 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 8 Aug 2011 14:03:18 -0400 Subject: [OpenAFS] OpenAFS 1.6.0pre7 and Mac OS 10.7 (Lion) cannot aklog In-Reply-To: <941254EC-1056-4D5F-AD68-CB80C2E7F990@indiana.edu> References: <941254EC-1056-4D5F-AD68-CB80C2E7F990@indiana.edu> Message-ID: On Mon, Aug 8, 2011 at 1:56 PM, Kallback-Rose, Kristy A wrote: > Hello, > > I am trying to get OpenAFS client working under Lion. I have tried the tr= icks listed in other posts such as allow_weak_crypto =3D true and directly = specifying encryption types in the Kerberos config file. However, I still c= annot get AFS to startup or aklog to work. Kinit works fine. > > I should note that a co-worker upgraded to Lion using this version + the = weak crypto option and all worked fine. I am trying this on 2 test MacBooks= with the same results --so I don't think it's on the OpenAFS server side. > > Let me know what other details I should provide that might be useful in t= roubleshooting this. > > When I try to start the client I see in the console: > > 8/4/11 3:36:37.178 PM System Preferences: Starting up afs > 8/4/11 3:36:41.284 PM authexec: executing /Library/OpenAFS/Tools/root.cli= ent/usr/vice/etc/afs.rc > 8/4/11 3:36:41.359 PM com.apple.kextd: Failed to load /var/db/openafs/etc= /afs.kext - (libkern/kext) kext (kmod) start/stop routine failed. you're already lost once you see this. > > I feel confident that the allow_weak_crypto =3D true is recognized becaus= e when I aklog without that in the file I get: > ~ kallbac$ aklog -d > Authenticating to cell iu.edu (server rfs1). > Trying to authenticate to user's realm ADS.IU.EDU. > Getting tickets: afs/iu.edu@ADS.IU.EDU > Kerberos error code returned by get_cred : -1765328234 > aklog: Couldn't get iu.edu AFS tickets: > aklog: unknown RPC error (-1765328234) while getting AFS tickets > > --where =A0-1765328234 is -1765328234 =A0 =A0 KRB5_PROG_ETYPE_NOSUPP =A0P= rogram lacks support for encryption type > > while with allow_weak_crypto =A0I get (the rather strange looking) > ~ kallbac$ aklog -d > Authenticating to cell iu.edu (server rfs1). > Trying to authenticate to user's realm ADS.IU.EDU. > Getting tickets: afs/iu.edu@ADS.IU.EDU > Using Kerberos V5 ticket natively > About to resolve name =A0??d?. ??d? to id in cell iu.edu. > Id 32766 > Set username to =A0??d?. ??d? > Setting tokens. =A0??d?. ??d? @ iu.edu > aklog: a pioctl failed while obtaining tokens for cell iu.edu afs isn't running, so, aklog can't insert tokens > By running =A0kextutil -v 6 /var/db/openafs/etc/afs.kext as root I get > > Kext user-space log filter changed from 0xff2 to 0xfff. > Kext kernel-space log filter changed from 0xff2 to 0xfff. > Kext library architecture set to i386. > Kext library recording diagnostics for: validation authentication depende= ncies warnings. > /var/db/openafs/etc/afs.kext appears to be loadable (not including linkag= e for on-disk libraries). > Loading /var/db/openafs/etc/afs.kext. > Reading load info for 4 kexts. > Reading loaded kext info from kernel. > (kernel) User-space log flags changed from 0x0 to 0xfff. > (kernel) Received 'Get Loaded Kext Info' request from user space. > (kernel) Returning loaded kext info. > Loaded kext info: > { > "com.apple.kpi.mach" =3D { > =A0"OSBundleWiredSize" =3D 2032 > =A0"CFBundleIdentifier" =3D "com.apple.kpi.mach" > =A0"OSBundleUUID" =3D > =A0"OSBundleLoadAddress" =3D 9646080 > =A0"OSBundleLoadTag" =3D 5 > =A0"OSBundleStarted" =3D true > =A0"OSBundleCompatibleVersion" =3D "8.0.0d0" > =A0"OSKernelResource" =3D true > =A0"CFBundleVersion" =3D "11.0.0" > =A0"OSBundlePath" =3D > =A0 =A0 =A0"//System/Library/Extensions/System.kext/PlugIns/Mach.kext" > =A0"OSBundleRetainCount" =3D 86 > =A0"OSBundlePrelinked" =3D true > =A0"OSBundleIsInterface" =3D true > =A0"OSBundleLoadSize" =3D 2032 > } > "com.apple.kpi.bsd" =3D { > =A0"OSBundleWiredSize" =3D 23880 > =A0"CFBundleIdentifier" =3D "com.apple.kpi.bsd" > =A0"OSBundleUUID" =3D <2c050b25 bab14483 92f4bf45 e21665c5> > =A0"OSBundleLoadAddress" =3D 9719808 > =A0"OSBundleLoadTag" =3D 1 > =A0"OSBundleStarted" =3D true > =A0"OSBundleCompatibleVersion" =3D "8.0.0b1" > =A0"OSKernelResource" =3D true > =A0"CFBundleVersion" =3D "11.0.0" > =A0"OSBundlePath" =3D > =A0 =A0 =A0"//System/Library/Extensions/System.kext/PlugIns/BSDKernel.kex= t" > =A0"OSBundleRetainCount" =3D 69 > =A0"OSBundlePrelinked" =3D true > =A0"OSBundleIsInterface" =3D true > =A0"OSBundleLoadSize" =3D 23880 > } > "com.apple.kpi.libkern" =3D { > =A0"OSBundleWiredSize" =3D 36448 > =A0"CFBundleIdentifier" =3D "com.apple.kpi.libkern" > =A0"OSBundleUUID" =3D > =A0"OSBundleLoadAddress" =3D 9682944 > =A0"OSBundleLoadTag" =3D 4 > =A0"OSBundleStarted" =3D true > =A0"OSBundleCompatibleVersion" =3D "8.0.0d0" > =A0"OSKernelResource" =3D true > =A0"CFBundleVersion" =3D "11.0.0" > =A0"OSBundlePath" =3D > =A0 =A0 =A0"//System/Library/Extensions/System.kext/PlugIns/Libkern.kext" > =A0"OSBundleRetainCount" =3D 95 > =A0"OSBundlePrelinked" =3D true > =A0"OSBundleIsInterface" =3D true > =A0"OSBundleLoadSize" =3D 36448 > } > } > > /System/Library/Extensions/System.kext/PlugIns/Mach.kext (version 11.0, U= UID BAC1677F-9CAD-49E7-ABEF-4D6E6D42C02E) is loaded. > /System/Library/Extensions/System.kext/PlugIns/Mach.kext (version 11.0): = is started. > setting load address of /System/Library/Extensions/System.kext/PlugIns/Ma= ch.kext to 0x933000 > /System/Library/Extensions/System.kext/PlugIns/BSDKernel.kext (version 11= .0, UUID 2C050B25-BAB1-4483-92F4-BF45E21665C5) is loaded. > /System/Library/Extensions/System.kext/PlugIns/BSDKernel.kext (version 11= .0): is started. > setting load address of /System/Library/Extensions/System.kext/PlugIns/BS= DKernel.kext to 0x945000 > /System/Library/Extensions/System.kext/PlugIns/Libkern.kext (version 11.0= , UUID A9EE657F-1F7B-45E9-8A18-53A0C2D00BAD) is loaded. > /System/Library/Extensions/System.kext/PlugIns/Libkern.kext (version 11.0= ): is started. > setting load address of /System/Library/Extensions/System.kext/PlugIns/Li= bkern.kext to 0x93c000 > Omitting loaded kext /System/Library/Extensions/System.kext/PlugIns/Mach.= kext from mkext for kernel load. > Omitting loaded kext /System/Library/Extensions/System.kext/PlugIns/BSDKe= rnel.kext from mkext for kernel load. > Omitting loaded kext /System/Library/Extensions/System.kext/PlugIns/Libke= rn.kext from mkext for kernel load. > Adding /var/db/openafs/etc/afs.kext to mkext. > /var/db/openafs/etc/afs.kext added 1054644-byte noncompressed executable = to mkext. > Created mkext for architecture i386 containing 1 kexts. i386 tells us 32 bit, which is not the default. > Loading /var/db/openafs/etc/afs.kext. > (kernel) User-space log flags changed from 0x0 to 0xfff. > (kernel) Received kext load request from user space. > (kernel) Received request from user space to load kext org.openafs.filesy= stems.afs. > (kernel) Loading kext org.openafs.filesystems.afs. > (kernel) Kext com.apple.kpi.mach is already loaded. > (kernel) Kext com.apple.kpi.bsd is already loaded. > (kernel) Kext com.apple.kpi.libkern is already loaded. > (kernel) Allocated link buffer for kext org.openafs.filesystems.afs at 0x= 1562000 (851968 bytes). > (kernel) Kext org.openafs.filesystems.afs executable loaded; 208 pages at= 0x1562000 (load tag 117). > (kernel) Kext org.openafs.filesystems.afs calling module start function. > (kernel) Kext org.openafs.filesystems.afs did not start (return code 0x5)= . > (kernel) Kext org.openafs.filesystems.afs start failed (result 0x5). > (kernel) Kext org.openafs.filesystems.afs failed to load (0xdc008017). > (kernel) Failed to load kext org.openafs.filesystems.afs (error 0xdc00801= 7). which sadly isn't that useful, but is the error. is there a reason you're booting 32 bit? --=20 Derrick From adeason@sinenomine.net Mon Aug 8 19:46:59 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 8 Aug 2011 13:46:59 -0500 Subject: [OpenAFS] Re: Unable to delete volume References: <20110731195436.b34202c5cc7bffe6fc954c40@sinenomine.net> <20110801091831.3539844a.adeason@sinenomine.net> <20110801102101.0d8b7fa2.adeason@sinenomine.net> Message-ID: <20110808134659.66d1c76f.adeason@sinenomine.net> On Mon, 1 Aug 2011 11:33:49 -0400 Aaron Knister wrote: > sigh...the vldb clearly is unhappy. I'll send you a copy of the > vldb.DB0 To follow up on this a bit: the issue here was a little different than the direction Derrick and I were going in, as the vldb was actually corrupt (as in, not in a way you can get by just issuing "vos" commands, I think). We perhaps should log this case in the vlserver, as it represents database corruption and it's hard to tell that that is the case from the "vos" error messages (we just see "volume does not exist"). This is an example of what can "break" due to errors reported by vldb_check. I thought people might want to see that, since we talked about it at the workshop. In this case, running vldb_check shows (among a bunch of other stuff): VLDB_CHECK_WARNING: VLDB entry 'g.ilin22.local' has no RW volume 23193: Volume 'g.ilin22.local' not found in name hash 2214 (next 20655 next in chain) 23193: Record 3432572 (type 0x7071) not in a name chain That first one isn't necessarily corruption, but just a strange condition (you _can_ have RO volumes without RWs, but most of the time that's not what you want). "Record 3432572" is one of the g.ilin22.local entries. In this particular case there are two different vl entries for the name g.ilin22.local, but one of them is not anywhere in the name hash table. The vlserver refuses to delete an entry that does not exist on the name hash table, so the entry is undeletable. This can only be fixed by 'vldb_check -fix', or by altering the database by hand or something. It doesn't really have much to do with the fact that the volume is specified on a server that doesn't exist, or that there's another vl entry with the same name, though of course that made things more confusing. -- Andrew Deason adeason@sinenomine.net From kallbac@iu.edu Mon Aug 8 20:13:19 2011 From: kallbac@iu.edu (Kristy Kallback-Rose) Date: Mon, 8 Aug 2011 15:13:19 -0400 Subject: [OpenAFS] OpenAFS 1.6.0pre7 and Mac OS 10.7 (Lion) cannot aklog In-Reply-To: References: <941254EC-1056-4D5F-AD68-CB80C2E7F990@indiana.edu> Message-ID: <3CB5919B-C4A5-4617-9277-FC20575776D6@iu.edu> >>=20 >=20 > i386 tells us 32 bit, which is not the default. >=20 >> Loading /var/db/openafs/etc/afs.kext. >> (kernel) User-space log flags changed from 0x0 to 0xfff. >> (kernel) Received kext load request from user space. >> (kernel) Received request from user space to load kext = org.openafs.filesystems.afs. >> (kernel) Loading kext org.openafs.filesystems.afs. >> (kernel) Kext com.apple.kpi.mach is already loaded. >> (kernel) Kext com.apple.kpi.bsd is already loaded. >> (kernel) Kext com.apple.kpi.libkern is already loaded. >> (kernel) Allocated link buffer for kext org.openafs.filesystems.afs = at 0x1562000 (851968 bytes). >> (kernel) Kext org.openafs.filesystems.afs executable loaded; 208 = pages at 0x1562000 (load tag 117). >> (kernel) Kext org.openafs.filesystems.afs calling module start = function. >> (kernel) Kext org.openafs.filesystems.afs did not start (return code = 0x5). >> (kernel) Kext org.openafs.filesystems.afs start failed (result 0x5). >> (kernel) Kext org.openafs.filesystems.afs failed to load = (0xdc008017). >> (kernel) Failed to load kext org.openafs.filesystems.afs (error = 0xdc008017). >=20 > which sadly isn't that useful, but is the error. >=20 > is there a reason you're booting 32 bit? >=20 >=20 >=20 Thanks for your quick reply. I didn't realize it was going into 32-bit = mode by default, but now under=20 About This Mac --> More Info.. -> System Report --> Software --> 64-bit = Kernel and Extension=20 I see "No". In theory this Macbook can be made to run in 64-bit mode. Let me work on = that and report back. Cheers, Kristy= From kallbac@indiana.edu Wed Aug 10 21:56:17 2011 From: kallbac@indiana.edu (Kristy Kallback-Rose) Date: Wed, 10 Aug 2011 16:56:17 -0400 Subject: [OpenAFS] OpenAFS 1.6.0pre7 and Mac OS 10.7 (Lion) cannot aklog In-Reply-To: <0B2F97A1-724C-4DB6-8716-E821A27BC1ED@iu.edu> References: <941254EC-1056-4D5F-AD68-CB80C2E7F990@indiana.edu> <0B2F97A1-724C-4DB6-8716-E821A27BC1ED@iu.edu> Message-ID: Well, it appears I have one of the Macs which is blacklisted from = booting into 64-bit mode --though I can run Lion. If you want to know = more about what I'm referring to you can see the workaround that people = did for Snow Leopard and the same issue here http://netkas.org/?p=3D189 = Though I have not seem similar fixes for Lion, nor do I want to support = users doing that sort of fix. Regarding getting OpenAFS working on this type of system, do you have a = recommendation for next steps? Should I build openafs from source on = this machine? We will no doubt have users running into this problem as = the semester begins. Thanks, Kristy On Aug 8, 2011, at 3:08 PM, Kristy Kallback-Rose wrote: >>>=20 >>=20 >> i386 tells us 32 bit, which is not the default. >>=20 >>> Loading /var/db/openafs/etc/afs.kext. >>> (kernel) User-space log flags changed from 0x0 to 0xfff. >>> (kernel) Received kext load request from user space. >>> (kernel) Received request from user space to load kext = org.openafs.filesystems.afs. >>> (kernel) Loading kext org.openafs.filesystems.afs. >>> (kernel) Kext com.apple.kpi.mach is already loaded. >>> (kernel) Kext com.apple.kpi.bsd is already loaded. >>> (kernel) Kext com.apple.kpi.libkern is already loaded. >>> (kernel) Allocated link buffer for kext org.openafs.filesystems.afs = at 0x1562000 (851968 bytes). >>> (kernel) Kext org.openafs.filesystems.afs executable loaded; 208 = pages at 0x1562000 (load tag 117). >>> (kernel) Kext org.openafs.filesystems.afs calling module start = function. >>> (kernel) Kext org.openafs.filesystems.afs did not start (return code = 0x5). >>> (kernel) Kext org.openafs.filesystems.afs start failed (result 0x5). >>> (kernel) Kext org.openafs.filesystems.afs failed to load = (0xdc008017). >>> (kernel) Failed to load kext org.openafs.filesystems.afs (error = 0xdc008017). >>=20 >> which sadly isn't that useful, but is the error. >>=20 >> is there a reason you're booting 32 bit? >>=20 >>=20 >>=20 >=20 >=20 > Thanks for your quick reply. I didn't realize it was going into 32-bit = mode by default, but now under=20 > About This Mac --> More Info.. -> System Report --> Software --> = 64-bit Kernel and Extension=20 >=20 > I see "No". >=20 > In theory this Macbook can be made to run in 64-bit mode. Let me work = on that and report back. >=20 > Cheers, > Kristy Kristy Kallback-Rose Research Storage kallbac@iu.edu From shadow@gmail.com Thu Aug 11 05:31:13 2011 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 11 Aug 2011 00:31:13 -0400 Subject: [OpenAFS] OpenAFS 1.6.0pre7 and Mac OS 10.7 (Lion) cannot aklog In-Reply-To: References: <941254EC-1056-4D5F-AD68-CB80C2E7F990@indiana.edu> <0B2F97A1-724C-4DB6-8716-E821A27BC1ED@iu.edu> Message-ID: On Wed, Aug 10, 2011 at 4:56 PM, Kristy Kallback-Rose wrote: > Well, it appears I have one of the Macs which is blacklisted from booting= into 64-bit mode --though I can run Lion. If you want to know more about w= hat I'm referring to you can see the workaround that people did for Snow Le= opard and the same issue here http://netkas.org/?p=3D189 Though I have not = seem similar fixes for Lion, nor do I want to support users doing that sort= of fix. > > Regarding getting OpenAFS working on this type of system, do you have a r= ecommendation for next steps? Should I build openafs from source on this ma= chine? We will no doubt have users running into this problem as the semeste= r begins. given that the existing build is a universal build, i suspect something more subtle is in play. you can try building for source but i suspect you will have the same error. --=20 Derrick From aaronk@umbc.edu Thu Aug 11 15:11:27 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Thu, 11 Aug 2011 10:11:27 -0400 Subject: [OpenAFS] Re: kernel panic in 1.4.14.1 on rhel6 In-Reply-To: References: <4E304FF4.8070403@sinenomine.net> <20110727140512.72b69950.adeason@sinenomine.net> <20110727153702.e85ebd45.adeason@sinenomine.net> Message-ID: --485b39618be4b8929504aa3b600d Content-Type: text/plain; charset=ISO-8859-1 I know everyone's busy, but out of curiosity has anybody had a chance to look at this? Could it be that it's not even an AFS bug, but a linux kernel issue? I'm getting ready to put a webserver into production and am concerned because I can reliably trigger the mentioned kernel panic using "ab". Thanks, -Aaron On Wed, Jul 27, 2011 at 4:59 PM, Aaron Knister wrote: > On Wed, Jul 27, 2011 at 4:37 PM, Andrew Deason wrote: > >> On Wed, 27 Jul 2011 15:25:15 -0400 >> Aaron Knister wrote: >> >> > Hi Andrew, >> > >> > Here's the output of the last few, what I think are relavent lines: >> >> Well, this is basically the same information as you originally gave. >> There's nothing else around these, or immediately before? >> > > The lines preceding what I originally sent are: > > openafs: module license 'http://www.openafs.org/dl/license10.html' taints > kernel. > Disabling lock debugging due to kernel taint > enabling dynamically allocated vcaches > Starting AFS cache scan...found 993 non-empty cache files (31%). > eth0: no IPv6 routers present > general protection fault: 0000 [#1] SMP > > I don't know if that's any more helpful. > > >> >> Also, could you run in crash >> >> dis rxi_NewCall >> dis rxi_ResetCall >> dis mutex_lock >> >> > Here's a link to a directory containing the output of those commands: > http://userpages.umbc.edu/~aaronk/afs/linux-kernel-panic/. > > Thanks, > -Aaron > > >> and put the result somewhere? (It's probably a bit large for the list) >> >> -- >> Andrew Deason >> adeason@sinenomine.net >> >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info >> > > > > -- > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --485b39618be4b8929504aa3b600d Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I know everyone's busy, but out of curiosity has anybody had a chance t= o look at this? Could it be that it's not even an AFS bug, but a linux = kernel issue? I'm getting ready to put a webserver into production and = am concerned because I can reliably trigger the mentioned kernel panic usin= g "ab".

Thanks,
-Aaron

On Wed,= Jul 27, 2011 at 4:59 PM, Aaron Knister <aaronk@umbc.edu> wrote:
On Wed, Jul 27, 2011 at 4:37 PM, Andrew Deason <ad= eason@sinenomine.net> wrote:
On Wed, 27 Jul 2011 15:25:15 -0400
Aaron Knister <aaro= nk@umbc.edu> wrote:

> Hi Andrew,
>
> Here's the output of the last few, what I think are relavent lines= :

Well, this is basically the same information as you originally gave.<= br> There's nothing else around these, or immediately before?

The lines=A0preceding=A0what I originally sent= are:

openafs: module license 'http://www.ope= nafs.org/dl/license10.html' taints kernel.
Disabling lock debugging due to kernel taint
enabling dynami= cally allocated vcaches
Starting AFS cache scan...found 993 non-e= mpty cache files (31%).
eth0: no IPv6 routers present
general protection fault: 0000 [#1] SMP=A0

I= don't know if that's any more helpful.
=A0

Also, could you run in crash

dis rxi_NewCall
dis rxi_ResetCall
dis mutex_lock


Here's a link to a directory= containing the output of those commands:=A0http://userpages.u= mbc.edu/~aaronk/afs/linux-kernel-panic/.

Thanks,
-Aaron
= =A0
and put the result somewhere? (It's probably a bit large for the list)<= br>
--
Andrew Deason
adeason@sinenom= ine.net

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


=

--
Aaron Knister
Systems Administrator
Divi= sion of Information Technology
University of Maryland, Baltimore County<= br> aaronk@umbc.edu



--
Aaron Knist= er
Systems Administrator
Division of Information Technology
Univer= sity of Maryland, Baltimore County
aaronk@umbc.edu
--485b39618be4b8929504aa3b600d-- From shadow@gmail.com Thu Aug 11 15:37:19 2011 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 11 Aug 2011 10:37:19 -0400 Subject: [OpenAFS] OpenAFS 1.6.0pre7 and Mac OS 10.7 (Lion) cannot aklog In-Reply-To: References: <941254EC-1056-4D5F-AD68-CB80C2E7F990@indiana.edu> <0B2F97A1-724C-4DB6-8716-E821A27BC1ED@iu.edu> Message-ID: The fix will be in 1.6.0 final. In the meantime, you can have something for now from /afs/your-file-system.com/user/shadow/OpenAFS-1.6.0pre7-31-g24471-Lion.dmg On Thu, Aug 11, 2011 at 12:31 AM, Derrick Brashear wrote= : > On Wed, Aug 10, 2011 at 4:56 PM, Kristy Kallback-Rose > wrote: >> Well, it appears I have one of the Macs which is blacklisted from bootin= g into 64-bit mode --though I can run Lion. If you want to know more about = what I'm referring to you can see the workaround that people did for Snow L= eopard and the same issue here http://netkas.org/?p=3D189 Though I have not= seem similar fixes for Lion, nor do I want to support users doing that sor= t of fix. >> >> Regarding getting OpenAFS working on this type of system, do you have a = recommendation for next steps? Should I build openafs from source on this m= achine? We will no doubt have users running into this problem as the semest= er begins. > > given that the existing build is a universal build, i suspect > something more subtle is in play. you can try building for source but > i suspect you will have the same error. > > > > > > -- > Derrick > --=20 Derrick From jblaine@kickflop.net Thu Aug 11 15:53:02 2011 From: jblaine@kickflop.net (Jeff Blaine) Date: Thu, 11 Aug 2011 10:53:02 -0400 Subject: [OpenAFS] ETA for 1.4.15? Message-ID: <4E43ECCE.3010402@kickflop.net> Do we have an ETA for 1.4.15 by any chance? Last I heard it was March/April 2011. Looking to have an official/bundled fix for the Solaris 10 hang at shutdown thing. Anything I can do to help the cause? From shadow@gmail.com Thu Aug 11 16:12:32 2011 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 11 Aug 2011 11:12:32 -0400 Subject: [OpenAFS] ETA for 1.4.15? In-Reply-To: <4E43ECCE.3010402@kickflop.net> References: <4E43ECCE.3010402@kickflop.net> Message-ID: send a time machine? a security fix will be included. final testing is happening. On Thu, Aug 11, 2011 at 10:53 AM, Jeff Blaine wrote: > Do we have an ETA for 1.4.15 by any chance? =A0Last > I heard it was March/April 2011. =A0Looking to have > an official/bundled fix for the Solaris 10 hang > at shutdown thing. > > Anything I can do to help the cause? > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > --=20 Derrick From pontius@btv.ibm.com Thu Aug 11 17:44:43 2011 From: pontius@btv.ibm.com (Dale Pontius) Date: Thu, 11 Aug 2011 12:44:43 -0400 Subject: [OpenAFS] ETA for 1.4.15? In-Reply-To: References: <4E43ECCE.3010402@kickflop.net> Message-ID: <4E4406FB.3050907@btv.ibm.com> On 08/11/2011 11:12 AM, Derrick Brashear wrote: > send a time machine? > > a security fix will be included. final testing is happening. > > On Thu, Aug 11, 2011 at 10:53 AM, Jeff Blaine wrote: >> Do we have an ETA for 1.4.15 by any chance? Last >> I heard it was March/April 2011. Looking to have >> an official/bundled fix for the Solaris 10 hang >> at shutdown thing. >> >> Anything I can do to help the cause? >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info >> I was under the impression that 1.4.14.1 was it, and that 1.6.0 would be the next release out, which begs the next question. Will 1.4.15 support the linux-3.0.x kernel series? It was actually linux-2.6.39 that drove me off of 1.4.14.1. -- Dale Pontius Senior Engineer IBM Corporation Phone: (802) 769-6850 Tie-Line: 446-6850 email: pontius@us.ibm.com This e-mail and its attachments, if any, may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply e-mail and delete all copies of this message from your system without copying it and notify sender of the misdirection by reply e-mail. From sxw@inf.ed.ac.uk Thu Aug 11 17:58:25 2011 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Thu, 11 Aug 2011 17:58:25 +0100 Subject: [OpenAFS] ETA for 1.4.15? In-Reply-To: <4E4406FB.3050907@btv.ibm.com> References: <4E43ECCE.3010402@kickflop.net> <4E4406FB.3050907@btv.ibm.com> Message-ID: <83CAEE32-C426-4AE1-AD80-699DC486B63E@inf.ed.ac.uk> On 11 Aug 2011, at 17:44, Dale Pontius wrote: > I was under the impression that 1.4.14.1 was it, and that 1.6.0 would = be the next release out, which begs the next question. >=20 > Will 1.4.15 support the linux-3.0.x kernel series? It was actually = linux-2.6.39 that drove me off of 1.4.14.1. So, I think the current plan is that we will do 1.4.15 as a security = release (this will be 1.4.14.1 with only a couple of security fixes = applied), and 1.4.16 (which has all of the bug fixes that have been = collecting in the 1.4 branch since 1.4.14). After that, we'll see. If the uptake of 1.6 is rapid enough, then 1.4.16 = may be the last 1.4 series release. If there are still a large number of = users on 1.4.x, and sufficiently serious bugs appear, then there may be = further 1.4.x series releases. None of the recent Linux fixes (including the change for Linux 3.0 = support) have been backported to the 1.4 branch yet. This is primarily = because we've all be focussing on getting 1.6 out of the door. If those = backports are simple then I'd imagine that they'll appear in 1.4.16, = otherwise it's going to depend on whether there is sufficient demand for = that work to be done. It's worth bearing in mind that anyone can submit backports to gerrit. = Use git cherry-pick -x to pull the change that you are interested in = onto a checkout of the openafs-stable-1_4_x branch. Test that it works = as expected, then edit the change's commit message to remove the = existing Commit-Id line. Finally, push the change to the = refs/for/openafs-stable-1_4_x branch of gerrit. Cheers, Simon From aaronk@umbc.edu Thu Aug 11 18:13:57 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Thu, 11 Aug 2011 13:13:57 -0400 Subject: [OpenAFS] Re: kernel panic in 1.4.14.1 on rhel6 In-Reply-To: References: <4E304FF4.8070403@sinenomine.net> <20110727140512.72b69950.adeason@sinenomine.net> <20110727153702.e85ebd45.adeason@sinenomine.net> Message-ID: --0022152d632945898004aa3ded80 Content-Type: text/plain; charset=ISO-8859-1 For what it's worth, the panic doesn't occur in 1.6.0pre7 even with repeated hammering with ab. On Thu, Aug 11, 2011 at 10:11 AM, Aaron Knister wrote: > I know everyone's busy, but out of curiosity has anybody had a chance to > look at this? Could it be that it's not even an AFS bug, but a linux kernel > issue? I'm getting ready to put a webserver into production and am concerned > because I can reliably trigger the mentioned kernel panic using "ab". > > Thanks, > -Aaron > > > On Wed, Jul 27, 2011 at 4:59 PM, Aaron Knister wrote: > >> On Wed, Jul 27, 2011 at 4:37 PM, Andrew Deason wrote: >> >>> On Wed, 27 Jul 2011 15:25:15 -0400 >>> Aaron Knister wrote: >>> >>> > Hi Andrew, >>> > >>> > Here's the output of the last few, what I think are relavent lines: >>> >>> Well, this is basically the same information as you originally gave. >>> There's nothing else around these, or immediately before? >>> >> >> The lines preceding what I originally sent are: >> >> openafs: module license 'http://www.openafs.org/dl/license10.html' taints >> kernel. >> Disabling lock debugging due to kernel taint >> enabling dynamically allocated vcaches >> Starting AFS cache scan...found 993 non-empty cache files (31%). >> eth0: no IPv6 routers present >> general protection fault: 0000 [#1] SMP >> >> I don't know if that's any more helpful. >> >> >>> >>> Also, could you run in crash >>> >>> dis rxi_NewCall >>> dis rxi_ResetCall >>> dis mutex_lock >>> >>> >> Here's a link to a directory containing the output of those commands: >> http://userpages.umbc.edu/~aaronk/afs/linux-kernel-panic/. >> >> Thanks, >> -Aaron >> >> >>> and put the result somewhere? (It's probably a bit large for the list) >>> >>> -- >>> Andrew Deason >>> adeason@sinenomine.net >>> >>> _______________________________________________ >>> OpenAFS-info mailing list >>> OpenAFS-info@openafs.org >>> https://lists.openafs.org/mailman/listinfo/openafs-info >>> >> >> >> >> -- >> Aaron Knister >> Systems Administrator >> Division of Information Technology >> University of Maryland, Baltimore County >> aaronk@umbc.edu >> > > > > -- > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --0022152d632945898004aa3ded80 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable For what it's worth, the panic doesn't occur in 1.6.0pre7 even with= repeated hammering with ab.

On Thu, Aug = 11, 2011 at 10:11 AM, Aaron Knister <aaronk@umbc.edu> wrote:
I know everyone's busy, but out of curio= sity has anybody had a chance to look at this? Could it be that it's no= t even an AFS bug, but a linux kernel issue? I'm getting ready to put a= webserver into production and am concerned because I can reliably trigger = the mentioned kernel panic using "ab".

Thanks,
-Aaron


On Wed, Jul 27, 2011 at 4:59 PM, Aaron Knister <aaronk@umbc.ed= u> wrote:
On Wed, Jul 27, 2011 at 4:37 PM, Andrew Deason <<= a href=3D"mailto:adeason@sinenomine.net" target=3D"_blank">adeason@sinenomi= ne.net> wrote:
On Wed, 27 Jul 2011 15:25:15 -0400
Aaron Knister <aaro= nk@umbc.edu> wrote:

> Hi Andrew,
>
> Here's the output of the last few, what I think are relavent lines= :

Well, this is basically the same information as you originally gave.<= br> There's nothing else around these, or immediately before?

The lines=A0preceding=A0what I originally sent= are:

openafs: module license 'http://www.ope= nafs.org/dl/license10.html' taints kernel.
Disabling lock debugging due to kernel taint
enabling dynami= cally allocated vcaches
Starting AFS cache scan...found 993 non-e= mpty cache files (31%).
eth0: no IPv6 routers present
general protection fault: 0000 [#1] SMP=A0

I= don't know if that's any more helpful.
=A0

Also, could you run in crash

dis rxi_NewCall
dis rxi_ResetCall
dis mutex_lock


Here's a link to a directory= containing the output of those commands:=A0http://userpages.u= mbc.edu/~aaronk/afs/linux-kernel-panic/.

Thanks,
-Aaron
=A0
and put the result somewhere? (It's probably a bit large for the list)<= br>
--
Andrew Deason
adeason@sinenom= ine.net

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


=

--
Aaron Knister
Systems Administrator
Division of Infor= mation Technology
University of Maryland, Baltimore County
aaronk@umbc.edu



--
Aaron Knist= er
Systems Administrator
Division of Information Technology
Univer= sity of Maryland, Baltimore County
aaronk@umbc.edu



--
Aaron Knister
System= s Administrator
Division of Information Technology
University of Mary= land, Baltimore County
aaronk@umbc.edu
--0022152d632945898004aa3ded80-- From adeason@sinenomine.net Thu Aug 11 18:56:36 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Thu, 11 Aug 2011 12:56:36 -0500 Subject: [OpenAFS] Re: ETA for 1.4.15? References: <4E43ECCE.3010402@kickflop.net> <4E4406FB.3050907@btv.ibm.com> <83CAEE32-C426-4AE1-AD80-699DC486B63E@inf.ed.ac.uk> Message-ID: <20110811125636.2ce4a6c3.adeason@sinenomine.net> On Thu, 11 Aug 2011 17:58:25 +0100 Simon Wilkinson wrote: > and 1.4.16 (which has all of the bug fixes that have been collecting > in the 1.4 branch since 1.4.14). I think that's "since 1.4.12". -- Andrew Deason adeason@sinenomine.net From dorian.taylor.lists@gmail.com Thu Aug 11 20:23:35 2011 From: dorian.taylor.lists@gmail.com (dorian taylor) Date: Thu, 11 Aug 2011 12:23:35 -0700 Subject: [OpenAFS] endianness problem somewhere in pts lookup, 1.5.77 Message-ID: Hi there, I think I found a byte-sex problem somewhere in the system of aklog and DNS SRV records. My Mac (10.5.8 running OpenAFS 1.5.77) kept taking inordinately long to run aklog, so I decided to investigate. Upon retrieving the SRV record for the prserver, aklog makes multiple attempts to query the server on port 23067 rather than 7002. Of course: $ perl -le 'printf "%x %x\n", 7002, 23067' 1b5a 5a1b I'm not sure if this was fixed in future releases, just letting you know. Cheers, -- Dorian Taylor http://doriantaylor.com/ From adeason@sinenomine.net Fri Aug 12 17:01:03 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Fri, 12 Aug 2011 11:01:03 -0500 Subject: [OpenAFS] Re: kernel panic in 1.4.14.1 on rhel6 References: <4E304FF4.8070403@sinenomine.net> <20110727140512.72b69950.adeason@sinenomine.net> <20110727153702.e85ebd45.adeason@sinenomine.net> Message-ID: <20110812110103.54e316dd.adeason@sinenomine.net> On Thu, 11 Aug 2011 10:11:27 -0400 Aaron Knister wrote: > I know everyone's busy, but out of curiosity has anybody had a chance > to look at this? Could it be that it's not even an AFS bug, but a > linux kernel issue? I'm getting ready to put a webserver into > production and am concerned because I can reliably trigger the > mentioned kernel panic using "ab". I don't think it's likely that it's a linux kernel issue. If you want something to try, you can try applying these two patches: http://git.openafs.org/?p=openafs.git;a=commitdiff_plain;h=910f3d4cbba4cd78a8e6afa0b32e37ed7f292d18 http://git.openafs.org/?p=openafs.git;a=commitdiff_plain;h=35835af9bb85b358e4bdd11113b637ac0e52c56b Those fix what may be a similar issue, but I'm not really sure if it's the same thing as here. If that doesn't do it (it's good that you have a reliable way of reproducing this), we'll need to look more closely at the core (if you don't want to provide it since it probably contains file data, instructions can probably be provided, though it's a lot more detailed than what we've done so far). Or we can proably go the route of giving you debug patches, but I'm not quite sure what to add quite yet. -- Andrew Deason adeason@sinenomine.net From ktdreyer@ktdreyer.com Sat Aug 13 00:13:30 2011 From: ktdreyer@ktdreyer.com (Ken Dreyer) Date: Fri, 12 Aug 2011 17:13:30 -0600 Subject: [OpenAFS] Re: kernel panic in 1.4.14.1 on rhel6 In-Reply-To: References: <4E304FF4.8070403@sinenomine.net> <20110727140512.72b69950.adeason@sinenomine.net> <20110727153702.e85ebd45.adeason@sinenomine.net> Message-ID: On Thu, Aug 11, 2011 at 8:11 AM, Aaron Knister wrote: > I know everyone's busy, but out of curiosity has anybody had a chance to > look at this? Could it be that it's not even an AFS bug, but a linux kernel > issue? I'm getting ready to put a webserver into production and am concerned > because I can reliably trigger the mentioned kernel panic using "ab". For the record, what's the full ab command that you're using? - Ken From dcolson5@uncc.edu Wed Aug 10 20:16:39 2011 From: dcolson5@uncc.edu (Colson, Daniel) Date: Wed, 10 Aug 2011 15:16:39 -0400 Subject: [OpenAFS] 64-bit mod_waklog / filedrawers on afs Message-ID: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> This is a multi-part message in MIME format. ------_=_NextPart_001_01CC5792.07AC056F Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Is anyone else running mod_waklog / filedrawers on afs on a 64-bit = machine? I've been unable to compile mod_waklog (an essential component = of filedrawers) on 64-bit (RHEL5.6). ------_=_NextPart_001_01CC5792.07AC056F Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable 64-bit mod_waklog / filedrawers on afs

Is anyone else running mod_waklog / filedrawers on afs = on a 64-bit machine? I've been unable to compile mod_waklog (an = essential component of filedrawers) on 64-bit (RHEL5.6).

------_=_NextPart_001_01CC5792.07AC056F-- From aaronk@umbc.edu Mon Aug 15 17:32:48 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Mon, 15 Aug 2011 12:32:48 -0400 Subject: [OpenAFS] 64-bit mod_waklog / filedrawers on afs In-Reply-To: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> References: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> Message-ID: --0022152d63297594e404aa8dd174 Content-Type: text/plain; charset=ISO-8859-1 Hi Daniel, While I'm not running filedrawers I am using mod_waklog, albeit with RHEL6. I had to apply some patches to get it to build correctly. If you send your build errors I can see if my patches will help. -Aaron On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel wrote: > ** > > Is anyone else running mod_waklog / filedrawers on afs on a 64-bit machine? > I've been unable to compile mod_waklog (an essential component of > filedrawers) on 64-bit (RHEL5.6). > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --0022152d63297594e404aa8dd174 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi Daniel,

While I'm not running filedrawers I am us= ing mod_waklog, albeit with RHEL6. I had to apply some patches to get it to= build correctly. If you send your build errors I can see if my patches wil= l help.

-Aaron

On Wed, Aug 10= , 2011 at 3:16 PM, Colson, Daniel <dcolson5@uncc.edu> wrote:

Is anyone else running mod_waklog / filedrawers on afs = on a 64-bit machine? I've been unable to compile mod_waklog (an essenti= al component of filedrawers) on 64-bit (RHEL5.6).




--
Aaron Kniste= r
Systems Administrator
Division of Information Technology
Univers= ity of Maryland, Baltimore County
aaronk@umbc.edu
--0022152d63297594e404aa8dd174-- From botsch@cnf.cornell.edu Mon Aug 15 17:37:09 2011 From: botsch@cnf.cornell.edu (Dave Botsch) Date: Mon, 15 Aug 2011 12:37:09 -0400 Subject: [OpenAFS] 64-bit mod_waklog / filedrawers on afs In-Reply-To: References: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> Message-ID: <20110815163708.GE11153@cnf.cornell.edu> Can you post your patches someplace? I do plan on compiling under 64-bit RHEL6... I'm currently running under 32-bit RHEL4. thanks. On Mon, Aug 15, 2011 at 12:32:48PM -0400, Aaron Knister wrote: > Hi Daniel, > > While I'm not running filedrawers I am using mod_waklog, albeit with RHEL6. > I had to apply some patches to get it to build correctly. If you send your > build errors I can see if my patches will help. > > -Aaron > > On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel wrote: > > > ** > > > > Is anyone else running mod_waklog / filedrawers on afs on a 64-bit machine? > > I've been unable to compile mod_waklog (an essential component of > > filedrawers) on 64-bit (RHEL5.6). > > > > > > -- > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu -- ******************************** David William Botsch Programmer/Analyst CNF Computing botsch@cnf.cornell.edu ******************************** From aaronk@umbc.edu Mon Aug 15 17:47:26 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Mon, 15 Aug 2011 12:47:26 -0400 Subject: [OpenAFS] 64-bit mod_waklog / filedrawers on afs In-Reply-To: <20110815163708.GE11153@cnf.cornell.edu> References: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> <20110815163708.GE11153@cnf.cornell.edu> Message-ID: --0022152d6329cba38304aa8e05b8 Content-Type: text/plain; charset=ISO-8859-1 I feel your pain. We were running some 32 bit RHEL3 up until 8 months ago. Here are the patches: http://userpages.umbc.edu/~aaronk/waklog/ Let me know if they don't work for you -Aaron On Mon, Aug 15, 2011 at 12:37 PM, Dave Botsch wrote: > Can you post your patches someplace? I do plan on compiling under 64-bit > RHEL6... I'm currently running under 32-bit RHEL4. > > thanks. > > On Mon, Aug 15, 2011 at 12:32:48PM -0400, Aaron Knister wrote: > > Hi Daniel, > > > > While I'm not running filedrawers I am using mod_waklog, albeit with > RHEL6. > > I had to apply some patches to get it to build correctly. If you send > your > > build errors I can see if my patches will help. > > > > -Aaron > > > > On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel > wrote: > > > > > ** > > > > > > Is anyone else running mod_waklog / filedrawers on afs on a 64-bit > machine? > > > I've been unable to compile mod_waklog (an essential component of > > > filedrawers) on 64-bit (RHEL5.6). > > > > > > > > > > > -- > > Aaron Knister > > Systems Administrator > > Division of Information Technology > > University of Maryland, Baltimore County > > aaronk@umbc.edu > > -- > ******************************** > David William Botsch > Programmer/Analyst > CNF Computing > botsch@cnf.cornell.edu > ******************************** > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --0022152d6329cba38304aa8e05b8 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I fe= el your pain. We were running some 32 bit RHEL3 up until 8 months ago.

Here are the patches:

Let me know if they don't work for you

-= Aaron

On Mon, Aug 15, 2011 at 12:37 PM, Dave Botsch <botsch@cnf.corne= ll.edu> wrote:
Can you post your patches someplace? I do p= lan on compiling under 64-bit
RHEL6... I'm currently running under 32-bit RHEL4.

thanks.

On Mon, Aug 15, 2011 at 12:32:48PM -0400, Aaron Knister wrote:
> Hi Daniel,
>
> While I'm not running filedrawers I am using mod_waklog, albeit wi= th RHEL6.
> I had to apply some patches to get it to build correctly. If you send = your
> build errors I can see if my patches will help.
>
> -Aaron
>
> On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel <dcolson5@uncc.edu> wrote:
>
> > **
> >
> > Is anyone else running mod_waklog / filedrawers on afs on a 64-bi= t machine?
> > I've been unable to compile mod_waklog (an essential componen= t of
> > filedrawers) on 64-bit (RHEL5.6).
> >
>
>
>
> --
> Aaron Knister
> Systems Administrator
> Division of Information Technology
> University of Maryland, Baltimore County
> aaronk@umbc.edu

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



--
Aaron Kniste= r
Systems Administrator
Division of Information Technology
Univers= ity of Maryland, Baltimore County
aaronk@umbc.edu
--0022152d6329cba38304aa8e05b8-- From jblaine@kickflop.net Mon Aug 15 20:29:35 2011 From: jblaine@kickflop.net (Jeff Blaine) Date: Mon, 15 Aug 2011 15:29:35 -0400 Subject: [OpenAFS] screen loses tokens - Solaris 10 Message-ID: <4E49739F.1080102@kickflop.net> How might I go about debugging this? This happens on a host with Generic_142900-03 but not on a host with Generic_144488-17 (nor ever on this latter host at any patch rev -- I have been using/resuming screen on it for years). 1. Connect to host with PuTTY 2. Confirm krb5 creds and tokens gotten from PAM 3. Start screen 4. Confirm krb5 creds and tokens in screen shell 5. Close PuTTY, "Yes, disconnect" 6. Connect to host with PuTTY 7. Confirm krb5 creds and tokens gotten from PAM 8. Resume screen session 9. Tokens and krb5 creds in screen shell are gone Common ------ OpenAFS 1.4.14 MIT Kerberos 1.6.3 Screen 4.00.02 sshd_config pam.conf pam_afs_session pam_krb5RA (Russ Alberry's) No kdestroy in shell dot files Different --------- SunOS faron.our.org 5.10 Generic_142900-03 sun4u sparc SUNW,Sun-Fire-V490 SunOS cairo.our.org 5.10 Generic_144488-17 sun4u sparc SUNW,Sun-Fire-280R From rra@stanford.edu Mon Aug 15 20:34:19 2011 From: rra@stanford.edu (Russ Allbery) Date: Mon, 15 Aug 2011 12:34:19 -0700 Subject: [OpenAFS] screen loses tokens - Solaris 10 In-Reply-To: <4E49739F.1080102@kickflop.net> (Jeff Blaine's message of "Mon, 15 Aug 2011 15:29:35 -0400") References: <4E49739F.1080102@kickflop.net> Message-ID: <87sjp2a2pw.fsf@windlord.stanford.edu> Jeff Blaine writes: > How might I go about debugging this? This happens on a host with > Generic_142900-03 but not on a host with Generic_144488-17 (nor ever on > this latter host at any patch rev -- I have been using/resuming screen > on it for years). > 1. Connect to host with PuTTY > 2. Confirm krb5 creds and tokens gotten from PAM > 3. Start screen > 4. Confirm krb5 creds and tokens in screen shell > 5. Close PuTTY, "Yes, disconnect" > 6. Connect to host with PuTTY > 7. Confirm krb5 creds and tokens gotten from PAM > 8. Resume screen session > 9. Tokens and krb5 creds in screen shell are gone When you log out of the session from which you started screen, PAM will destroy your AFS tokens. If you don't want PAM to destroy AFS tokens on session close, you need to give the retain_after_close option to pam-afs-session and pam-krb5. Alternately, start screen with krenew -t -- screen, which will create isolated tickets and tokens for the screen process that are disconnected from your login session. -- Russ Allbery (rra@stanford.edu) From kevin@umd.edu Mon Aug 15 20:34:47 2011 From: kevin@umd.edu (Kevin Hildebrand) Date: Mon, 15 Aug 2011 15:34:47 -0400 (EDT) Subject: [OpenAFS] screen loses tokens - Solaris 10 In-Reply-To: <4E49739F.1080102@kickflop.net> References: <4E49739F.1080102@kickflop.net> Message-ID: We had problems with tokens disappearing until I added: pam = { retain_after_close = true } pam-afs-session = { retain_after_close = true } to /etc/krb5.conf. Kevin On Mon, 15 Aug 2011, Jeff Blaine wrote: > How might I go about debugging this? This happens > on a host with Generic_142900-03 but not on a host > with Generic_144488-17 (nor ever on this latter host > at any patch rev -- I have been using/resuming screen > on it for years). > > 1. Connect to host with PuTTY > 2. Confirm krb5 creds and tokens gotten from PAM > 3. Start screen > 4. Confirm krb5 creds and tokens in screen shell > 5. Close PuTTY, "Yes, disconnect" > 6. Connect to host with PuTTY > 7. Confirm krb5 creds and tokens gotten from PAM > 8. Resume screen session > 9. Tokens and krb5 creds in screen shell are gone > > Common > ------ > OpenAFS 1.4.14 > MIT Kerberos 1.6.3 > Screen 4.00.02 > sshd_config > pam.conf > pam_afs_session > pam_krb5RA (Russ Alberry's) > No kdestroy in shell dot files > > Different > --------- > SunOS faron.our.org 5.10 Generic_142900-03 sun4u sparc SUNW,Sun-Fire-V490 > > SunOS cairo.our.org 5.10 Generic_144488-17 sun4u sparc SUNW,Sun-Fire-280R > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > From kevin@umd.edu Mon Aug 15 20:38:07 2011 From: kevin@umd.edu (Kevin Hildebrand) Date: Mon, 15 Aug 2011 15:38:07 -0400 (EDT) Subject: [OpenAFS] screen loses tokens - Solaris 10 In-Reply-To: References: <4E49739F.1080102@kickflop.net> Message-ID: Sorry, should have mentioned, those are in the section of krb5.conf labeled: [appdefaults] Kevin On Mon, 15 Aug 2011, Kevin Hildebrand wrote: > > We had problems with tokens disappearing until I added: > > pam = { > retain_after_close = true > } > > pam-afs-session = { > retain_after_close = true > } > > to /etc/krb5.conf. > > Kevin > > > On Mon, 15 Aug 2011, Jeff Blaine wrote: > >> How might I go about debugging this? This happens >> on a host with Generic_142900-03 but not on a host >> with Generic_144488-17 (nor ever on this latter host >> at any patch rev -- I have been using/resuming screen >> on it for years). >> >> 1. Connect to host with PuTTY >> 2. Confirm krb5 creds and tokens gotten from PAM >> 3. Start screen >> 4. Confirm krb5 creds and tokens in screen shell >> 5. Close PuTTY, "Yes, disconnect" >> 6. Connect to host with PuTTY >> 7. Confirm krb5 creds and tokens gotten from PAM >> 8. Resume screen session >> 9. Tokens and krb5 creds in screen shell are gone >> >> Common >> ------ >> OpenAFS 1.4.14 >> MIT Kerberos 1.6.3 >> Screen 4.00.02 >> sshd_config >> pam.conf >> pam_afs_session >> pam_krb5RA (Russ Alberry's) >> No kdestroy in shell dot files >> >> Different >> --------- >> SunOS faron.our.org 5.10 Generic_142900-03 sun4u sparc SUNW,Sun-Fire-V490 >> >> SunOS cairo.our.org 5.10 Generic_144488-17 sun4u sparc SUNW,Sun-Fire-280R >> >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info >> > From aaronk@umbc.edu Mon Aug 15 20:56:19 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Mon, 15 Aug 2011 15:56:19 -0400 Subject: [OpenAFS] Re: kernel panic in 1.4.14.1 on rhel6 In-Reply-To: References: <4E304FF4.8070403@sinenomine.net> <20110727140512.72b69950.adeason@sinenomine.net> <20110727153702.e85ebd45.adeason@sinenomine.net> Message-ID: --0016e6d262324ba54504aa90a9b1 Content-Type: text/plain; charset=ISO-8859-1 ab -n 1000 -c 10 Also, I'm running httpd 2.2.15-9.el6 with the mpm-itk patch. On Fri, Aug 12, 2011 at 7:13 PM, Ken Dreyer wrote: > On Thu, Aug 11, 2011 at 8:11 AM, Aaron Knister wrote: > > I know everyone's busy, but out of curiosity has anybody had a chance to > > look at this? Could it be that it's not even an AFS bug, but a linux > kernel > > issue? I'm getting ready to put a webserver into production and am > concerned > > because I can reliably trigger the mentioned kernel panic using "ab". > > For the record, what's the full ab command that you're using? > > - Ken > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --0016e6d262324ba54504aa90a9b1 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable ab -n 1000 -c 10 <url>

Also, I'm running httpd= 2.2.15-9.el6 with the mpm-itk patch.

On = Fri, Aug 12, 2011 at 7:13 PM, Ken Dreyer <ktdreyer@ktdreyer.com> wrote:
On Thu, Aug 11, 2011 at 8= :11 AM, Aaron Knister <aaronk@umbc.ed= u> wrote:
> I know everyone's busy, but out of curiosity has anybody had a cha= nce to
> look at this? Could it be that it's not even an AFS bug, but a lin= ux kernel
> issue? I'm getting ready to put a webserver into production and am= concerned
> because I can reliably trigger the mentioned kernel panic using "= ab".

For the record, what's the full ab command that you're using?=

- Ken



--
Aaron= Knister
Systems Administrator
Division of Information Technology
= University of Maryland, Baltimore County
aaronk@umbc.edu
--0016e6d262324ba54504aa90a9b1-- From jblaine@kickflop.net Mon Aug 15 20:58:58 2011 From: jblaine@kickflop.net (Jeff Blaine) Date: Mon, 15 Aug 2011 15:58:58 -0400 Subject: [OpenAFS] screen loses tokens - Solaris 10 In-Reply-To: <87sjp2a2pw.fsf@windlord.stanford.edu> References: <4E49739F.1080102@kickflop.net> <87sjp2a2pw.fsf@windlord.stanford.edu> Message-ID: <4E497A82.4020608@kickflop.net> On 8/15/2011 3:34 PM, Russ Allbery wrote: > Jeff Blaine writes: > >> How might I go about debugging this? This happens on a host with >> Generic_142900-03 but not on a host with Generic_144488-17 (nor ever on >> this latter host at any patch rev -- I have been using/resuming screen >> on it for years). > >> 1. Connect to host with PuTTY >> 2. Confirm krb5 creds and tokens gotten from PAM >> 3. Start screen >> 4. Confirm krb5 creds and tokens in screen shell >> 5. Close PuTTY, "Yes, disconnect" >> 6. Connect to host with PuTTY >> 7. Confirm krb5 creds and tokens gotten from PAM >> 8. Resume screen session >> 9. Tokens and krb5 creds in screen shell are gone > > When you log out of the session from which you started screen, PAM will > destroy your AFS tokens. If you don't want PAM to destroy AFS tokens on > session close, you need to give the retain_after_close option to > pam-afs-session and pam-krb5. Thanks Russ (and Kevin!). Both hosts are using that option. Identical /etc/pam.conf and /etc/krb5.conf files on both the working and failing hosts. login session optional pam_krb5RA.so minimum_uid=92 retain_after_close I'll play around though. From aaronk@umbc.edu Mon Aug 15 21:26:53 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Mon, 15 Aug 2011 16:26:53 -0400 Subject: [OpenAFS] Re: kernel panic in 1.4.14.1 on rhel6 In-Reply-To: <20110812110103.54e316dd.adeason@sinenomine.net> References: <4E304FF4.8070403@sinenomine.net> <20110727140512.72b69950.adeason@sinenomine.net> <20110727153702.e85ebd45.adeason@sinenomine.net> <20110812110103.54e316dd.adeason@sinenomine.net> Message-ID: --20cf307810a89b832304aa9116f7 Content-Type: text/plain; charset=ISO-8859-1 On Fri, Aug 12, 2011 at 12:01 PM, Andrew Deason wrote: > On Thu, 11 Aug 2011 10:11:27 -0400 > Aaron Knister wrote: > > > I know everyone's busy, but out of curiosity has anybody had a chance > > to look at this? Could it be that it's not even an AFS bug, but a > > linux kernel issue? I'm getting ready to put a webserver into > > production and am concerned because I can reliably trigger the > > mentioned kernel panic using "ab". > > I don't think it's likely that it's a linux kernel issue. If you want > something to try, you can try applying these two patches: > > > http://git.openafs.org/?p=openafs.git;a=commitdiff_plain;h=910f3d4cbba4cd78a8e6afa0b32e37ed7f292d18 > > http://git.openafs.org/?p=openafs.git;a=commitdiff_plain;h=35835af9bb85b358e4bdd11113b637ac0e52c56b > > Those fix what may be a similar issue, but I'm not really sure if it's > the same thing as here. If that doesn't do it (it's good that you have a > reliable way of reproducing this), we'll need to look more closely at > the core (if you don't want to provide it since it probably contains > file data, instructions can probably be provided, though it's a lot more > detailed than what we've done so far). Or we can proably go the route of > giving you debug patches, but I'm not quite sure what to add quite yet. > > Hi Andrew, Those patches do prevent it from crashing, but also results in a number of "permission denied" errors from the apache processes under heavy load from ab. The permission denied errors don't occur with 1.6.0pre7. Thanks, Aaron > -- > Andrew Deason > adeason@sinenomine.net > > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --20cf307810a89b832304aa9116f7 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
On Fri, Aug 12, 2011 at 12:01 PM, Andrew Deason = <adeason@sin= enomine.net> wrote:
On Thu, 11 Aug 2011 10:11:27 -0400
Aaron Knister <aaro= nk@umbc.edu> wrote:

> I know everyone's busy, but out of curiosi= ty has anybody had a chance
> to look at this? Could it be that it's not even an AFS bug, but a<= br> > linux kernel issue? I'm getting ready to put a webserver into
> production and am concerned because I can reliably trigger the
> mentioned kernel panic using "ab".

I don't think it's likely that it's a linux kernel issue.= If you want
something to try, you can try applying these two patches:

http://git.o= penafs.org/?p=3Dopenafs.git;a=3Dcommitdiff_plain;h=3D910f3d4cbba4cd78a8e6af= a0b32e37ed7f292d18
http://git.o= penafs.org/?p=3Dopenafs.git;a=3Dcommitdiff_plain;h=3D35835af9bb85b358e4bdd1= 1113b637ac0e52c56b

Those fix what may be a similar issue, but I'm not really sure if it= 9;s
the same thing as here. If that doesn't do it (it's good that you h= ave a
reliable way of reproducing this), we'll need to look more closely at the core (if you don't want to provide it since it probably contains file data, instructions can probably be provided, though it's a lot mor= e
detailed than what we've done so far). Or we can proably go the route o= f
giving you debug patches, but I'm not quite sure what to add quite yet.=


Hi Andr= ew,=A0

Those patches do prevent it from crashing, = but also results in a number of "permission denied" errors from t= he apache processes under heavy load from ab. The permission denied errors = don't occur with 1.6.0pre7.

Thanks,
Aaron
=A0
--
Andrew Deason
adeason@sinenomine.net

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



--
= Aaron Knister
Systems Administrator
Division of Information Technolog= y
University of Maryland, Baltimore County
aaronk@umbc.edu
--20cf307810a89b832304aa9116f7-- From shadow@gmail.com Mon Aug 15 23:02:58 2011 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 15 Aug 2011 18:02:58 -0400 Subject: [OpenAFS] Re: kernel panic in 1.4.14.1 on rhel6 In-Reply-To: References: <4E304FF4.8070403@sinenomine.net> <20110727140512.72b69950.adeason@sinenomine.net> <20110727153702.e85ebd45.adeason@sinenomine.net> <20110812110103.54e316dd.adeason@sinenomine.net> Message-ID: On Mon, Aug 15, 2011 at 4:26 PM, Aaron Knister wrote: > On Fri, Aug 12, 2011 at 12:01 PM, Andrew Deason > wrote: >> >> On Thu, 11 Aug 2011 10:11:27 -0400 >> Aaron Knister wrote: >> >> > I know everyone's busy, but out of curiosity has anybody had a chance >> > to look at this? Could it be that it's not even an AFS bug, but a >> > linux kernel issue? I'm getting ready to put a webserver into >> > production and am concerned because I can reliably trigger the >> > mentioned kernel panic using "ab". >> >> I don't think it's likely that it's a linux kernel issue. If you want >> something to try, you can try applying these two patches: >> >> >> http://git.openafs.org/?p=openafs.git;a=commitdiff_plain;h=910f3d4cbba4cd78a8e6afa0b32e37ed7f292d18 >> >> http://git.openafs.org/?p=openafs.git;a=commitdiff_plain;h=35835af9bb85b358e4bdd11113b637ac0e52c56b >> >> Those fix what may be a similar issue, but I'm not really sure if it's >> the same thing as here. If that doesn't do it (it's good that you have a >> reliable way of reproducing this), we'll need to look more closely at >> the core (if you don't want to provide it since it probably contains >> file data, instructions can probably be provided, though it's a lot more >> detailed than what we've done so far). Or we can proably go the route of >> giving you debug patches, but I'm not quite sure what to add quite yet. >> > > Hi Andrew, > Those patches do prevent it from crashing, but also results in a number of > "permission denied" errors from the apache processes under heavy load from > ab. The permission denied errors don't occur with 1.6.0pre7. is there a reason 1.6.0pre7 isn't suitable, incidentally? -- Derrick From rra@stanford.edu Mon Aug 15 23:13:08 2011 From: rra@stanford.edu (Russ Allbery) Date: Mon, 15 Aug 2011 15:13:08 -0700 Subject: [OpenAFS] screen loses tokens - Solaris 10 In-Reply-To: <4E497A82.4020608@kickflop.net> (Jeff Blaine's message of "Mon, 15 Aug 2011 15:58:58 -0400") References: <4E49739F.1080102@kickflop.net> <87sjp2a2pw.fsf@windlord.stanford.edu> <4E497A82.4020608@kickflop.net> Message-ID: <87aaba9vd7.fsf@windlord.stanford.edu> Jeff Blaine writes: > Thanks Russ (and Kevin!). Both hosts are using that option. > Identical /etc/pam.conf and /etc/krb5.conf files on both > the working and failing hosts. > login session optional pam_krb5RA.so minimum_uid=92 retain_after_close > I'll play around though. You need it for pam_afs_session as well. Try running with debug set for both and make sure that syslog says that it's not deleting tickets and tokens during the logout. -- Russ Allbery (rra@stanford.edu) From jblaine@kickflop.net Mon Aug 15 23:28:52 2011 From: jblaine@kickflop.net (Jeff Blaine) Date: Mon, 15 Aug 2011 18:28:52 -0400 Subject: [OpenAFS] screen loses tokens - Solaris 10 In-Reply-To: <87aaba9vd7.fsf@windlord.stanford.edu> References: <4E49739F.1080102@kickflop.net> <87sjp2a2pw.fsf@windlord.stanford.edu> <4E497A82.4020608@kickflop.net> <87aaba9vd7.fsf@windlord.stanford.edu> Message-ID: <4E499DA4.2070109@kickflop.net> On 8/15/2011 6:13 PM, Russ Allbery wrote: > Jeff Blaine writes: > >> Thanks Russ (and Kevin!). Both hosts are using that option. > >> Identical /etc/pam.conf and /etc/krb5.conf files on both >> the working and failing hosts. > >> login session optional pam_krb5RA.so minimum_uid=92 retain_after_close > >> I'll play around though. > > You need it for pam_afs_session as well. Try running with debug set for > both and make sure that syslog says that it's not deleting tickets and > tokens during the logout. That solved it. Now I wish I could explain why it worked fine on the one box and not the other. Thanks. From aaronk@umbc.edu Mon Aug 15 23:55:48 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Mon, 15 Aug 2011 18:55:48 -0400 Subject: [OpenAFS] Re: kernel panic in 1.4.14.1 on rhel6 In-Reply-To: References: <4E304FF4.8070403@sinenomine.net> <20110727140512.72b69950.adeason@sinenomine.net> <20110727153702.e85ebd45.adeason@sinenomine.net> <20110812110103.54e316dd.adeason@sinenomine.net> Message-ID: --bcaec54ee9422cfd0a04aa932bd0 Content-Type: text/plain; charset=ISO-8859-1 On Mon, Aug 15, 2011 at 6:02 PM, Derrick Brashear wrote: > On Mon, Aug 15, 2011 at 4:26 PM, Aaron Knister wrote: > > On Fri, Aug 12, 2011 at 12:01 PM, Andrew Deason > > wrote: > >> > >> On Thu, 11 Aug 2011 10:11:27 -0400 > >> Aaron Knister wrote: > >> > >> > I know everyone's busy, but out of curiosity has anybody had a chance > >> > to look at this? Could it be that it's not even an AFS bug, but a > >> > linux kernel issue? I'm getting ready to put a webserver into > >> > production and am concerned because I can reliably trigger the > >> > mentioned kernel panic using "ab". > >> > >> I don't think it's likely that it's a linux kernel issue. If you want > >> something to try, you can try applying these two patches: > >> > >> > >> > http://git.openafs.org/?p=openafs.git;a=commitdiff_plain;h=910f3d4cbba4cd78a8e6afa0b32e37ed7f292d18 > >> > >> > http://git.openafs.org/?p=openafs.git;a=commitdiff_plain;h=35835af9bb85b358e4bdd11113b637ac0e52c56b > >> > >> Those fix what may be a similar issue, but I'm not really sure if it's > >> the same thing as here. If that doesn't do it (it's good that you have a > >> reliable way of reproducing this), we'll need to look more closely at > >> the core (if you don't want to provide it since it probably contains > >> file data, instructions can probably be provided, though it's a lot more > >> detailed than what we've done so far). Or we can proably go the route of > >> giving you debug patches, but I'm not quite sure what to add quite yet. > >> > > > > Hi Andrew, > > Those patches do prevent it from crashing, but also results in a number > of > > "permission denied" errors from the apache processes under heavy load > from > > ab. The permission denied errors don't occur with 1.6.0pre7. > > is there a reason 1.6.0pre7 isn't suitable, incidentally? > Not really, other than some of my colleagues had reservations about using a pre-release in production. > > -- > Derrick > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --bcaec54ee9422cfd0a04aa932bd0 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
On Mon, Aug 15, 2011 at 6:02 PM, Derrick Brashea= r <shadow@gmail.co= m> wrote:
On Mon, Aug 15, 2011 at 4:26 PM, Aaron Kn= ister <aaronk@umbc.edu> wrote:=
> On Fri, Aug 12, 2011 at 12:01 PM, Andrew Deason <adeason@sinenomine.net>
> wrote:
>>
>> On Thu, 11 Aug 2011 10:11:27 -0400
>> Aaron Knister <aaronk@umbc.e= du> wrote:
>>
>> > I know everyone's busy, but out of curiosity has anybody = had a chance
>> > to look at this? Could it be that it's not even an AFS bu= g, but a
>> > linux kernel issue? I'm getting ready to put a webserver = into
>> > production and am concerned because I can reliably trigger th= e
>> > mentioned kernel panic using "ab".
>>
>> I don't think it's likely that it's a linux kernel iss= ue. If you want
>> something to try, you can try applying these two patches:
>>
>>
>> http:= //git.openafs.org/?p=3Dopenafs.git;a=3Dcommitdiff_plain;h=3D910f3d4cbba4cd7= 8a8e6afa0b32e37ed7f292d18
>>
>> http:= //git.openafs.org/?p=3Dopenafs.git;a=3Dcommitdiff_plain;h=3D35835af9bb85b35= 8e4bdd11113b637ac0e52c56b
>>
>> Those fix what may be a similar issue, but I'm not really sure= if it's
>> the same thing as here. If that doesn't do it (it's good t= hat you have a
>> reliable way of reproducing this), we'll need to look more clo= sely at
>> the core (if you don't want to provide it since it probably co= ntains
>> file data, instructions can probably be provided, though it's = a lot more
>> detailed than what we've done so far). Or we can proably go th= e route of
>> giving you debug patches, but I'm not quite sure what to add q= uite yet.
>>
>
> Hi Andrew,
> Those patches do prevent it from crashing, but also results in a numbe= r of
> "permission denied" errors from the apache processes under h= eavy load from
> ab. The permission denied errors don't occur with 1.6.0pre7.

is there a reason 1.6.0pre7 isn't suitable, incidentally?

Not really, other than some of my collea= gues had reservations about using a pre-release in production.=A0
=A0

--
Derrick



--
Aaron= Knister
Systems Administrator
Division of Information Technology
= University of Maryland, Baltimore County
aaronk@umbc.edu
--bcaec54ee9422cfd0a04aa932bd0-- From adeason@sinenomine.net Tue Aug 16 02:51:17 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 15 Aug 2011 21:51:17 -0400 Subject: [OpenAFS] Re: kernel panic in 1.4.14.1 on rhel6 References: <4E304FF4.8070403@sinenomine.net> <20110727140512.72b69950.adeason@sinenomine.net> <20110727153702.e85ebd45.adeason@sinenomine.net> <20110812110103.54e316dd.adeason@sinenomine.net> Message-ID: <20110815215117.75f5559c.adeason@sinenomine.net> On Mon, 15 Aug 2011 16:26:53 -0400 Aaron Knister wrote: > Those patches do prevent it from crashing, but also results in a > number of "permission denied" errors from the apache processes under > heavy load from ab. The permission denied errors don't occur with > 1.6.0pre7. The easiest explanation would be the tokens being bad or corrupted or something... but I don't suppose you see any "afs: Tokens for user X" messages in the kernel log around this time? -- Andrew Deason adeason@sinenomine.net From wayne_greene@unc.edu Tue Aug 16 03:01:44 2011 From: wayne_greene@unc.edu (Wayne Greene) Date: Mon, 15 Aug 2011 22:01:44 -0400 Subject: [OpenAFS] Win7 64-bit not getting tokens on login Message-ID: <4E49CF88.4080307@unc.edu> I recently started using the Windows (WAIK) rather than Symantec Ghost imaging method to create images to apply to our systems. It works well except that I can no longer gain AFS tokens on login on the Windows 64-bit domain machines. I'm using MIT Kerberos 3.22 with the NetID Manager included, and the latest 64-bit 1.5.9906 package with the 32-bit tools installed. We can gain tokens by using the "Obtain new Credentials" choice from the "Credential" menu choice. This lack of credentials on login doesn't occur with the 32-bit image. Is anyone else out there seeing this and does anybody know if there a fix for it or more clues to lead to a fix? More details can be provided as necessary. Thanks. Wayne Greene Computer Services UNC Computer Science From l.schimmer@cgv.tugraz.at Tue Aug 16 09:26:00 2011 From: l.schimmer@cgv.tugraz.at (Lars Schimmer) Date: Tue, 16 Aug 2011 10:26:00 +0200 Subject: [OpenAFS] Win7 64-bit not getting tokens on login In-Reply-To: <4E49CF88.4080307@unc.edu> References: <4E49CF88.4080307@unc.edu> Message-ID: <4E4A2998.3010309@cgv.tugraz.at> On 16.08.2011 04:01, Wayne Greene wrote: > I recently started using the Windows (WAIK) rather than Symantec Ghost > imaging method to create images to apply to our systems. It works well > except that I can no longer gain AFS tokens on login on the Windows > 64-bit domain machines. I'm using MIT Kerberos 3.22 with the NetID > Manager included, and the latest 64-bit 1.5.9906 package with the 32-bi= t > tools installed. We can gain tokens by using the "Obtain new > Credentials" choice from the "Credential" menu choice. This lack of > credentials on login doesn't occur with the 32-bit image. Is anyone els= e > out there seeing this and does anybody know if there a fix for it or > more clues to lead to a fix? More details can be provided as necessary. > Thanks. What does the Windows logs tell you on the error? And the waitin gperiod after booting up is taken, or? > Wayne Greene > Computer Services > UNC Computer Science MfG, Lars Schimmer --=20 ------------------------------------------------------------- TU Graz, Institut f=FCr ComputerGraphik & WissensVisualisierung Tel: +43 316 873-5405 E-Mail: l.schimmer@cgv.tugraz.at Fax: +43 316 873-5402 PGP-Key-ID: 0x4A9B1723 From kallbac@iu.edu Tue Aug 16 14:38:59 2011 From: kallbac@iu.edu (Kristy Kallback-Rose) Date: Tue, 16 Aug 2011 09:38:59 -0400 Subject: [OpenAFS] OpenAFS 1.6.0pre7 and Mac OS 10.7 (Lion) cannot aklog In-Reply-To: References: <941254EC-1056-4D5F-AD68-CB80C2E7F990@indiana.edu> <0B2F97A1-724C-4DB6-8716-E821A27BC1ED@iu.edu> Message-ID: Many thanks. However, when I run the installer it causes a kernel panic. = It happens after the installer says that the installation is complete. I = imagine it is when it is trying to load the extension. After that, when I go back into the system and setup the CellServDB and = try to startup OpenAFS it also kernel panics. Let me know what additional information you might like. Cheers, Kristy PS - Sorry I didn't send this sooner. I had to be away for a few days. On Aug 11, 2011, at 10:37 AM, Derrick Brashear wrote: > The fix will be in 1.6.0 final. >=20 > In the meantime, you can have something for now from > = /afs/your-file-system.com/user/shadow/OpenAFS-1.6.0pre7-31-g24471-Lion.dmg= >=20 >=20 >=20 > On Thu, Aug 11, 2011 at 12:31 AM, Derrick Brashear = wrote: >> On Wed, Aug 10, 2011 at 4:56 PM, Kristy Kallback-Rose >> wrote: >>> Well, it appears I have one of the Macs which is blacklisted from = booting into 64-bit mode --though I can run Lion. If you want to know = more about what I'm referring to you can see the workaround that people = did for Snow Leopard and the same issue here http://netkas.org/?p=3D189 = Though I have not seem similar fixes for Lion, nor do I want to support = users doing that sort of fix. >>>=20 >>> Regarding getting OpenAFS working on this type of system, do you = have a recommendation for next steps? Should I build openafs from source = on this machine? We will no doubt have users running into this problem = as the semester begins. >>=20 >> given that the existing build is a universal build, i suspect >> something more subtle is in play. you can try building for source but >> i suspect you will have the same error. >>=20 >>=20 >>=20 >>=20 >>=20 >> -- >> Derrick >>=20 >=20 >=20 >=20 > --=20 > Derrick > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info Kristy Kallback-Rose Research Storage kallbac@iu.edu From aaronk@umbc.edu Wed Aug 17 15:59:30 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Wed, 17 Aug 2011 10:59:30 -0400 Subject: [OpenAFS] 64-bit mod_waklog / filedrawers on afs In-Reply-To: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBE2@EXEVS04.its.uncc.edu> References: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBE2@EXEVS04.its.uncc.edu> Message-ID: --0022152d63298da04c04aab4bfdf Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable Hi Daniel, Well... I've never built filedrawers, but this line: tpupload.c:18:19: error: mysql.h: No such file or directory suggests you're missing the mysql headers. If its redhat try a yum install mysql-devel. -Aaron On Tue, Aug 16, 2011 at 4:50 PM, Colson, Daniel wrote: > ** > > Thanks. Any help is appreciated. Here's the error output from "make". > > make[1]: Entering directory `/build/filedrawers/libcgi= ' > gcc -Wall -Wstrict-prototypes -g -fPIC -I/usr/include/apr-1 -c -o cgi.o > cgi.c > cgi.c: In function =91post_getline=92: > cgi.c:91: warning: function declaration isn=92t a prototype > cgi.c: In function =91cgi_init=92: > cgi.c:148: warning: function declaration isn=92t a prototype > cgi.c: In function =91mp_get_file=92: > cgi.c:361: warning: function declaration isn=92t a prototype > cgi.c:364: warning: function declaration isn=92t a prototype > cgi.c:375: warning: function declaration isn=92t a prototype > cgi.c:378: warning: function declaration isn=92t a prototype > cgi.c:385: warning: function declaration isn=92t a prototype > cgi.c:388: warning: function declaration isn=92t a prototype > cgi.c:400: warning: function declaration isn=92t a prototype > cgi.c:420: warning: function declaration isn=92t a prototype > cgi.c:424: warning: function declaration isn=92t a prototype > cgi.c: In function =91cgi_multipart=92: > cgi.c:539: warning: function declaration isn=92t a prototype > cgi.c:614: warning: function declaration isn=92t a prototype > building normal libcgi.a > make[1]: Leaving directory `/build/filedrawers/libcgi' > ( cd upload; make ) > make[1]: Entering directory `/build/filedrawers/upload= ' > gcc -Wall -Wmissing-prototypes -g -O2 -I../libcgi -I/usr/include/mysql -= g > -pipe -Wp,-D_FORTIFY_SOURCE=3D2 -fexceptions -fstack-protector > --param=3Dssp-buffer-size=3D4 -m64 -D_GNU_SOURCE -D_FILE_OFFSET_BITS=3D64 > -D_LARGEFILE_SOURCE -fno-strict-aliasing -fwrapv -Wall -Wmissing-prototyp= es > \ > -D_URL_ERROR_PAGE=3D\"https://?error=3Dtrue\" \ > > -D_FILEDRAWERS_CONFIG_PATH=3D\"/var/www/filedrawers/lib/filedrawers.conf\= " \ > -c tpupload.c > tpupload.c:18:19: error: mysql.h: No such file or directory > tpupload.c:34: error: expected =91=3D=92, =91,=92, =91;=92, =91asm=92 or = =91__attribute__=92 > before =91*=92 token > tpupload.c:35: error: expected =91=3D=92, =91,=92, =91;=92, =91asm=92 or = =91__attribute__=92 > before =91*=92 token > tpupload.c: In function =91upload_progress_db_insert=92: > tpupload.c:103: error: =91MYSQL_BIND=92 undeclared (first use in this fun= ction) > tpupload.c:103: error: (Each undeclared identifier is reported only once > tpupload.c:103: error: for each function it appears in.) > tpupload.c:103: error: expected =91;=92 before =91bind=92 > tpupload.c:104: error: =91my_bool=92 undeclared (first use in this functi= on) > tpupload.c:104: error: expected =91;=92 before =91is_null=92 > tpupload.c:110: error: =91stmt=92 undeclared (first use in this function) > tpupload.c:111: warning: implicit declaration of function =91mysql_stmt_i= nit=92 > tpupload.c:111: error: =91mysql=92 undeclared (first use in this function= ) > tpupload.c:112: warning: implicit declaration of function =91mysql_error= =92 > tpupload.c:112: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c:123: warning: implicit declaration of function > =91mysql_stmt_prepare=92 > tpupload.c:124: warning: implicit declaration of function > =91mysql_stmt_error=92 > tpupload.c:124: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c:130: error: =91bind=92 undeclared (first use in this function) > tpupload.c:130: warning: left-hand operand of comma expression has no > effect > tpupload.c:132: error: =91MYSQL_TYPE_STRING=92 undeclared (first use in t= his > function) > tpupload.c:132: error: expected expression before =91)=92 token > tpupload.c:132: error: =91is_null=92 undeclared (first use in this functi= on) > tpupload.c:133: error: expected expression before =91)=92 token > tpupload.c:134: error: =91MYSQL_TYPE_LONGLONG=92 undeclared (first use in= this > function) > tpupload.c:134: error: expected expression before =91)=92 token > tpupload.c:135: error: expected expression before =91)=92 token > tpupload.c:136: error: =91MYSQL_TYPE_TINY=92 undeclared (first use in thi= s > function) > tpupload.c:136: error: expected expression before =91)=92 token > tpupload.c:138: warning: implicit declaration of function > =91mysql_stmt_bind_param=92 > tpupload.c:139: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c:148: warning: implicit declaration of function > =91mysql_stmt_execute=92 > tpupload.c:149: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c: In function =91upload_progress_db_update=92: > tpupload.c:160: error: =91MYSQL_BIND=92 undeclared (first use in this fun= ction) > tpupload.c:160: error: expected =91;=92 before =91bind=92 > tpupload.c:161: error: =91my_bool=92 undeclared (first use in this functi= on) > tpupload.c:161: error: expected =91;=92 before =91is_null=92 > tpupload.c:166: error: =91stmt=92 undeclared (first use in this function) > tpupload.c:167: error: =91mysql=92 undeclared (first use in this function= ) > tpupload.c:168: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c:180: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c:186: error: =91bind=92 undeclared (first use in this function) > tpupload.c:186: warning: left-hand operand of comma expression has no > effect > tpupload.c:188: error: =91MYSQL_TYPE_STRING=92 undeclared (first use in t= his > function) > tpupload.c:188: error: expected expression before =91)=92 token > tpupload.c:188: error: =91is_null=92 undeclared (first use in this functi= on) > tpupload.c:189: error: =91MYSQL_TYPE_LONGLONG=92 undeclared (first use in= this > function) > tpupload.c:189: error: expected expression before =91)=92 token > tpupload.c:190: error: =91MYSQL_TYPE_TINY=92 undeclared (first use in thi= s > function) > tpupload.c:190: error: expected expression before =91)=92 token > tpupload.c:191: error: expected expression before =91)=92 token > tpupload.c:194: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c:203: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c: In function =91upload_init=92: > tpupload.c:317: error: =91mysql=92 undeclared (first use in this function= ) > tpupload.c:322: error: =91stmt=92 undeclared (first use in this function) > tpupload.c:323: warning: implicit declaration of function > =91mysql_stmt_close=92 > tpupload.c:324: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c: In function =91upload_progress=92: > tpupload.c:347: error: =91mysql=92 undeclared (first use in this function= ) > tpupload.c: In function =91upload_complete=92: > tpupload.c:373: error: =91mysql=92 undeclared (first use in this function= ) > tpupload.c: In function =91main=92: > tpupload.c:528: error: =91mysql=92 undeclared (first use in this function= ) > tpupload.c:528: warning: implicit declaration of function =91mysql_init= =92 > tpupload.c:530: warning: implicit declaration of function > =91mysql_real_connect=92 > tpupload.c:532: warning: format =91%s=92 expects type =91char *=92, but a= rgument 4 > has type =91int=92 > tpupload.c:553: error: =91stmt=92 undeclared (first use in this function) > tpupload.c:556: warning: implicit declaration of function =91mysql_close= =92 > make[1]: *** [tpupload.o] Error 1 > make[1]: Leaving directory `/build/filedrawers/upload' > make: *** [upload/upload.cgi] Error 2 > > -Daniel > > > -----Original Message----- > From: Aaron Knister [mailto:aaronk@umbc.edu ] > Sent: Mon 8/15/2011 12:32 PM > To: Colson, Daniel > Cc: openafs-info@openafs.org > Subject: Re: [OpenAFS] 64-bit mod_waklog / filedrawers on afs > > Hi Daniel, > > While I'm not running filedrawers I am using mod_waklog, albeit with RHEL= 6. > I had to apply some patches to get it to build correctly. If you send you= r > build errors I can see if my patches will help. > > -Aaron > > On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel wrote= : > > > ** > > > > Is anyone else running mod_waklog / filedrawers on afs on a 64-bit > machine? > > I've been unable to compile mod_waklog (an essential component of > > filedrawers) on 64-bit (RHEL5.6). > > > > > > -- > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu > > --=20 Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --0022152d63298da04c04aab4bfdf Content-Type: text/html; charset=windows-1252 Content-Transfer-Encoding: quoted-printable
Hi Daniel,

Well... I've never built filedrawer= s, but this line:

tpupload.c:18:19: error: mysql.h: No s= uch file or directory

suggests you're missing the mys= ql headers. If its redhat try a yum install mysql-devel.

-Aaron

On Tue,= Aug 16, 2011 at 4:50 PM, Colson, Daniel <dcolson5@uncc.edu> wrote:

Thanks. Any help is appreciated. Here's the error o= utput from "make".

make[1]: Entering directory `<path-to-build-dir>/build/filedrawers/li= bcgi'
gcc -Wall -Wstrict-prototypes -g=A0 -fPIC -I/usr/include/apr-1=A0 -c -o cgi= .o cgi.c
cgi.c: In function =91post_getline=92:
cgi.c:91: warning: function declaration isn=92t a prototype
cgi.c: In function =91cgi_init=92:
cgi.c:148: warning: function declaration isn=92t a prototype
cgi.c: In function =91mp_get_file=92:
cgi.c:361: warning: function declaration isn=92t a prototype
cgi.c:364: warning: function declaration isn=92t a prototype
cgi.c:375: warning: function declaration isn=92t a prototype
cgi.c:378: warning: function declaration isn=92t a prototype
cgi.c:385: warning: function declaration isn=92t a prototype
cgi.c:388: warning: function declaration isn=92t a prototype
cgi.c:400: warning: function declaration isn=92t a prototype
cgi.c:420: warning: function declaration isn=92t a prototype
cgi.c:424: warning: function declaration isn=92t a prototype
cgi.c: In function =91cgi_multipart=92:
cgi.c:539: warning: function declaration isn=92t a prototype
cgi.c:614: warning: function declaration isn=92t a prototype
building normal libcgi.a
make[1]: Leaving directory `<path-to-build-dir>/build/filedrawers/lib= cgi'
( cd upload; make=A0 )
make[1]: Entering directory `<path-to-build-dir>/build/filedrawers/up= load'
gcc -Wall -Wmissing-prototypes -g -O2 -I../libcgi -I/usr/include/mysql=A0 -= g -pipe -Wp,-D_FORTIFY_SOURCE=3D2 -fexceptions -fstack-protector --param=3D= ssp-buffer-size=3D4 -m64 -D_GNU_SOURCE -D_FILE_OFFSET_BITS=3D64 -D_LARGEFIL= E_SOURCE -fno-strict-aliasing -fwrapv -Wall -Wmissing-prototypes \
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 -D_URL_ERROR_PAGE=3D\"https://<site>?error=3Dtrue\" \
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 -D_FILEDRAWERS_CONFIG_PATH=3D= \"/var/www/filedrawers/lib/filedrawers.conf\" \
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 -c tpupload.c
tpupload.c:18:19: error: mysql.h: No such file or directory
tpupload.c:34: error: expected =91=3D=92, =91,=92, =91;=92, =91asm=92 or = =91__attribute__=92 before =91*=92 token
tpupload.c:35: error: expected =91=3D=92, =91,=92, =91;=92, =91asm=92 or = =91__attribute__=92 before =91*=92 token
tpupload.c: In function =91upload_progress_db_insert=92:
tpupload.c:103: error: =91MYSQL_BIND=92 undeclared (first use in this funct= ion)
tpupload.c:103: error: (Each undeclared identifier is reported only once tpupload.c:103: error: for each function it appears in.)
tpupload.c:103: error: expected =91;=92 before =91bind=92
tpupload.c:104: error: =91my_bool=92 undeclared (first use in this function= )
tpupload.c:104: error: expected =91;=92 before =91is_null=92
tpupload.c:110: error: =91stmt=92 undeclared (first use in this function) tpupload.c:111: warning: implicit declaration of function =91mysql_stmt_ini= t=92
tpupload.c:111: error: =91mysql=92 undeclared (first use in this function)<= br> tpupload.c:112: warning: implicit declaration of function =91mysql_error=92=
tpupload.c:112: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c:123: warning: implicit declaration of function =91mysql_stmt_pre= pare=92
tpupload.c:124: warning: implicit declaration of function =91mysql_stmt_err= or=92
tpupload.c:124: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c:130: error: =91bind=92 undeclared (first use in this function) tpupload.c:130: warning: left-hand operand of comma expression has no effec= t
tpupload.c:132: error: =91MYSQL_TYPE_STRING=92 undeclared (first use in thi= s function)
tpupload.c:132: error: expected expression before =91)=92 token
tpupload.c:132: error: =91is_null=92 undeclared (first use in this function= )
tpupload.c:133: error: expected expression before =91)=92 token
tpupload.c:134: error: =91MYSQL_TYPE_LONGLONG=92 undeclared (first use in t= his function)
tpupload.c:134: error: expected expression before =91)=92 token
tpupload.c:135: error: expected expression before =91)=92 token
tpupload.c:136: error: =91MYSQL_TYPE_TINY=92 undeclared (first use in this = function)
tpupload.c:136: error: expected expression before =91)=92 token
tpupload.c:138: warning: implicit declaration of function =91mysql_stmt_bin= d_param=92
tpupload.c:139: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c:148: warning: implicit declaration of function =91mysql_stmt_exe= cute=92
tpupload.c:149: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c: In function =91upload_progress_db_update=92:
tpupload.c:160: error: =91MYSQL_BIND=92 undeclared (first use in this funct= ion)
tpupload.c:160: error: expected =91;=92 before =91bind=92
tpupload.c:161: error: =91my_bool=92 undeclared (first use in this function= )
tpupload.c:161: error: expected =91;=92 before =91is_null=92
tpupload.c:166: error: =91stmt=92 undeclared (first use in this function) tpupload.c:167: error: =91mysql=92 undeclared (first use in this function)<= br> tpupload.c:168: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c:180: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c:186: error: =91bind=92 undeclared (first use in this function) tpupload.c:186: warning: left-hand operand of comma expression has no effec= t
tpupload.c:188: error: =91MYSQL_TYPE_STRING=92 undeclared (first use in thi= s function)
tpupload.c:188: error: expected expression before =91)=92 token
tpupload.c:188: error: =91is_null=92 undeclared (first use in this function= )
tpupload.c:189: error: =91MYSQL_TYPE_LONGLONG=92 undeclared (first use in t= his function)
tpupload.c:189: error: expected expression before =91)=92 token
tpupload.c:190: error: =91MYSQL_TYPE_TINY=92 undeclared (first use in this = function)
tpupload.c:190: error: expected expression before =91)=92 token
tpupload.c:191: error: expected expression before =91)=92 token
tpupload.c:194: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c:203: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c: In function =91upload_init=92:
tpupload.c:317: error: =91mysql=92 undeclared (first use in this function)<= br> tpupload.c:322: error: =91stmt=92 undeclared (first use in this function) tpupload.c:323: warning: implicit declaration of function =91mysql_stmt_clo= se=92
tpupload.c:324: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c: In function =91upload_progress=92:
tpupload.c:347: error: =91mysql=92 undeclared (first use in this function)<= br> tpupload.c: In function =91upload_complete=92:
tpupload.c:373: error: =91mysql=92 undeclared (first use in this function)<= br> tpupload.c: In function =91main=92:
tpupload.c:528: error: =91mysql=92 undeclared (first use in this function)<= br> tpupload.c:528: warning: implicit declaration of function =91mysql_init=92<= br> tpupload.c:530: warning: implicit declaration of function =91mysql_real_con= nect=92
tpupload.c:532: warning: format =91%s=92 expects type =91char *=92, but arg= ument 4 has type =91int=92
tpupload.c:553: error: =91stmt=92 undeclared (first use in this function) tpupload.c:556: warning: implicit declaration of function =91mysql_close=92=
make[1]: *** [tpupload.o] Error 1
make[1]: Leaving directory `<path-to-build-dir>/build/filedrawers/upl= oad'
make: *** [upload/upload.cgi] Error 2

-Daniel



-----Original Message-----
From: Aaron Knister [m= ailto:aaronk@umbc.edu]
Sent: Mon 8/15/2011 12:32 PM
To: Colson, Daniel
Cc: openafs-i= nfo@openafs.org
Subject: Re: [OpenAFS] 64-bit mod_waklog / filedrawers on afs

Hi Daniel,

While I'm not running filedrawers I am using mod_waklog, albeit with RH= EL6.
I had to apply some patches to get it to build correctly. If you send your<= br> build errors I can see if my patches will help.

-Aaron

On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel <dcolson5@uncc.edu> wrote:

> **
>
> Is anyone else running mod_waklog / filedrawers on afs on a 64-bit mac= hine?
> I've been unable to compile mod_waklog (an essential component of<= br> > filedrawers) on 64-bit (RHEL5.6).
>



--
Aaron Knister
Systems Administrator
Division of Information Technology
University of Maryland, Baltimore County
aaronk@umbc.edu




--
Aaron Kniste= r
Systems Administrator
Division of Information Technology
Univers= ity of Maryland, Baltimore County
aaronk@umbc.edu
--0022152d63298da04c04aab4bfdf-- From jwinius@umrk.nl Wed Aug 17 22:03:44 2011 From: jwinius@umrk.nl (Jaap Winius) Date: Wed, 17 Aug 2011 23:03:44 +0200 Subject: [OpenAFS] Performance issues Message-ID: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> Hi folks, Some six months after launching my first AFS project, I've learned some things but would also like to make some improvements. My site has three locations, each with one physical server machine running Debian squeeze, as well as two Internet connections. To avoid Kerberos identity problems among the three servers, I've set up one virtual host per server machine, each with a single interface and a public IP address. At first these virtual machines hosted both our integrated Kerberos and OpenLDAP servers, as well as our OpenAFS database and file servers. In this configuration, all AFS traffic, even local traffic, was forced to traverse a firewall and a NAT. In tests it all seemed okay, but in practice performance could be quite poor, often getting worse after several hours of use. At first we thought the cause was due to the UDP connections timing out, but the problems persisted even after we had compensated and no more AFS packets were being dropped. Last weekend I decided to make a major change to our AFS server architecture. I created a new AFS file server on the base OS of each of the three physical servers, moved all of the AFS volumes there and removed the file server processes from the virtual hosts. With NetInfo and NetRestrict, I set each of the new file servers to register only one public IP address in the VLDB. Now local AFS file server traffic is no longer forced to traverse the firewall/NAT. This was a great improvement, so that people with local user volumes are at last experiencing performance that can be described as normal. However, if anything this has only made the contrast between local and remote performance even more acute. After all, nothing has changed for the remote users: not only do their connections have less bandwidth (6 Mbps) and higher latency (25-30 ms), but their traffic must also traverse two firewalls and a local NAT. What is currently likely the main reason for this difference in performance? If it's mostly due to the NAT and the fact that the firewalls must be traversed and the UPD connections kept track of, then I could set up some tunnels to avoid most of that. But, if it's just the higher latency and limited bandwidth, then I suppose there's not much to do about it. Thanks, Jaap From peter.van.der.reest@desy.de Sun Aug 21 14:49:15 2011 From: peter.van.der.reest@desy.de (Peter van der Reest) Date: Sun, 21 Aug 2011 15:49:15 +0200 Subject: [OpenAFS] Announcing the 2011 European AFS and Kerberos Conference & Call for Participation Message-ID: <86BAD050-4C5D-45F6-AB6D-FB826979DDBC@desy.de> --Apple-Mail-17-619906267 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 This is the long due public announcement of the=20 2011 European AFS and Kerberos Conference taking place in Hamburg, Germany from Oct 4th through Oct 7th 2011. Please find details at https://indico.desy.de/event/eakc2011 The call for abstracts is open and do feel free to submit your = presentation=20 proposals. As always the conference will examine the development outlook for AFS = and=20 Kerberos implementations, it will highlight current projects and will = offer=20 space to proposals and new ideas. Also, sites will be able to present = their AFS and Kerberos activities in site reports. On the last day of the conference, we are planning to have time for = tutorials, hands-on workshops and for those interested tours of the DESY research = site. As always, the program will allow for informal discussions and exchanges = as=20 well. Due to some formal constraints, the conference attendance registration = will=20 open in the course of the coming week. The conference fee for academic attendees will be under =80 150, for = industrial=20 attendees under =80 500. Vendor representatives fees are not yet fixed. Please refer to the conference secretariat or myself for questions - a = formal=20 e-mail adress is in preparation and will be published on the conference = site. With best regards and hoping to see you in Hamburg, for the organizing team, Peter van der Reest. = __________________________________________________________________________= ______ Peter van der Reest Information = Technology Deutsches Elektronen-Synchrotron - = DESY - A Research Centre of the Helmholtz = Association phone: +49 40 8998 3315 = Notkestrasse 85 fax: +49 40 8994 3315 D-22607 = Hamburg e-mail: peter.van.der.reest@desy.de = Germany = __________________________________________________________________________= ______ --Apple-Mail-17-619906267 Content-Disposition: attachment; filename=smime.p7s Content-Type: application/pkcs7-signature; name=smime.p7s Content-Transfer-Encoding: base64 MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIIFTDCCBUgw ggQwoAMCAQICBBGLEbMwDQYJKoZIhvcNAQEFBQAwgYIxCzAJBgNVBAYTAkRFMS4wLAYDVQQKEyVE ZXV0c2NoZXMgRWxla3Ryb25lbi1TeW5jaHJvdHJvbiBERVNZMQswCQYDVQQLEwJJVDEWMBQGA1UE AxMNREVTWSBDQSAtIEcwMjEeMBwGCSqGSIb3DQEJARYPZGVzeS1jYUBkZXN5LmRlMB4XDTExMDIy ODEyMjUwN1oXDTE0MDIyNzEyMjUwN1owfDELMAkGA1UEBhMCREUxLjAsBgNVBAoTJURldXRzY2hl cyBFbGVrdHJvbmVuLVN5bmNocm90cm9uIERFU1kxHzAdBgNVBAsTFkluZm9ybWF0aW9uIFRlY2hu b2xvZ3kxHDAaBgNVBAMTE1BldGVyIHZhbiBkZXIgUmVlc3QwggEiMA0GCSqGSIb3DQEBAQUAA4IB DwAwggEKAoIBAQC/rNIwpVWl9hstkpRMckPI/WPWJQcofXwDxeXNbEjeLeZ2nigRJmVDkQZjMG7K Kiau46+UiDCy5fqGHBXtOzJgyF9DPeREuvX293NzsiQvnUNMu90tOYVg5DMIljWUE3rjeiRD7Ftx e9j87fR0HcvIECqQNI/ax17rq1JGrYrRgTp+UIi/++gaOuaglMjdBi11zwhcoWSm1pHLPn61Ynh+ QeaAC9SF5qzLb70XiUkOi90FMnW1AwR+v2BH4wKFjuFYZ6o0pECOIWpMsnJZgXUlk3PQu23thXtP N7RE27dfHRTp4E9f2b1J21nJEjhOwdbyMc0YYZabey89n+x1WtAtAgMBAAGjggHJMIIBxTAJBgNV HRMEAjAAMAsGA1UdDwQEAwIF4DApBgNVHSUEIjAgBggrBgEFBQcDAgYIKwYBBQUHAwQGCisGAQQB gjcUAgIwHQYDVR0OBBYEFI/ES8NGdDKp+N+rKWP/fV1VaB7KMB8GA1UdIwQYMBaAFJSX58dzgIIi jnL6/92eNIXjtHYUMCYGA1UdEQQfMB2BG3BldGVyLnZhbi5kZXIucmVlc3RAZGVzeS5kZTB9BgNV HR8EdjB0MDigNqA0hjJodHRwOi8vY2RwMS5wY2EuZGZuLmRlL2Rlc3ktY2EvcHViL2NybC9nX2Nh Y3JsLmNybDA4oDagNIYyaHR0cDovL2NkcDIucGNhLmRmbi5kZS9kZXN5LWNhL3B1Yi9jcmwvZ19j YWNybC5jcmwwgZgGCCsGAQUFBwEBBIGLMIGIMEIGCCsGAQUFBzAChjZodHRwOi8vY2RwMS5wY2Eu ZGZuLmRlL2Rlc3ktY2EvcHViL2NhY2VydC9nX2NhY2VydC5jcnQwQgYIKwYBBQUHMAKGNmh0dHA6 Ly9jZHAyLnBjYS5kZm4uZGUvZGVzeS1jYS9wdWIvY2FjZXJ0L2dfY2FjZXJ0LmNydDANBgkqhkiG 9w0BAQUFAAOCAQEAg5IoNEPpwACklnsGj6lLvRBHFweqxe4AkqjmCUcV0Ul08Jx9cY2u+nl8YIR/ j7fpftge7YgSsATtKcX5yLRG4UcqN4TYZdmjLH4ZbzxVzXPgaGghJlDWnW/t8/pTH5G1zoie/SE2 C31C+T4/lmh9lwfCspwEiLfcoG9B/uJIGjVgfKWeSSjcOEQ8pI7PCPlbadVlN272C2xy4r5cWZsQ zlOMUF3U0iSthF9eD9D743PGDBIXAk06AU8ltcgw9sNIHWBLZsxOekJ96cGjOJdJrGvZ3gSDHKvP jOUCectKLEzoteH8tB1bvzs7G+wq7OKK1CJdWPCeEqQMss47FC7pNTGCA1QwggNQAgEBMIGLMIGC MQswCQYDVQQGEwJERTEuMCwGA1UEChMlRGV1dHNjaGVzIEVsZWt0cm9uZW4tU3luY2hyb3Ryb24g REVTWTELMAkGA1UECxMCSVQxFjAUBgNVBAMTDURFU1kgQ0EgLSBHMDIxHjAcBgkqhkiG9w0BCQEW D2Rlc3ktY2FAZGVzeS5kZQIEEYsRszAJBgUrDgMCGgUAoIIBnTAYBgkqhkiG9w0BCQMxCwYJKoZI hvcNAQcBMBwGCSqGSIb3DQEJBTEPFw0xMTA4MjExMzQ5MTVaMCMGCSqGSIb3DQEJBDEWBBQ9SHgB ow9fILMppCPbEC0MXGzRuTCBnAYJKwYBBAGCNxAEMYGOMIGLMIGCMQswCQYDVQQGEwJERTEuMCwG A1UEChMlRGV1dHNjaGVzIEVsZWt0cm9uZW4tU3luY2hyb3Ryb24gREVTWTELMAkGA1UECxMCSVQx FjAUBgNVBAMTDURFU1kgQ0EgLSBHMDIxHjAcBgkqhkiG9w0BCQEWD2Rlc3ktY2FAZGVzeS5kZQIE EYsRszCBngYLKoZIhvcNAQkQAgsxgY6ggYswgYIxCzAJBgNVBAYTAkRFMS4wLAYDVQQKEyVEZXV0 c2NoZXMgRWxla3Ryb25lbi1TeW5jaHJvdHJvbiBERVNZMQswCQYDVQQLEwJJVDEWMBQGA1UEAxMN REVTWSBDQSAtIEcwMjEeMBwGCSqGSIb3DQEJARYPZGVzeS1jYUBkZXN5LmRlAgQRixGzMA0GCSqG SIb3DQEBAQUABIIBAGExP0EhQxZqr9TgRoiQSwg+jkaW/NcUo1TLWSyBX+315CLj5UOnY7wWmaIZ vE3WYg8K8aKcCJQKZxzicTO5t2VciyhLjn0WCZkWfKtHhUBneqMW3gjpU6w4ykvXRQoW0Hbd+lil wKbaGbaQHpt0Bzke01dNtMJcvq4rmZ66+YFEHKoASlyCz9s+H+3j2wKPtAqbnpGuMEPTGOHFisq5 EKJ3snDKm2g7dMd1SXcssIMlDchX10O9ZowKRgwFrLMkj4TC+KE3fL5kBO3xREonNKw+pozD6tHg jQkdjdk4DAoMUxmEYcg5oALJBlWnElyb1qyHREtulki+q4GiLEIw+csAAAAAAAA= --Apple-Mail-17-619906267-- From yentj@infowrap.com.tw Mon Aug 22 05:49:51 2011 From: yentj@infowrap.com.tw (Ting-jen Yen) Date: Mon, 22 Aug 2011 12:49:51 +0800 Subject: [OpenAFS] Install OpenAFS on ppc64? Message-ID: <1313988591.6154.17.camel@yentj.infowrap.com.tw> Hello, Has anyone successfully installed OpenAFS on RHEL5 machines with ppc64 arch? (Power6 CPU, virtualized by PowerVM). I was using SRPMs from the OpenAFS site. Rpmbuild showed some errors, and I have to specify to build userspace, modules, and authlibs packages, then modify the source, 'flock64' to 'flock' in src/afs/afs.h. This way, I could build the RPMs, and install them. However, even though the installation succeeded, I cannot start the client service. If I added options '-debug' and '-verbose' to afsd, it will show afsd: enabling dynamically allocated vcaches afsd: My home cell is 'mycell' ParseCacheInfoFile: Opening cache info file '/usr/vice/etc/cacheinfo'... ParseCacheInfoFile: Cache info file successfully parsed: cacheMountDir: '/afs' cacheBaseDir: '/usr/vice/cache' cacheBlocks: 100000 afsd: cacheFiles autotuned to 3125 afsd: dCacheSize autotuned to 2000 afsd: cacheStatEntries autotuned to 3000 SScall(137, 28, 35)=-1 SScall(137, 28, 17)=-1 afsd: Forking rx listener daemon. SScall(137, 28, 48)=-1 afsd: Forking rx callback listener. SScall(137, 28, 0)=-1 afsd: Forking rxevent daemon. SScall(137, 28, 19)=-1 SScall(137, 28, 36)=-1 afsd: Error -1 in basic initialization. Anybody could help me with this? I tried 1.4.7, 1.4.14, 1.5.78, and 1.6.0-pre1, and all with the same result. Other machines with x86_64 work fine. -- -------------------------------- Ting-jen Yen, Assistant Manager R&D Dep. INFOWRAP TECHNOLOGIES, INC. No.2060, Yucheng Rd., Gushan District, Kaohsiung City 80458, Taiwan Tel: 886-7-5228611 Ext.22 Fax: 886-7-5228986 E-mail: yentj@infowrap.com.tw From cedric.hernalsteens@gmail.com Mon Aug 22 11:00:30 2011 From: cedric.hernalsteens@gmail.com (=?ISO-8859-1?Q?C=E9dric_Hernalsteens?=) Date: Mon, 22 Aug 2011 12:00:30 +0200 Subject: [OpenAFS] OpenAFS 'preference panel' and 'menu' broken in OS X Lion ? Message-ID: Hi all, I'm quite new to OpenAFS but I noticed that since I reinstalled completely OpenAFS 1.6.0pre7 for Lion I can't have the OpenAFS menu and any action in the preference panel seems to be without effect. Is that a real issue ? Thanks Cedric From shadow@gmail.com Mon Aug 22 14:00:41 2011 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 22 Aug 2011 09:00:41 -0400 Subject: [OpenAFS] OpenAFS 'preference panel' and 'menu' broken in OS X Lion ? In-Reply-To: References: Message-ID: There are bugs in pre7's backgrounder due to the unfortunate Kerberos changes in Lion. 1.6.0 final includes fixes. On Mon, Aug 22, 2011 at 6:00 AM, C=E9dric Hernalsteens wrote: > Hi all, > > I'm quite new to OpenAFS but I noticed that since I reinstalled > completely OpenAFS 1.6.0pre7 for Lion I can't have the OpenAFS menu > and any action in the preference panel seems to be without effect. > > Is that a real issue ? > > Thanks > Cedric > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > --=20 Derrick From rcc2@cornell.edu Mon Aug 22 15:07:28 2011 From: rcc2@cornell.edu (Rick Cochran) Date: Mon, 22 Aug 2011 10:07:28 -0400 Subject: [OpenAFS] OpenAFS 'preference panel' and 'menu' broken in OS X Lion ? In-Reply-To: References: Message-ID: <4E5262A0.4070304@cornell.edu> My understanding of the Kerberos changes is Lion is: o Weak encryption types are now dis-allowed. o TCP protocol is used when large TGTs (eg. from Active Directory KDCs)= are=20 involved. Although potentially disruptive, these seem logical to me. Do you know of any others? -Rick On 08/22/2011 09:00 AM, Derrick Brashear wrote: > There are bugs in pre7's backgrounder due to the unfortunate Kerberos > changes in Lion. 1.6.0 final includes fixes. > > On Mon, Aug 22, 2011 at 6:00 AM, C=EF=BF=BDdric Hernalsteens > wrote: >> Hi all, >> >> I'm quite new to OpenAFS but I noticed that since I reinstalled >> completely OpenAFS 1.6.0pre7 for Lion I can't have the OpenAFS menu >> and any action in the preference panel seems to be without effect. >> >> Is that a real issue ? >> >> Thanks >> Cedric >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info >> > > > From shadow@gmail.com Mon Aug 22 15:41:51 2011 From: shadow@gmail.com (Derrick Brashear) Date: Mon, 22 Aug 2011 10:41:51 -0400 Subject: [OpenAFS] OpenAFS 'preference panel' and 'menu' broken in OS X Lion ? In-Reply-To: <4E5262A0.4070304@cornell.edu> References: <4E5262A0.4070304@cornell.edu> Message-ID: On Mon, Aug 22, 2011 at 10:07 AM, Rick Cochran wrote: > My understanding of the Kerberos changes is Lion is: > > =A0o Weak encryption types are now dis-allowed. > > =A0o TCP protocol is used when large TGTs (eg. from Active Directory KDCs= ) are > involved. > > Although potentially disruptive, these seem logical to me. > > Do you know of any others? oh, sadly, yes. most of the API is missing, including the "yes, we know we have weak crypto on" API calls, and the calls to translate error messages. in some cases "missing" means not "no routine" but instead "routine which returns something, possibly success, despite not being implemented" --=20 Derrick From sxw@inf.ed.ac.uk Mon Aug 22 17:07:53 2011 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Mon, 22 Aug 2011 17:07:53 +0100 Subject: [OpenAFS] OpenAFS 'preference panel' and 'menu' broken in OS X Lion ? In-Reply-To: <4E5262A0.4070304@cornell.edu> References: <4E5262A0.4070304@cornell.edu> Message-ID: <8067B96E-E4C1-49C4-BFA0-6B7A342E9153@inf.ed.ac.uk> On 22 Aug 2011, at 15:07, Rick Cochran wrote: > Although potentially disruptive, these seem logical to me. >=20 > Do you know of any others? The whole of MIT Kerberos has been ripped out and replaced with Heimdal. The= Heimdal API is private. Bits of the old MIT API are supported via a shim la= yer, but many of the functions we require are implemented as stubs. Some of t= hese stubs also have unhelpful return codes, making it difficult to tell the= difference between an unimplemented function and one returning an error. All in all, Kerberos in Lion is a bit of a mess. S. > -Rick >=20 > On 08/22/2011 09:00 AM, Derrick Brashear wrote: >> There are bugs in pre7's backgrounder due to the unfortunate Kerberos >> changes in Lion. 1.6.0 final includes fixes. >>=20 >> On Mon, Aug 22, 2011 at 6:00 AM, C=EF=BF=BDdric Hernalsteens >> wrote: >>> Hi all, >>>=20 >>> I'm quite new to OpenAFS but I noticed that since I reinstalled >>> completely OpenAFS 1.6.0pre7 for Lion I can't have the OpenAFS menu >>> and any action in the preference panel seems to be without effect. >>>=20 >>> Is that a real issue ? >>>=20 >>> Thanks >>> Cedric >>> _______________________________________________ >>> OpenAFS-info mailing list >>> OpenAFS-info@openafs.org >>> https://lists.openafs.org/mailman/listinfo/openafs-info >>>=20 >>=20 >>=20 >>=20 > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info >=20 From chas@cmf.nrl.navy.mil Mon Aug 22 18:08:33 2011 From: chas@cmf.nrl.navy.mil (chas williams - CONTRACTOR) Date: Mon, 22 Aug 2011 13:08:33 -0400 Subject: [OpenAFS] Install OpenAFS on ppc64? In-Reply-To: <1313988591.6154.17.camel@yentj.infowrap.com.tw> References: <1313988591.6154.17.camel@yentj.infowrap.com.tw> Message-ID: <20110822130833.6109e6ec@thirdoffive.cmf.nrl.navy.mil> i believe we have one ppc64 machine running openafs. the biggest deal was to make sure that the binaries were built 64-bit. the 32-bit versions of the ucontext syscalls are broken/incomplete. just based on your errors though it looks like the afs module isnt loaded? On Mon, 22 Aug 2011 12:49:51 +0800 Ting-jen Yen wrote: > Hello, > > Has anyone successfully installed OpenAFS on RHEL5 machines with > ppc64 arch? (Power6 CPU, virtualized by PowerVM). > I was using SRPMs from the OpenAFS site. Rpmbuild showed some > errors, and I have to specify to build userspace, modules, and > authlibs packages, then modify the source, 'flock64' to 'flock' in > src/afs/afs.h. This way, I could build the RPMs, and install > them. However, even though the installation succeeded, I cannot start > the client service. > > If I added options '-debug' and '-verbose' to afsd, it will show > > > afsd: enabling dynamically allocated vcaches > afsd: My home cell is 'mycell' > ParseCacheInfoFile: Opening cache info file '/usr/vice/etc/cacheinfo'... > ParseCacheInfoFile: Cache info file successfully parsed: > cacheMountDir: '/afs' > cacheBaseDir: '/usr/vice/cache' > cacheBlocks: 100000 > afsd: cacheFiles autotuned to 3125 > afsd: dCacheSize autotuned to 2000 > afsd: cacheStatEntries autotuned to 3000 > SScall(137, 28, 35)=-1 SScall(137, 28, 17)=-1 afsd: Forking rx listener > daemon. > SScall(137, 28, 48)=-1 afsd: Forking rx callback listener. > SScall(137, 28, 0)=-1 afsd: Forking rxevent daemon. > SScall(137, 28, 19)=-1 SScall(137, 28, 36)=-1 afsd: Error -1 in basic > initialization. > > Anybody could help me with this? I tried 1.4.7, 1.4.14, 1.5.78, and > 1.6.0-pre1, and all with the same result. Other machines with x86_64 > work fine. > From yentj@infowrap.com.tw Tue Aug 23 01:56:11 2011 From: yentj@infowrap.com.tw (Ting-jen Yen) Date: Tue, 23 Aug 2011 08:56:11 +0800 Subject: [OpenAFS] Install OpenAFS on ppc64? In-Reply-To: <20110822130833.6109e6ec@thirdoffive.cmf.nrl.navy.mil> References: <1313988591.6154.17.camel@yentj.infowrap.com.tw> <20110822130833.6109e6ec@thirdoffive.cmf.nrl.navy.mil> Message-ID: <1314060971.6543.10.camel@yentj.infowrap.com.tw> Hello, I know that judging from those error code -1 of the debug output, it looks like the problem is about module. However, I can see openafs module from lsmod even after the failure. 'dmesg' also shows: Cache bypass pathed libafs module init. Moreover, as I mentioned, I compiled those RPMs from the SRPM, and all of those RPMs are ppc64 arch. (Though I did notice when I used 'file' to check those binary files, more of them are reported as 32-bit. The module itself is 64-bit.) Regards, Ting-jen =A9=F3 =A4@=A1A2011-08-22 =A9=F3 13:08 -0400=A1Achas williams - CONTRACTO= R =B4=A3=A8=EC=A1G > i believe we have one ppc64 machine running openafs. the biggest deal > was to make sure that the binaries were built 64-bit. the 32-bit > versions of the ucontext syscalls are broken/incomplete. >=20 > just based on your errors though it looks like the afs module isnt > loaded? >=20 > On Mon, 22 Aug 2011 12:49:51 +0800 > Ting-jen Yen wrote: >=20 > > Hello, > >=20 > > Has anyone successfully installed OpenAFS on RHEL5 machines with > > ppc64 arch? (Power6 CPU, virtualized by PowerVM). > > I was using SRPMs from the OpenAFS site. Rpmbuild showed some > > errors, and I have to specify to build userspace, modules, and > > authlibs packages, then modify the source, 'flock64' to 'flock' in > > src/afs/afs.h. This way, I could build the RPMs, and install > > them. However, even though the installation succeeded, I cannot star= t > > the client service. > >=20 > > If I added options '-debug' and '-verbose' to afsd, it will show > >=20 > >=20 > > afsd: enabling dynamically allocated vcaches > > afsd: My home cell is 'mycell' > > ParseCacheInfoFile: Opening cache info file '/usr/vice/etc/cacheinfo'= ... > > ParseCacheInfoFile: Cache info file successfully parsed: > > cacheMountDir: '/afs' > > cacheBaseDir: '/usr/vice/cache'=20 > > cacheBlocks: 100000 > > afsd: cacheFiles autotuned to 3125=20 > > afsd: dCacheSize autotuned to 2000 > > afsd: cacheStatEntries autotuned to 3000 > > SScall(137, 28, 35)=3D-1 SScall(137, 28, 17)=3D-1 afsd: Forking rx li= stener > > daemon. > > SScall(137, 28, 48)=3D-1 afsd: Forking rx callback listener. > > SScall(137, 28, 0)=3D-1 afsd: Forking rxevent daemon. > > SScall(137, 28, 19)=3D-1 SScall(137, 28, 36)=3D-1 afsd: Error -1 in b= asic > > initialization. > >=20 > > Anybody could help me with this? I tried 1.4.7, 1.4.14, 1.5.78, an= d > > 1.6.0-pre1, and all with the same result. Other machines with x86_64 > > work fine. > >=20 >=20 From jwinius@umrk.nl Tue Aug 23 02:25:03 2011 From: jwinius@umrk.nl (Jaap Winius) Date: Tue, 23 Aug 2011 03:25:03 +0200 Subject: [OpenAFS] Deleted db servers still being queried Message-ID: <20110823032503.31267cw51wfrks08@bitis.umrk.nl> Hi folks, My AFS network used to have six hosts: thee db servers and three fs servers. Thinking that I might be able to end some unexplained behavior, I decided to consolidate them and have three hosts. I did this by creating a new db server on each fs server host and then deleting the db server that was closest to it. I deleted the old db servers by applying both the "bos removehost" command to each of the existing servers,"vos changeaddr" to remove the old IP addresses from the VLDB, as well as to restart each server after the entries had been removed. However, I must not have done this in the correct order or made a mistake, because even though the system seems to behave properly, I see evidence that the old db servers are still being queried. This despite the fact that they do not exist in any of the CellServDB files, or even the BosConfig files. Can anyone suggest steps that I might take to get rid of these old entries? Both "bos listhosts" on the different hosts and "vos listaddrs" do not show the old addresses, plus I've also tried shutting down all of the servers and then stating them up again. But, none of that makes any difference. Where could these old IP addresses be hiding? Thanks, Jaap From oreilly@qualcomm.com Tue Aug 23 02:35:21 2011 From: oreilly@qualcomm.com (Patricia O'Reilly) Date: Mon, 22 Aug 2011 18:35:21 -0700 Subject: [OpenAFS] Deleted db servers still being queried In-Reply-To: <20110823032503.31267cw51wfrks08@bitis.umrk.nl> References: <20110823032503.31267cw51wfrks08@bitis.umrk.nl> Message-ID: <4E5303D9.506@qualcomm.com> I believe you will need to restart all your client machines. Jaap Winius wrote: > Hi folks, > > My AFS network used to have six hosts: thee db servers and three fs > servers. Thinking that I might be able to end some unexplained behavior, > I decided to consolidate them and have three hosts. > > I did this by creating a new db server on each fs server host and then > deleting the db server that was closest to it. I deleted the old db > servers by applying both the "bos removehost" command to each of the > existing servers,"vos changeaddr" to remove the old IP addresses from > the VLDB, as well as to restart each server after the entries had been > removed. > > However, I must not have done this in the correct order or made a > mistake, because even though the system seems to behave properly, I see > evidence that the old db servers are still being queried. This despite > the fact that they do not exist in any of the CellServDB files, or even > the BosConfig files. > > Can anyone suggest steps that I might take to get rid of these old entries? > > Both "bos listhosts" on the different hosts and "vos listaddrs" do not > show the old addresses, plus I've also tried shutting down all of the > servers and then stating them up again. But, none of that makes any > difference. Where could these old IP addresses be hiding? > > Thanks, > > Jaap > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info From jaltman@secure-endpoints.com Tue Aug 23 02:40:53 2011 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Mon, 22 Aug 2011 21:40:53 -0400 Subject: [OpenAFS] Deleted db servers still being queried In-Reply-To: <20110823032503.31267cw51wfrks08@bitis.umrk.nl> References: <20110823032503.31267cw51wfrks08@bitis.umrk.nl> Message-ID: <4E530525.2060108@secure-endpoints.com> This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigECDEE9CFFD079D6C3D3D508E Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On 8/22/2011 9:25 PM, Jaap Winius wrote: > Hi folks, >=20 > My AFS network used to have six hosts: thee db servers and three fs > servers. Thinking that I might be able to end some unexplained behavior= , > I decided to consolidate them and have three hosts. >=20 > I did this by creating a new db server on each fs server host and then > deleting the db server that was closest to it. I deleted the old db > servers by applying both the "bos removehost" command to each of the > existing servers,"vos changeaddr" to remove the old IP addresses from > the VLDB, as well as to restart each server after the entries had been > removed. Each client will continue to probe all file and volume location service instances known to them until the client is restarted. There is no mechanism at the present time to make a client forget about a service instance without a restart. Jeffrey Altman --------------enigECDEE9CFFD079D6C3D3D508E Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (MingW32) iQEcBAEBAgAGBQJOUwUnAAoJENxm1CNJffh4FH0IAM0ycdefOMXan28u8qQ+unw4 0F827KmGLrjVdcdTAp8wcwD/c+IO+p50PVpYenhSySM6NPBp7SjKVufuJP4OG4t7 e87Q5BogZIbnSlQrJJknU/AHfM0dy9ZgoTYgkEKFEkPr5FMcibRZbaKd4XaCpaBe 9dMS6st4jzUiyaIr56wi1qQ2Q8u/TO+guPF2AFkfvuL+t9EnMpjCMF8rMVG0hQ0h IyUUM+F9l2qZm+VZyefYwxrp/7rx8wAVIiINYWMo8GPvQqnUv/NAqG9ItkA2BLwo zqFX6XB/OGRmii6063ZfSpVW5QTW0EHINm54u8s1YvAPsvKBJuzrHXAlP6OACyY= =oW7c -----END PGP SIGNATURE----- --------------enigECDEE9CFFD079D6C3D3D508E-- From jwinius@umrk.nl Tue Aug 23 02:57:01 2011 From: jwinius@umrk.nl (Jaap Winius) Date: Tue, 23 Aug 2011 03:57:01 +0200 Subject: [OpenAFS] Deleted db servers still being queried In-Reply-To: <1112946558.12006.1314063085455.JavaMail.root@mailstore01.gopc.net> References: <1112946558.12006.1314063085455.JavaMail.root@mailstore01.gopc.net> Message-ID: <20110823035701.13704y5julodj7cw@bitis.umrk.nl> Quoting Chris Hoy Poy : > if you still have the old IPs in your control, put an iptables (or > some other firewall rule) up to log any traffic to/from the ports. That's how I know where the packets are coming from and where they are going. All of the queries are coming from the current ubik master db server and being sent out to the IP addresses of two of the three non-existent db servers. > Most likely an old client hiding in a closet somewhere But, then I'd see one of the client IP address in the syslog, wouldn't I? Plus, all AFS clients are accounted for and none of them are up and running anyway. Cheers, Jaap From adeason@sinenomine.net Tue Aug 23 05:27:04 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 22 Aug 2011 23:27:04 -0500 Subject: [OpenAFS] Re: Deleted db servers still being queried References: <1112946558.12006.1314063085455.JavaMail.root@mailstore01.gopc.net> <20110823035701.13704y5julodj7cw@bitis.umrk.nl> Message-ID: <20110822232704.6d8cb43b7c241991d8dc4c04@sinenomine.net> On Tue, 23 Aug 2011 03:57:01 +0200 Jaap Winius wrote: > Quoting Chris Hoy Poy : > > > if you still have the old IPs in your control, put an iptables (or > > some other firewall rule) up to log any traffic to/from the ports. > > That's how I know where the packets are coming from and where they are > going. All of the queries are coming from the current ubik master db > server and being sent out to the IP addresses of two of the three > non-existent db servers. What's the source port on these packets from the 'master' dbserver? -- Andrew Deason adeason@sinenomine.net From adeason@sinenomine.net Tue Aug 23 05:33:11 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 22 Aug 2011 23:33:11 -0500 Subject: [OpenAFS] Re: Performance issues References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> Message-ID: <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> On Wed, 17 Aug 2011 23:03:44 +0200 Jaap Winius wrote: > If it's mostly due to the NAT and the fact that the firewalls must be > traversed and the UPD connections kept track of, then I could set up > some tunnels to avoid most of that. But, if it's just the higher > latency and limited bandwidth, then I suppose there's not much to do > about it. What approximate throughput are you seeing in the two cases? I would guess that this has more to do with the bandwidth and latency, but unless there are more dropped or reordered packets or something, I'm not sure either way what we would see from the OpenAFS fileserver point of view besides worse bandwidth/latency. -- Andrew Deason adeason@sinenomine.net From jwinius@umrk.nl Tue Aug 23 16:55:29 2011 From: jwinius@umrk.nl (Jaap Winius) Date: Tue, 23 Aug 2011 17:55:29 +0200 Subject: [OpenAFS] Re: Deleted db servers still being queried In-Reply-To: <20110822232704.6d8cb43b7c241991d8dc4c04@sinenomine.net> References: <1112946558.12006.1314063085455.JavaMail.root@mailstore01.gopc.net> <20110823035701.13704y5julodj7cw@bitis.umrk.nl> <20110822232704.6d8cb43b7c241991d8dc4c04@sinenomine.net> Message-ID: <20110823175529.20168fqaawnujy40@bitis.umrk.nl> Quoting Andrew Deason : > What's the source port on these packets from the 'master' dbserver? The source port is usually 7001 with destination port 7003. However, for less than 30 minutes this morning, I saw packets arriving for destination port 7002 from arbitrary unprivileged port numbers.* In all cases the queries were coming from the ubik master server.** In the mean time, the problem has been solved. It was indeed an AFS client that was looking for the old db servers -- a Windows AFS client that I had overlooked. Incidentally, I tried to remove the old server names from the volume location server preferences list in the Windows client, but could not figure out how to do that. In the end I simply reinstalled it. After that it had the correct db server names and the queries for the old server had stopped. Cheers, Jaap *) Probably when someone attempted to use the Windows machine to access an AFS directory, although it wasn't working at the time because of it's outdated and incorrect configuration. **) But that's probably because the ubik master server is installed on the one physical server at that location, which also functions as the router that connects that local network to the Internet. From chas@cmf.nrl.navy.mil Tue Aug 23 19:50:25 2011 From: chas@cmf.nrl.navy.mil (chas williams - CONTRACTOR) Date: Tue, 23 Aug 2011 14:50:25 -0400 Subject: [OpenAFS] Install OpenAFS on ppc64? In-Reply-To: <1314060971.6543.10.camel@yentj.infowrap.com.tw> References: <1313988591.6154.17.camel@yentj.infowrap.com.tw> <20110822130833.6109e6ec@thirdoffive.cmf.nrl.navy.mil> <1314060971.6543.10.camel@yentj.infowrap.com.tw> Message-ID: <20110823145025.50a629db@thirdoffive.cmf.nrl.navy.mil> do you have /proc/fs/openafs/afs_ioctl? i believe your kernel is new enough that you shouldnt be using the traditional syscall interface. On Tue, 23 Aug 2011 08:56:11 +0800 Ting-jen Yen wrote: > Hello, >=20 > I know that judging from those error code -1 of the debug output, > it looks like the problem is about module. However, I can see > openafs module from lsmod even after the failure. 'dmesg' also shows: >=20 > Cache bypass pathed libafs module init. >=20 > Moreover, as I mentioned, I compiled those RPMs from the SRPM, and > all of those RPMs are ppc64 arch. > (Though I did notice when I used 'file' to check those binary files, > more of them are reported as 32-bit. The module itself is 64-bit.) >=20 > Regards, > Ting-jen >=20 >=20 > =A9=F3 =A4@=A1A2011-08-22 =A9=F3 13:08 -0400=A1Achas williams - CONTRAC= TOR =B4=A3=A8=EC=A1G > > i believe we have one ppc64 machine running openafs. the biggest dea= l > > was to make sure that the binaries were built 64-bit. the 32-bit > > versions of the ucontext syscalls are broken/incomplete. > >=20 > > just based on your errors though it looks like the afs module isnt > > loaded? > >=20 > > On Mon, 22 Aug 2011 12:49:51 +0800 > > Ting-jen Yen wrote: > >=20 > > > Hello, > > >=20 > > > Has anyone successfully installed OpenAFS on RHEL5 machines with > > > ppc64 arch? (Power6 CPU, virtualized by PowerVM). > > > I was using SRPMs from the OpenAFS site. Rpmbuild showed some > > > errors, and I have to specify to build userspace, modules, and > > > authlibs packages, then modify the source, 'flock64' to 'flock' in > > > src/afs/afs.h. This way, I could build the RPMs, and install > > > them. However, even though the installation succeeded, I cannot st= art > > > the client service. > > >=20 > > > If I added options '-debug' and '-verbose' to afsd, it will show > > >=20 > > >=20 > > > afsd: enabling dynamically allocated vcaches > > > afsd: My home cell is 'mycell' > > > ParseCacheInfoFile: Opening cache info file '/usr/vice/etc/cacheinf= o'... > > > ParseCacheInfoFile: Cache info file successfully parsed: > > > cacheMountDir: '/afs' > > > cacheBaseDir: '/usr/vice/cache'=20 > > > cacheBlocks: 100000 > > > afsd: cacheFiles autotuned to 3125=20 > > > afsd: dCacheSize autotuned to 2000 > > > afsd: cacheStatEntries autotuned to 3000 > > > SScall(137, 28, 35)=3D-1 SScall(137, 28, 17)=3D-1 afsd: Forking rx = listener > > > daemon. > > > SScall(137, 28, 48)=3D-1 afsd: Forking rx callback listener. > > > SScall(137, 28, 0)=3D-1 afsd: Forking rxevent daemon. > > > SScall(137, 28, 19)=3D-1 SScall(137, 28, 36)=3D-1 afsd: Error -1 in= basic > > > initialization. > > >=20 > > > Anybody could help me with this? I tried 1.4.7, 1.4.14, 1.5.78, = and > > > 1.6.0-pre1, and all with the same result. Other machines with x86_= 64 > > > work fine. > > >=20 > >=20 >=20 > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info >=20 From dvorkias@umdnj.edu Tue Aug 23 20:20:03 2011 From: dvorkias@umdnj.edu (Dvorkin, Asya) Date: Tue, 23 Aug 2011 15:20:03 -0400 Subject: [OpenAFS] OpenAFS inconsistent behavior - Windows 7 32 bit Message-ID: <5FE0774E-7E40-458A-813F-F247CD02F633@umdnj.edu> Hi everyone, I've been working on getting OpenAFS to work on Windows 7 32 bit and having= some issues. A little bit about the system: Virtual Machine running on Mac (tested on a desktop as well with similar re= sults) Windows 7 32bit fully patched. OpenAFS - 1.5.9906 MIT Kerberos - 1.3.1.0 My problem that it works very inconsistently. It's working (case 1) I rebo= ot and all of a sudden it stops working (case 2). Could it be related to t= he network performance? There is also a third case where it can't even suc= cessfully get initial kerberos credentials (case 3). All logs are below. = Did anyone get OpenAFS work properly on Windows 7 32 bit? Thank you for al= l your help. 1. Everything works.. kerberos authenticates and openafs works as it's supp= osed to. Log file: 15[10044]() 2011-8-23 15:25:49.384 Debug(1): Found principal [user@CO= RE.UMDNJ.EDU] 15[10044]() 2011-8-23 15:25:49.384 Debug(1): Setting default identity= to [user@CORE.UMDNJ.EDU] 15 2011-8-23 15:25:49.384 >> Context end -- 14[10044](KMM) 2011-8-23 15:25:49.290 Info: Initialization of the plug-i= n returned code 0. 14[10044](KMM) 2011-8-23 15:25:49.384 Info: Plug-in running 14 2011-8-23 15:25:49.384 >> Context end -- 16[10048](KMM) 2011-8-23 15:25:49.384 Info: Initialization of the plug-i= n returned code 0. 16[10048](KMM) 2011-8-23 15:25:49.384 Info: Plug-in running 16 2011-8-23 15:25:49.384 >> Context end -- 17 2011-8-23 15:25:54.223 << Context begin -- 17 Description: Refreshing timers 17[9912](NetIDMgr) 2011-8-23 15:25:54.223 Debug(1): Starting with 0 = timers 17[9912](NetIDMgr) 2011-8-23 15:25:54.223 Debug(1): Looking at cred = [krbtgt/CORE.UMDNJ.EDU@CORE.UMDNJ.EDU] 17[9912](NetIDMgr) 2011-8-23 15:25:54.223 Debug(1): Updating identit= y marker timer for user@CORE.UMDNJ.EDU exp(49920-30203-18 0:0:0) 17[9912](NetIDMgr) 2011-8-23 15:25:54.223 Debug(1): Updating identit= y renew timer for user@CORE.UMDNJ.EDU exp(2011-8-23 16:24:43) 17[9912](NetIDMgr) 2011-8-23 15:25:54.223 Debug(1): Skipping credent= ial [krbtgt/CORE.UMDNJ.EDU@CORE.UMDNJ.EDU]. The expiry time is too close t= o the identity expiry. 17[9912](NetIDMgr) 2011-8-23 15:25:54.223 Debug(1): Leaving with 2 t= imers 17 2011-8-23 15:25:54.223 >> Context end -- 13[10036](KMM) 2011-8-23 15:26:05.645 Info: Initialization of the plug-i= n returned code 0. 13[10036](KMM) 2011-8-23 15:26:05.645 Info: Plug-in running 13 2011-8-23 15:26:05.645 >> Context end -- 18 2011-8-23 15:26:05.645 << Context begin -- 18 Description: Importing credentials from Windows 18[10032]() 2011-8-23 15:26:05.645 Debug(1): Begin : khm_krb5_ms2mit.= save_cred=3D1 18[10032]() 2011-8-23 15:26:05.645 Debug(1): Resolving MSLSA 18[10032]() 2011-8-23 15:26:05.645 Debug(1): Getting principal 19 2011-8-23 15:26:05.645 << Context begin -- 19 Description: Renewing credentials 20 2011-8-23 15:26:05.645 << Context begin -- 20 Description: Renewing credentials for user@CORE.UMDNJ.EDU 21 2011-8-23 15:26:05.645 << Context begin -- 21 Description: Renewing Krb5 credentials 21[10032]() 2011-8-23 15:26:06.785 Debug(1): Tickets successfully ren= ewed 21 2011-8-23 15:26:06.785 >> Context end -- 22 2011-8-23 15:26:06.785 << Context begin -- 22 Description: Getting AFS tokens... 22[10036](AfsCred) 2011-8-23 15:26:06.785 Info: AFS Renew Creds :: i= dent 01214838 22[10036](AfsCred) 2011-8-23 15:26:14.301 Info: Getting tokens for c= ell rwjresearch.umdnj.edu with realm using method 0 22[10036]() 2011-8-23 15:26:14.301 Debug(1): Trying Kerberos 5 22[10036]() 2011-8-23 15:26:22.270 Debug(1): Code -1765328370 while g= etting credentials 22[10036]() 2011-8-23 15:26:22.270 Debug(1): Trying Kerberos 4 22[10036]() 2011-8-23 15:26:22.270 Debug(1): Trying to find afs.rwjre= search.umdnj.edu@ 22[10036]() 2011-8-23 15:26:22.270 Debug(1): krb_get_cred returns 76 = (no ticket file) 22[10036](AfsCred) 2011-8-23 15:26:22.270 Info: klog returns code 1 22[10036](AfsCred) 2011-8-23 15:26:22.270 ERROR: Credentials could n= ot be obtained for cell rwjresearch.umdnj.edu. 22 2011-8-23 15:26:22.270 >> Context end -- 20[10048]() 2011-8-23 15:26:06.785 Debug(1): Kerberos 4 is not enable= d for this identity. Skipping 20[10032]() 2011-8-23 15:26:22.302 Debug(1): Getting tickets from cac= he [API:user@CORE.UMDNJ.EDU] 20[10032]() 2011-8-23 15:26:22.302 Debug(1): Found principal [user@CO= RE.UMDNJ.EDU] 20[10032]() 2011-8-23 15:26:22.302 Debug(1): Setting properties for i= dentity (count=3D1) 20[10032]() 2011-8-23 15:26:22.302 Debug(1): Getting tickets from cac= he [MSLSA:] 20 2011-8-23 15:26:22.302 >> Context end -- 19 2011-8-23 15:26:22.302 >> Context end -- 18[10032]() 2011-8-23 15:26:05.645 Debug(1): Received code=3D-17653= 28243 18 2011-8-23 15:26:22.302 >> Context end -- 23 2011-8-23 15:26:27.302 << Context begin -- 23 Description: Refreshing timers 23[9912](NetIDMgr) 2011-8-23 15:26:27.302 Debug(1): Starting with 2 = timers 23[9912](NetIDMgr) 2011-8-23 15:26:27.302 Debug(1): Looking at cred = [krbtgt/CORE.UMDNJ.EDU@CORE.UMDNJ.EDU] 23[9912](NetIDMgr) 2011-8-23 15:26:27.302 Debug(1): Updating identit= y renew timer for user@CORE.UMDNJ.EDU exp(2011-8-23 16:26:6) 23[9912](NetIDMgr) 2011-8-23 15:26:27.302 Debug(1): Skipping credent= ial [krbtgt/CORE.UMDNJ.EDU@CORE.UMDNJ.EDU]. The expiry time is too close t= o the identity expiry. 23[9912](NetIDMgr) 2011-8-23 15:26:27.302 Debug(1): 2. Kerberos and OpenAFS start, but it seems like Kerberos authenticates AFT= ER OpenAFS starts up, thus OpenAFS doesn't work and only if I destroy old /= obtain new credentials, it starts working. Log file: 15[10072]() 2011-8-23 17:08:18.667 Debug(1): Found principal [user@CO= RE.UMDNJ.EDU] 15[10072]() 2011-8-23 17:08:18.667 Debug(1): Setting default identity= to [user@CORE.UMDNJ.EDU] 15 2011-8-23 17:08:18.667 >> Context end -- 14[10072](KMM) 2011-8-23 17:08:18.620 Info: Initialization of the plug-i= n returned code 0. 14[10072](KMM) 2011-8-23 17:08:18.667 Info: Plug-in running 14 2011-8-23 17:08:18.667 >> Context end -- 16[9168](KMM) 2011-8-23 17:08:18.667 Info: Initialization of the plug-i= n returned code 0. 16[9168](KMM) 2011-8-23 17:08:18.667 Info: Plug-in running 16 2011-8-23 17:08:18.667 >> Context end -- 13[9184](KMM) 2011-8-23 17:08:19.894 Info: Initialization of the plug-i= n returned code 0. 13[9184](KMM) 2011-8-23 17:08:19.894 Info: Plug-in running 13 2011-8-23 17:08:19.894 >> Context end -- 17 2011-8-23 17:08:19.894 << Context begin -- 17 Description: Importing credentials from Windows 17[9544]() 2011-8-23 17:08:19.894 Debug(1): Begin : khm_krb5_ms2mit.= save_cred=3D1 17[9544]() 2011-8-23 17:08:19.894 Debug(1): Resolving MSLSA 17[9544]() 2011-8-23 17:08:19.894 Debug(1): Getting principal 18 2011-8-23 17:08:19.894 << Context begin -- 18 Description: Renewing credentials 19 2011-8-23 17:08:19.894 << Context begin -- 19 Description: Renewing credentials for user@CORE.UMDNJ.EDU 20 2011-8-23 17:08:19.894 << Context begin -- 20 Description: Renewing Krb5 credentials 20[9544]() 2011-8-23 17:08:20.109 Debug(1): Tickets successfully ren= ewed 20 2011-8-23 17:08:20.109 >> Context end -- 21 2011-8-23 17:08:20.109 << Context begin -- 21 Description: Getting AFS tokens... 22 2011-8-23 17:08:24.906 << Context begin -- 22 Description: Refreshing timers 22[9396](NetIDMgr) 2011-8-23 17:08:24.906 Debug(1): Starting with 0 = timers 22[9396](NetIDMgr) 2011-8-23 17:08:24.906 Debug(1): Looking at cred = [krbtgt/CORE.UMDNJ.EDU@CORE.UMDNJ.EDU] 22[9396](NetIDMgr) 2011-8-23 17:08:24.906 Debug(1): Updating identit= y marker timer for user@CORE.UMDNJ.EDU exp(49920-30472-18 0:0:0) 22[9396](NetIDMgr) 2011-8-23 17:08:24.906 Debug(1): Updating identit= y renew timer for user@CORE.UMDNJ.EDU exp(2011-8-23 18:7:27) 22[9396](NetIDMgr) 2011-8-23 17:08:24.906 Debug(1): Skipping credent= ial [krbtgt/CORE.UMDNJ.EDU@CORE.UMDNJ.EDU]. The expiry time is too close t= o the identity expiry. 22[9396](NetIDMgr) 2011-8-23 17:08:24.906 Debug(1): Looking at cred = [afs@rwjresearch.umdnj.edu] 22[9396](NetIDMgr) 2011-8-23 17:08:24.906 Debug(1): Updating credent= ial renew timer for afs@rwjresearch.umdnj.edu exp(2011-8-23 22:49:10) 22[9396](NetIDMgr) 2011-8-23 17:08:24.906 Debug(1): Leaving with 3 t= imers 22 2011-8-23 17:08:24.906 >> Context end -- 21[9184](AfsCred) 2011-8-23 17:08:20.109 Info: AFS Renew Creds :: i= dent 01214838 21[9184](AfsCred) 2011-8-23 17:08:27.609 Info: Getting tokens for c= ell rwjresearch.umdnj.edu with realm using method 0 21[9184]() 2011-8-23 17:08:27.625 Debug(1): Trying Kerberos 5 21[9184]() 2011-8-23 17:08:35.344 Debug(1): Trying K5 SetToken 21[9184](AfsCred) 2011-8-23 17:08:42.876 Info: klog returns code 0 21 2011-8-23 17:08:42.876 >> Context end -- 19[9168]() 2011-8-23 17:08:20.109 Debug(1): Kerberos 4 is not enable= d for this identity. Skipping 19[9544]() 2011-8-23 17:08:42.891 Debug(1): Getting tickets from cac= he [API:user@CORE.UMDNJ.EDU] 19[9544]() 2011-8-23 17:08:42.891 Debug(1): Found principal [user@CO= RE.UMDNJ.EDU] 19[9544]() 2011-8-23 17:08:42.891 Debug(1): Setting properties for i= dentity (count=3D1) 19[9544]() 2011-8-23 17:08:42.891 Debug(1): Getting tickets from cac= he [MSLSA:] 19 2011-8-23 17:08:42.891 >> Context end -- 18 2011-8-23 17:08:42.891 >> Context end -- 17[9544]() 2011-8-23 17:08:19.894 Debug(1): Received code=3D-17653= 28243 17 2011-8-23 17:08:42.891 >> Context end -- 23 2011-8-23 17:08:47.907 << Context begin -- 23 Description: Refreshing timers 23[9396](NetIDMgr) 2011-8-23 17:08:47.907 Debug(1): Starting with 3 = timers 23[9396](NetIDMgr) 2011-8-23 17:08:47.907 Debug(1): Looking at cred = [krbtgt/CORE.UMDNJ.EDU@CORE.UMDNJ.EDU] 23[9396](NetIDMgr) 2011-8-23 17:08:47.907 Debug(1): Updating identit= y renew timer for user@CORE.UMDNJ.EDU exp(2011-8-23 18:8:20) 23[9396](NetIDMgr) 2011-8-23 17:08:47.907 Debug(1): Skipping credent= ial [krbtgt/CORE.UMDNJ.EDU@CORE.UMDNJ.EDU]. The expiry time is too close t= o the identity expiry. 23[9396](NetIDMgr) 2011-8-23 17:08:47.907 Debug(1): Looking at cred = [afs@rwjresearch.umdnj.edu] 23[9396](NetIDMgr) 2011-8-23 17:08:47.907 Debug(1): Skipping credent= ial [afs@rwjresearch.umdnj.edu]. The expiry time is too 3. For a different user often times I can't even get Kerberos credentials. = Once I manually add credentials, AFS works properly as well. 15[3264]() 2011-8-23 17:48:51.461 Debug(1): Found principal [diff_us= er@CORE.UMDNJ.EDU] 15[3264]() 2011-8-23 17:48:51.461 Debug(1): Setting default identity= to [diff_user@CORE.UMDNJ.EDU] 15 2011-8-23 17:48:51.461 >> Context end -- 14[3264](KMM) 2011-8-23 17:48:51.414 Info: Initialization of the plug-i= n returned code 0. 14[3264](KMM) 2011-8-23 17:48:51.461 Info: Plug-in running 14 2011-8-23 17:48:51.461 >> Context end -- 16[3268](KMM) 2011-8-23 17:48:51.461 Info: Initialization of the plug-i= n returned code 0. 16[3268](KMM) 2011-8-23 17:48:51.461 Info: Plug-in running 16 2011-8-23 17:48:51.461 >> Context end -- 13[3260](KMM) 2011-8-23 17:48:52.708 Info: Initialization of the plug-i= n returned code 0. 13[3260](KMM) 2011-8-23 17:48:52.708 Info: Plug-in running 13 2011-8-23 17:48:52.708 >> Context end -- 17 2011-8-23 17:48:52.708 << Context begin -- 17 Description: Importing credentials from Windows 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Begin : khm_krb5_ms2mit.= save_cred=3D1 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Resolving MSLSA 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Getting principal 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Unparsing name 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Unparsed name [diff_user= @CORE.UMDNJ.EDU] 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Setting CCache [API:diff= _user@CORE.UMDNJ.EDU] for identity [diff_user@CORE.UMDNJ.EDU] 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Resolving target cache [= API:diff_user@CORE.UMDNJ.EDU] 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Initializing ccache 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Copying credentials 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Received code=3D0 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Setting CCache [API:diff= _user@CORE.UMDNJ.EDU] for identity [diff_user@CORE.UMDNJ.EDU] 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Getting tickets from cac= he [API:diff_user@CORE.UMDNJ.EDU] 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Found principal [diff_us= er@CORE.UMDNJ.EDU] 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Getting tickets from cac= he [MSLSA:] 18 2011-8-23 17:48:52.724 << Context begin -- 18 Description: Obtaining new credentials 18 2011-8-23 17:48:52.739 >> Context end -- 17[3244]() 2011-8-23 17:48:52.708 Debug(1): Found principal [diff_us= er@CORE.UMDNJ.EDU] 17[3396](NetIDMgr) 2011-8-23 17:48:52.724 (None): Obtaining new cred= entials 17 2011-8-23 17:48:52.739 >> Context end -- 19 2011-8-23 17:48:57.724 << Context begin -- 19 Description: Refreshing timers 19[3396](NetIDMgr) 2011-8-23 17:48:57.724 Debug(1): Starting with 0 = timers 19[3396](NetIDMgr) 2011-8-23 17:48:57.724 Debug(1): Leaving with 0 t= imers 19 2011-8-23 17:48:57.724 >> Context end -- 20 2011-8-23 17:49:08.583 << Context begin -- 20 Description: Obtaining new credentials for diff_user@CORE.UMDNJ.E= DU 21 2011-8-23 17:49:08.583 << Context begin -- 21 Description: Obtaining initial Krb5 credentials 21[3244]() 2011-8-23 17:49:08.583 Debug(1): In khm_krb5_kinit 21[3244]() 2011-8-23 17:49:08.583 Debug(1): Setting CCache [API:diff= _user@CORE.UMDNJ.EDU] for identity [diff_user@CORE.UMDNJ.EDU] 21[3244]() 2011-8-23 17:49:08.583 Debug(1): krb5_cc_resolve returns = code 0 21[3244]() 2011-8-23 17:49:18.099 Debug(1): krb5_get_init_creds_pass= word returns code 0 21[3244]() 2011-8-23 17:49:18.099 Debug(1): krb5_cc_initialize retur= ns code 0 21[3244]() 2011-8-23 17:49:18.099 Debug(1): krb5_cc_store_cred retur= ns code 0 21[3244]() 2011-8-23 17:49:18.099 Debug(1): Tickets successfully acq= uired 21[3244]() 2011-8-23 17:49:18.115 Debug(1): Setting CCache [API:diff= _user@CORE.UMDNJ.EDU] for identity [diff_user@CORE.UMDNJ.EDU] 21[3244]() 2011-8-23 17:49:18.115 Debug(1): Getting tickets from cac= he [API:diff_user@CORE.UMDNJ.EDU] 21[3244]() 2011-8-23 17:49:18.115 Debug(1): Found principal [diff_us= er@CORE.UMDNJ.EDU] 21[3244]() 2011-8-23 17:49:18.115 Debug(1): Setting properties for i= dentity (count=3D1) 21[3244]() 2011-8-23 17:49:18.115 Debug(1): Getting tickets from cac= he [MSLSA:] 21[3244]() 2011-8-23 17:49:18.115 Debug(1): Found principal [diff_us= er@CORE.UMDNJ.EDU] 21 2011-8-23 17:49:18.130 >> Context end -- 22 2011-8-23 17:49:18.130 << Context begin -- 22 Description: Getting AFS tokens... 22[3260](AfsCred) 2011-8-23 17:49:18.130 Info: AFS New Creds :: ide= nt 01310EE0 22[3260](AfsCred) 2011-8-23 17:49:18.130 Info: Getting tokens for c= ell rwjresearch.umdnj.edu with realm using method 0 22[3260]() 2011-8-23 17:49:25.646 Debug(1): Trying Kerberos 5 Thank you! Asya From jaltman@secure-endpoints.com Tue Aug 23 21:38:36 2011 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Tue, 23 Aug 2011 16:38:36 -0400 Subject: [OpenAFS] OpenAFS inconsistent behavior - Windows 7 32 bit In-Reply-To: <5FE0774E-7E40-458A-813F-F247CD02F633@umdnj.edu> References: <5FE0774E-7E40-458A-813F-F247CD02F633@umdnj.edu> Message-ID: <4E540FCC.9020802@secure-endpoints.com> This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig4A6BAE227BFDC1C92A4FF9DD Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On 8/23/2011 3:20 PM, Dvorkin, Asya wrote: > Hi everyone, >=20 > I've been working on getting OpenAFS to work on Windows 7 32 bit and ha= ving some issues. A little bit about the system: >=20 > Virtual Machine running on Mac (tested on a desktop as well with simila= r results) > Windows 7 32bit fully patched. > OpenAFS - 1.5.9906 > MIT Kerberos - 1.3.1.0 >=20 > My problem that it works very inconsistently. It's working (case 1) I = reboot and all of a sudden it stops working (case 2). Could it be relate= d to the network performance? There is also a third case where it can't = even successfully get initial kerberos credentials (case 3). All logs ar= e below. Did anyone get OpenAFS work properly on Windows 7 32 bit? Than= k you for all your help. The logs you are quoting are Network Identity Manager logs. Network Identity Manager is an application that runs within the user logon session. OpenAFS will always have started prior to the logon session desktop appears. Kerberos authentication must always occur after OpenAFS starts. If OpenAFS has not started, there is no service with which to register AFS tokens. I suggest you read the OpenAFS for Windows Release Notes that are installed on your Windows system along with OpenAFS. In particular, read the Troubleshooting section. It describes the various logs used by OpenAFS and the tools you can use to assist with debugging. Jeffrey Altman --------------enig4A6BAE227BFDC1C92A4FF9DD Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (MingW32) iQEcBAEBAgAGBQJOVA/OAAoJENxm1CNJffh4jVQH/3yfL+qvnBzmZZHCTBKffPTs s24Pyvjvy0RFBNOGN9d26KnpMjY28E9xxRZ7ZS0n41xnp8023owIM8S4jf/z7O3b sWJcTsMkqI3SHh/BrxXOy4fb+R+dN+BISnL7KUHiMbk7fS2mIcQcJ1rmvigcvifB xQuk4v6sCbJzY00i7zDp2t9LW1CqrEF3gYp4GXQRklEqyVL6Lu7/sApP8p4+e6e1 Uw4DKP/6WOy1KyJHZ4AyTXkAsNFIIKKhU7NTIEkGfvT0TwdEXU1Yxa90b0D+7NlG Y1R5AMZNxfme0r2xND6SF00wdlt0/PECrRTGG3NCiIerMUVqLVCAjQ5WmEOlIJo= =1Ry1 -----END PGP SIGNATURE----- --------------enig4A6BAE227BFDC1C92A4FF9DD-- From jwinius@umrk.nl Wed Aug 24 00:10:04 2011 From: jwinius@umrk.nl (Jaap Winius) Date: Wed, 24 Aug 2011 01:10:04 +0200 Subject: [OpenAFS] Re: Performance issues In-Reply-To: <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> Message-ID: <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> Quoting Andrew Deason : > What approximate throughput are you seeing in the two cases? ... I've been monitoring bandwidth usage with Cacti, which shows that even internal throughput does not normally exceed 3 Mbps. The graphs for the main Internet connections are pretty much the same and the available bandwidth is almost never maxed out. It's probably safe to assume, however, that there are lots of spikes that don't show up in any of these graphs. > ... I would guess that this has more to do with the bandwidth and > latency, but unless there are more dropped or reordered packets or > something, I'm not sure either way what we would see from the OpenAFS > fileserver point of view besides worse bandwidth/latency. After both the fs and db servers were moved from the virtual hosts (and from behind the firewall and NAT) to the bare metal, we saw a marked improvement in local performance. For example, certain applications, particularly the Iceweasel (Firefox) browser, only became usable after the fs server was moved. But as I said, after the changes we've seen no performance improvement for when the user volumes are accessed remotely. (It has been interesting to note that, when dealing with poor file server performance, we found Chrome and Epiphany to work far better than the two other browsers we tested, Iceweasel and Opera.) I'm now leaning towards the idea that latency is the main factor slowing down our remote performance. Latency becomes more important as protocols become more chatty, and I've not been surprised to see AFS described as being such. When combined with the use of GUI desktops (even lightweight Xfce) instead of simple shell environments, that's not going to improve performance either. Originally I was hoping the AFS client cache would do more to compensate, but I guess that was a bit too much to expect from it. Cheers, Jaap From vittorlima@gmail.com Wed Aug 24 00:52:57 2011 From: vittorlima@gmail.com (vitor lima) Date: Tue, 23 Aug 2011 20:52:57 -0300 Subject: [OpenAFS] Home not found Message-ID: --0016e64983b21d2b0704ab34e5f4 Content-Type: text/plain; charset=ISO-8859-1 Hi! I have a problem with my OpenAFS server: When a user do the "log in", the home directory is not found and the user logs using the root (the / ). What can be this? Somebody can help me? Thank you! -- Vitor --0016e64983b21d2b0704ab34e5f4 Content-Type: text/html; charset=ISO-8859-1 Hi!

I have a problem with my OpenAFS server:

When a user do the "log in", the home directory is not found and the user logs using the root (the / ).

What can be this?
Somebody can help me?


Thank you!


--
Vitor

--0016e64983b21d2b0704ab34e5f4-- From vittorlima@gmail.com Wed Aug 24 01:02:42 2011 From: vittorlima@gmail.com (vitor lima) Date: Tue, 23 Aug 2011 21:02:42 -0300 Subject: [OpenAFS] Re: Home not found In-Reply-To: References: Message-ID: --0016e64c1af6050c2604ab350871 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable My OpenAFS version is the 1.4.14.1 and I use the Heimdal Kerberos 1.2.1 (kerberos 5 implementation). 2011/8/23 vitor lima > Hi! > > I have a problem with my OpenAFS server: > > When a user do the "log in", the home directory is not found and the user > logs using the root (the / ). > > What can be this? > Somebody can help me? > > > Thank you! > > > -- > Vitor > > --=20 Vitor Lima Bacharelando em Ci=EAncia da Computa=E7=E3o Instituto de Matem=E1tica e Estat=EDstica (IME) da Universidade de S=E3o Paulo (USP) --0016e64c1af6050c2604ab350871 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable My OpenAFS version is the 1.4.14.1

and I use the Heimdal Kerberos 1.= 2.1 (kerberos 5 implementation).




2011/8/23 vitor lima <vittorlima@gmail.com>
Hi!

I have= a problem with my OpenAFS server:

When a user do the "log in&q= uot;, the home directory is not found and the user logs using the root (the= / ).

What can be this?
Somebody can help me?


Thank you!


--
= Vitor




--
Vitor Lima
Ba= charelando em Ci=EAncia da Computa=E7=E3o
Instituto de Matem=E1tica e Es= tat=EDstica (IME)
da Universidade de S=E3o Paulo (USP)
--0016e64c1af6050c2604ab350871-- From allbery.b@gmail.com Wed Aug 24 03:08:05 2011 From: allbery.b@gmail.com (Brandon Allbery) Date: Tue, 23 Aug 2011 22:08:05 -0400 Subject: [OpenAFS] Re: Performance issues In-Reply-To: <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> Message-ID: --000e0cd156f45eb63d04ab36c87d Content-Type: text/plain; charset=UTF-8 On Tue, Aug 23, 2011 at 19:10, Jaap Winius wrote: > Originally I was hoping the AFS client cache would do more to compensate, > but I guess that was a bit too much to expect from it. First: is the cache big enough? The working set of even a lightweight desktop manager is pretty big. Second: AFS is really designed for mostly read-only data. Desktop environments are constantly updating the files in ~ so there will be a lot of writes. Third: especially with Iceweasel/Firefox, you'll find that it helps a lot to symlink the cache into /var/tmp or other fast local storage. (Desktop environments have caches as well, which again should be symlinked to local storage.) This still won't help with KDE/Gnome-like updating of application and system settings when you look at them funny, but it will help a lot. -- brandon s allbery allbery.b@gmail.com wandering unix systems administrator (available) (412) 475-9364 vm/sms --000e0cd156f45eb63d04ab36c87d Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
On Tue, Aug 23, 2011 at 19:10, Jaap Winius <jwinius@umrk.nl> wr= ote:
Originally I was hoping the AFS client cache would do more to compensate, b= ut I guess that was a bit too much to expect from it.
First: =C2=A0is the cache big enough? =C2=A0The working set of even a ligh= tweight desktop manager is pretty big.

Second: =C2=A0AFS is really designed for mostly read-only da= ta. =C2=A0Desktop environments are constantly updating the files in ~ so th= ere will be a lot of writes.

Third: =C2=A0especial= ly with Iceweasel/Firefox, you'll find that it helps a lot to symlink t= he cache into /var/tmp or other fast local storage. =C2=A0(Desktop environm= ents have caches as well, which again should be symlinked to local storage.= ) =C2=A0This still won't help with KDE/Gnome-like updating of applicati= on and system settings when you look at them funny, but it will help a lot.=

--
brandon s allbery =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0allbery.b@gmail.com
wandering unix systems administrator (avai= lable) =C2=A0 =C2=A0 (412) 475-9364 vm/sms

--000e0cd156f45eb63d04ab36c87d-- From jwinius@umrk.nl Wed Aug 24 04:03:20 2011 From: jwinius@umrk.nl (Jaap Winius) Date: Wed, 24 Aug 2011 05:03:20 +0200 Subject: [OpenAFS] Re: Performance issues In-Reply-To: References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> Message-ID: <20110824050320.15532dpqf4whury8@bitis.umrk.nl> Quoting Brandon Allbery : > First: is the cache big enough? The working set of even a lightweight > desktop manager is pretty big. It's currently set to 128 MB on all of the workstations. I could increase that, but by how much? I'm under the impression that, not far beyond this point, the returns tend to diminish. > Second: AFS is really designed for mostly read-only data. Desktop > environments are constantly updating the files in ~ so there will be a lot > of writes. I'm familiar with the description. But they put so much effort into supporting user volumes (client caches and all) that I guess I thought it might do better in this case than I had a right to expect. > Third: especially with Iceweasel/Firefox, you'll find that it helps a lot > to symlink the cache into /var/tmp or other fast local storage. ... Actually, each of our physical servers also runs a Squid proxy cache, so we decided to simply disable all of the browser caches. Nevertheless, when someone's user volume is remote, Iceweasel still runs too slow. > ... (Desktop environments have caches as well, which again should be > symlinked to local storage.) ... Oh, that sounds like it will be worth a little extra performance! > ... This still won't help with KDE/Gnome-like updating of application > and system settings when you look at them funny, but it will help a lot. When we were first setting up this system, we also tested the Gnome and KDE desktops just for fun. Gnome didn't do too badly with a remote user volume, but as I expected, KDE (my own favorite desktop) was just awful! Cheers, Jaap From allbery.b@gmail.com Wed Aug 24 04:26:18 2011 From: allbery.b@gmail.com (Brandon Allbery) Date: Tue, 23 Aug 2011 23:26:18 -0400 Subject: [OpenAFS] Re: Performance issues In-Reply-To: <20110824050320.15532dpqf4whury8@bitis.umrk.nl> References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> <20110824050320.15532dpqf4whury8@bitis.umrk.nl> Message-ID: --001517503b961c4b2504ab37e068 Content-Type: text/plain; charset=UTF-8 On Tue, Aug 23, 2011 at 23:03, Jaap Winius wrote: > Quoting Brandon Allbery : > > First: is the cache big enough? The working set of even a lightweight >> desktop manager is pretty big. >> > > It's currently set to 128 MB on all of the workstations. I could increase > that, but by how much? I'm under the impression that, not far beyond this > point, the returns tend to diminish. That was the old conventional wisdom, but these days people find 1GB caches useful (especially since the cache was reorganized to not be flat). Meanwhile, 128MB will thrash a lot. It *does* depend on usage to some extent, but for home directories you should almost always have a fairly large cache. Third: especially with Iceweasel/Firefox, you'll find that it helps a lot >> to symlink the cache into /var/tmp or other fast local storage. ... >> > > Actually, each of our physical servers also runs a Squid proxy cache, so we > decided to simply disable all of the browser caches. Nevertheless, when > someone's user volume is remote, Iceweasel still runs too slow. > It's still using the cache; it's just not persistent now. Last time I checked, any time Firefox loads images or media for a page, they are loaded into the cache and (dis)played from there. There used to be a memory cache once, but not at this (URL) level any more. You really want this to be local, and maybe even ramdisk. -- brandon s allbery allbery.b@gmail.com wandering unix systems administrator (available) (412) 475-9364 vm/sms --001517503b961c4b2504ab37e068 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
On Tue, Aug 23, 2011 at 23:03, Jaap Winius <jwinius@umrk.nl> wr= ote:
Quoting Brandon Allbery <allbery.b@gmail.com>:

First: =C2=A0is the cache big enough? =C2=A0The working set of even a light= weight
desktop manager is pretty big.

It's currently set to 128 MB on all of the workstations. I could increa= se that, but by how much? I'm under the impression that, not far beyond= this point, the returns tend to diminish.

That was the old conventional wisdom, but these days people find 1GB c= aches useful (especially since the cache was reorganized to not be flat). = =C2=A0Meanwhile, 128MB will thrash a lot. =C2=A0It *does* depend on usage t= o some extent, but for home directories you should almost always have a fai= rly large cache.

Third: =C2=A0especially with Iceweasel/Firefox, you'll find that it hel= ps a lot
to symlink the cache into /var/tmp or other fast local storage. ...

Actually, each of our physical servers also runs a Squid proxy cache, so we= decided to simply disable all of the browser caches. Nevertheless, when so= meone's user volume is remote, Iceweasel still runs too slow.

It's still using the cache; it's just not persi= stent now. =C2=A0Last time I checked, any time Firefox loads images or medi= a for a page, they are loaded into the cache and (dis)played from there. = =C2=A0There used to be a memory cache once, but not at this (URL) level any= more. =C2=A0You really want this to be local, and maybe even ramdisk.

--
brandon s allbery =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0allbery.b@gmail.com
wandering unix systems administrato= r (available) =C2=A0 =C2=A0 (412) 475-9364 vm/sms

--001517503b961c4b2504ab37e068-- From yentj@infowrap.com.tw Wed Aug 24 08:23:00 2011 From: yentj@infowrap.com.tw (Ting-jen Yen) Date: Wed, 24 Aug 2011 15:23:00 +0800 Subject: [OpenAFS] Install OpenAFS on ppc64? In-Reply-To: <20110823145025.50a629db@thirdoffive.cmf.nrl.navy.mil> References: <1313988591.6154.17.camel@yentj.infowrap.com.tw> <20110822130833.6109e6ec@thirdoffive.cmf.nrl.navy.mil> <1314060971.6543.10.camel@yentj.infowrap.com.tw> <20110823145025.50a629db@thirdoffive.cmf.nrl.navy.mil> Message-ID: <4E54A6D4.2050108@infowrap.com.tw> Hello, Yes, I can see the file /proc/fs/openafs/afs_ioctl. For some reason, the system became unstable after I browsed this directory. (Could be coincident. I don't know.) -- Ting-jen On Wed, 23 Aug 2011 14:50:25 -0400 chas williams - CONTRACTOR wrote: > do you have /proc/fs/openafs/afs_ioctl? i believe your kernel is new > enough that you shouldnt be using the traditional syscall interface. > > On Tue, 23 Aug 2011 08:56:11 +0800 > Ting-jen Yen wrote: > >> Hello, >> >> I know that judging from those error code -1 of the debug output, >> it looks like the problem is about module. However, I can see >> openafs module from lsmod even after the failure. 'dmesg' also shows: >> >> Cache bypass pathed libafs module init. >> >> Moreover, as I mentioned, I compiled those RPMs from the SRPM, and >> all of those RPMs are ppc64 arch. >> (Though I did notice when I used 'file' to check those binary files, >> more of them are reported as 32-bit. The module itself is 64-bit.) >> >> Regards, >> Ting-jen >> >> >> =A9=F3 =A4@=A1A2011-08-22 =A9=F3 13:08 -0400=A1Achas williams - CONTRA= CTOR =B4=A3=A8=EC=A1G >>> i believe we have one ppc64 machine running openafs. the biggest dea= l >>> was to make sure that the binaries were built 64-bit. the 32-bit >>> versions of the ucontext syscalls are broken/incomplete. >>> >>> just based on your errors though it looks like the afs module isnt >>> loaded? >>> >>> On Mon, 22 Aug 2011 12:49:51 +0800 >>> Ting-jen Yen wrote: >>> >>>> Hello, >>>> >>>> Has anyone successfully installed OpenAFS on RHEL5 machines with >>>> ppc64 arch? (Power6 CPU, virtualized by PowerVM). >>>> I was using SRPMs from the OpenAFS site. Rpmbuild showed some >>>> errors, and I have to specify to build userspace, modules, and >>>> authlibs packages, then modify the source, 'flock64' to 'flock' in >>>> src/afs/afs.h. This way, I could build the RPMs, and install >>>> them. However, even though the installation succeeded, I cannot sta= rt >>>> the client service. >>>> >>>> If I added options '-debug' and '-verbose' to afsd, it will show >>>> >>>> >>>> afsd: enabling dynamically allocated vcaches >>>> afsd: My home cell is 'mycell' >>>> ParseCacheInfoFile: Opening cache info file '/usr/vice/etc/cacheinfo= '... >>>> ParseCacheInfoFile: Cache info file successfully parsed: >>>> cacheMountDir: '/afs' >>>> cacheBaseDir: '/usr/vice/cache' >>>> cacheBlocks: 100000 >>>> afsd: cacheFiles autotuned to 3125 >>>> afsd: dCacheSize autotuned to 2000 >>>> afsd: cacheStatEntries autotuned to 3000 >>>> SScall(137, 28, 35)=3D-1 SScall(137, 28, 17)=3D-1 afsd: Forking rx l= istener >>>> daemon. >>>> SScall(137, 28, 48)=3D-1 afsd: Forking rx callback listener. >>>> SScall(137, 28, 0)=3D-1 afsd: Forking rxevent daemon. >>>> SScall(137, 28, 19)=3D-1 SScall(137, 28, 36)=3D-1 afsd: Error -1 in = basic >>>> initialization. >>>> >>>> Anybody could help me with this? I tried 1.4.7, 1.4.14, 1.5.78, = and >>>> 1.6.0-pre1, and all with the same result. Other machines with x86_6= 4 >>>> work fine. >>>> >>> >> >> _______________________________________________ >> OpenAFS-info mailing list >> OpenAFS-info@openafs.org >> https://lists.openafs.org/mailman/listinfo/openafs-info >> > > From dvorkias@umdnj.edu Wed Aug 24 13:55:13 2011 From: dvorkias@umdnj.edu (Dvorkin, Asya) Date: Wed, 24 Aug 2011 08:55:13 -0400 Subject: [OpenAFS] OpenAFS inconsistent behavior - Windows 7 32 bit In-Reply-To: <4E540FCC.9020802@secure-endpoints.com> References: <5FE0774E-7E40-458A-813F-F247CD02F633@umdnj.edu> <4E540FCC.9020802@secure-endpoints.com> Message-ID: <404F04F8-D2AF-452F-8871-C8C8C249A189@umdnj.edu> On Aug 23, 2011, at 4:38 PM, Jeffrey Altman wrote: > On 8/23/2011 3:20 PM, Dvorkin, Asya wrote: >> Hi everyone, >>=20 >> I've been working on getting OpenAFS to work on Windows 7 32 bit and hav= ing some issues. A little bit about the system: >>=20 >> Virtual Machine running on Mac (tested on a desktop as well with similar= results) >> Windows 7 32bit fully patched. >> OpenAFS - 1.5.9906 >> MIT Kerberos - 1.3.1.0 >>=20 > Identity Manager is an application that runs within the user logon > session. OpenAFS will always have started prior to the logon session > desktop appears. >=20 > Kerberos authentication must always occur after OpenAFS starts. If > OpenAFS has not started, there is no service with which to register AFS > tokens. >=20 > I suggest you read the OpenAFS for Windows Release Notes that are > installed on your Windows system along with OpenAFS. In particular, > read the Troubleshooting section. It describes the various logs used > by OpenAFS and the tools you can use to assist with debugging. >=20 > Jeffrey Altman >=20 Hi Jeffrey, Thank you. I will look into Release Notes. Has anyone gotten OpenAFS working on Windows 7 32 bit? If so, using which= versions of MIT Kerberos and OpenAFS? Thank you! Asya From jwinius@umrk.nl Wed Aug 24 14:17:34 2011 From: jwinius@umrk.nl (Jaap Winius) Date: Wed, 24 Aug 2011 15:17:34 +0200 Subject: [OpenAFS] Re: Performance issues In-Reply-To: References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> <20110824050320.15532dpqf4whury8@bitis.umrk.nl> Message-ID: <20110824151734.2108154adgcyaxgg@bitis.umrk.nl> Quoting Brandon Allbery : >> It's currently set to 128 MB on all of the workstations. I could increase >> that, but by how much? I'm under the impression that, not far beyond this >> point, the returns tend to diminish. > > That was the old conventional wisdom, but these days people find 1GB caches > useful (especially since the cache was reorganized to not be flat). ... But, wouldn't that cause trouble in the mornings when those workstations are turned on again and a number of them try to re-synchronize their 1GB caches across my relatively puny 6Mbps WAN links? Cheers, Jaap From ktdreyer@ktdreyer.com Wed Aug 24 14:20:02 2011 From: ktdreyer@ktdreyer.com (Ken Dreyer) Date: Wed, 24 Aug 2011 07:20:02 -0600 Subject: [OpenAFS] OpenAFS inconsistent behavior - Windows 7 32 bit In-Reply-To: <404F04F8-D2AF-452F-8871-C8C8C249A189@umdnj.edu> References: <5FE0774E-7E40-458A-813F-F247CD02F633@umdnj.edu> <4E540FCC.9020802@secure-endpoints.com> <404F04F8-D2AF-452F-8871-C8C8C249A189@umdnj.edu> Message-ID: On Wed, Aug 24, 2011 at 6:55 AM, Dvorkin, Asya wrote: > > Has anyone gotten OpenAFS =C2=A0working on Windows 7 32 bit? =C2=A0If so,= using which versions of MIT Kerberos and OpenAFS? Yes, at least several dozen clients on Win 7. KFW 3.2.2: http://web.mit.edu/kerberos/kfw-3.2/kfw-3.2.2.html OpenAFS 1.5.7x or above. Definitely recommend the latest version, found here: http://openafs.org/windows.html I don't run Windows myself as my main OS, but I have a Win 7 VM where AFS works fine. - Ken From mansaxel@besserwisser.org Wed Aug 24 14:28:31 2011 From: mansaxel@besserwisser.org (=?utf-8?B?TcOlbnM=?= Nilsson) Date: Wed, 24 Aug 2011 15:28:31 +0200 Subject: [OpenAFS] Re: Performance issues In-Reply-To: <20110824151734.2108154adgcyaxgg@bitis.umrk.nl> References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> <20110824050320.15532dpqf4whury8@bitis.umrk.nl> <20110824151734.2108154adgcyaxgg@bitis.umrk.nl> Message-ID: <20110824132831.GK11906@besserwisser.org> --wIc/V6YLA2QdyfT4 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Subject: Re: [OpenAFS] Re: Performance issues Date: Wed, Aug 24, 2011 at 03= :17:34PM +0200 Quoting Jaap Winius (jwinius@umrk.nl): =20 > But, wouldn't that cause trouble in the mornings when those > workstations are turned on again and a number of them try to > re-synchronize their 1GB caches across my relatively puny 6Mbps WAN > links? A cache refresh is almost always smaller than a cache miss. Being on disk, the cache persists over reboots. My mulituser machines all have at least 2G cache, some more. 128MB is painfully small; you really should up it considerably.=20 --=20 M=C3=A5ns Nilsson primary/secondary/besserwisser/machina MN-1334-RIPE +46 705 989668 While my BRAINPAN is being refused service in BURGER KING, Jesuit priests are DATING CAREER DIPLOMATS!! --wIc/V6YLA2QdyfT4 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) iEYEARECAAYFAk5U/H8ACgkQ02/pMZDM1cVwngCdGiRvKo9Y28apI1UHGnv8/PmV PLYAn103KNBdoNuJ9duLj77WT9AU2Kw0 =NRo5 -----END PGP SIGNATURE----- --wIc/V6YLA2QdyfT4-- From sxw@inf.ed.ac.uk Wed Aug 24 14:29:17 2011 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Wed, 24 Aug 2011 14:29:17 +0100 Subject: [OpenAFS] Re: Performance issues In-Reply-To: <20110824151734.2108154adgcyaxgg@bitis.umrk.nl> References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> <20110824050320.15532dpqf4whury8@bitis.umrk.nl> <20110824151734.2108154adgcyaxgg@bitis.umrk.nl> Message-ID: <2D439495-EF6B-4701-83B1-2E9CF6465F15@inf.ed.ac.uk> On 24 Aug 2011, at 14:17, Jaap Winius wrote: > But, wouldn't that cause trouble in the mornings when those = workstations are turned on again and a number of them try to = re-synchronize their 1GB caches across my relatively puny 6Mbps WAN = links? Caches aren't revalidated, or resyncronised, upon startup. Instead, when = an application tries to access a file, the cache manager sees that it no = longer has a valid callback, and issues a status request to the = fileserver. If that status request indicates that the cached copy is = still valid, then the cached copy will be used, otherwise the bits of = the file requested by the application will be refetched from the = fileserver. S. From l.schimmer@cgv.tugraz.at Wed Aug 24 14:32:28 2011 From: l.schimmer@cgv.tugraz.at (Lars Schimmer) Date: Wed, 24 Aug 2011 15:32:28 +0200 Subject: [OpenAFS] OpenAFS inconsistent behavior - Windows 7 32 bit In-Reply-To: <404F04F8-D2AF-452F-8871-C8C8C249A189@umdnj.edu> References: <5FE0774E-7E40-458A-813F-F247CD02F633@umdnj.edu> <4E540FCC.9020802@secure-endpoints.com> <404F04F8-D2AF-452F-8871-C8C8C249A189@umdnj.edu> Message-ID: <4E54FD6C.6000005@cgv.tugraz.at> On 2011-08-24 14:55, Dvorkin, Asya wrote: > Hi Jeffrey, > > Thank you. I will look into Release Notes. > > Has anyone gotten OpenAFS working on Windows 7 32 bit? If so, using w= hich versions of MIT Kerberos and OpenAFS? I do run OpoenAFS on windows 7 32bit for some years now on my laptop. I guess from 1.5.7x on up to latest 1.6.0 pre7 and the MIT kerberos=20 network ID manager 1.3.1.0. So far I could login and obtain tokens on login (if the win 7 bug does=20 not hit me). > Thank you! > Asya MfG, Lars Schimmer --=20 ------------------------------------------------------------- TU Graz, Institut f=FCr ComputerGraphik & WissensVisualisierung Tel: +43 316 873-5405 E-Mail: l.schimmer@cgv.tugraz.at Fax: +43 316 873-5402 PGP-Key-ID: 0x4A9B1723 From jaltman@secure-endpoints.com Wed Aug 24 14:34:25 2011 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Wed, 24 Aug 2011 09:34:25 -0400 Subject: [OpenAFS] Re: Performance issues In-Reply-To: <20110824151734.2108154adgcyaxgg@bitis.umrk.nl> References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> <20110824050320.15532dpqf4whury8@bitis.umrk.nl> <20110824151734.2108154adgcyaxgg@bitis.umrk.nl> Message-ID: <4E54FDE1.9030803@secure-endpoints.com> This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigAA8B2A3937E74A79D72866CA Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On 8/24/2011 9:17 AM, Jaap Winius wrote: > But, wouldn't that cause trouble in the mornings when those workstation= s > are turned on again and a number of them try to re-synchronize their 1G= B > caches across my relatively puny 6Mbps WAN links? Caches are not actively sync'd. Caches are passive data stores that are used to avoid copying data across the network when the data is known to already exist locally. If the required data is not present in the cache, then it is read from the file server. If the cache is too small, more data is read from the file server than would otherwise be necessary.= Cache coherency is maintained in AFS via the use of status information which has an expiration time and a callback protocol that is used by the file server to notify the client when the unexpired status information has been altered. AFS does not read and cache whole files. It reads file chunks which are tracked by File ID, File Offset, and Data Version. Data is used from the cache when the File ID, File Offset and Data Version required to satisfy an application request match. Otherwise, the data is read from the file server. For read/write data, if the cache is too small, the cache manager is required to flush data to the file server sooner than it would prefer. Since many files used today are in the GB range, it is not unusual to have caches sizes of 10GB to 20GB. The local disk is cheap; network bandwidth is not. Jeffrey Altman --------------enigAA8B2A3937E74A79D72866CA Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (MingW32) iQEcBAEBAgAGBQJOVP3jAAoJENxm1CNJffh4M2cH/12q57mSmmTXZJAWPkP1Gpfk hfPXQKbfm88tbBzkSLxh5S4rTnfW/EQOUPUCbRUhD/wDJwK/qjQfQB4hcMVC56Ei M3H9nainnmS6TNAH614rG4bP7XIDUZo8GAj4emx4UR9sePvHnznM5Uxb2FMsBbYx Gpd+gl04M4ds5CCBpd95kvx7eGrSygXIiPiCxGGOEtqhFBh9V7w7e6maKZaVfS/k t2fL1bUM+sf2Cl2AMftiKxzwDefLslJTYu8sKkJ+TS2s+4x/hC1auHgHyIF3NaBi WeXRZSA/h5GffToAbyWs+DL/dBdZ/zZ1OIFJaIBk2gQUJFKZGPrmXHtMz82KGRI= =sYM8 -----END PGP SIGNATURE----- --------------enigAA8B2A3937E74A79D72866CA-- From dvorkias@umdnj.edu Wed Aug 24 14:36:56 2011 From: dvorkias@umdnj.edu (Dvorkin, Asya) Date: Wed, 24 Aug 2011 09:36:56 -0400 Subject: [OpenAFS] OpenAFS inconsistent behavior - Windows 7 32 bit In-Reply-To: <4E54FD6C.6000005@cgv.tugraz.at> References: <5FE0774E-7E40-458A-813F-F247CD02F633@umdnj.edu> <4E540FCC.9020802@secure-endpoints.com> <404F04F8-D2AF-452F-8871-C8C8C249A189@umdnj.edu> <4E54FD6C.6000005@cgv.tugraz.at> Message-ID: <1039272C-81AD-460F-B596-522BAA1AC0FC@umdnj.edu> On Aug 24, 2011, at 9:32 AM, Lars Schimmer wrote: > On 2011-08-24 14:55, Dvorkin, Asya wrote: >=20 >> Hi Jeffrey, >>=20 >> Thank you. I will look into Release Notes. >>=20 >> Has anyone gotten OpenAFS working on Windows 7 32 bit? If so, using wh= ich versions of MIT Kerberos and OpenAFS? >=20 > I do run OpoenAFS on windows 7 32bit for some years now on my laptop. > I guess from 1.5.7x on up to latest 1.6.0 pre7 and the MIT kerberos=20 > network ID manager 1.3.1.0. > So far I could login and obtain tokens on login (if the win 7 bug does=20 > not hit me). Thank you Ken and Lars for providing me with all the information. It's rea= lly appreciated. Have a good day, Asya= From dcolson5@uncc.edu Tue Aug 16 21:50:51 2011 From: dcolson5@uncc.edu (Colson, Daniel) Date: Tue, 16 Aug 2011 16:50:51 -0400 Subject: [OpenAFS] 64-bit mod_waklog / filedrawers on afs References: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> Message-ID: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBE2@EXEVS04.its.uncc.edu> This is a multi-part message in MIME format. ------_=_NextPart_001_01CC5C56.DBA2EA2D Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Thanks. Any help is appreciated. Here's the error output from "make". make[1]: Entering directory = `/build/filedrawers/libcgi' gcc -Wall -Wstrict-prototypes -g -fPIC -I/usr/include/apr-1 -c -o = cgi.o cgi.c cgi.c: In function =91post_getline=92: cgi.c:91: warning: function declaration isn=92t a prototype cgi.c: In function =91cgi_init=92: cgi.c:148: warning: function declaration isn=92t a prototype cgi.c: In function =91mp_get_file=92: cgi.c:361: warning: function declaration isn=92t a prototype cgi.c:364: warning: function declaration isn=92t a prototype cgi.c:375: warning: function declaration isn=92t a prototype cgi.c:378: warning: function declaration isn=92t a prototype cgi.c:385: warning: function declaration isn=92t a prototype cgi.c:388: warning: function declaration isn=92t a prototype cgi.c:400: warning: function declaration isn=92t a prototype cgi.c:420: warning: function declaration isn=92t a prototype cgi.c:424: warning: function declaration isn=92t a prototype cgi.c: In function =91cgi_multipart=92: cgi.c:539: warning: function declaration isn=92t a prototype cgi.c:614: warning: function declaration isn=92t a prototype building normal libcgi.a make[1]: Leaving directory = `/build/filedrawers/libcgi' ( cd upload; make ) make[1]: Entering directory = `/build/filedrawers/upload' gcc -Wall -Wmissing-prototypes -g -O2 -I../libcgi -I/usr/include/mysql = -g -pipe -Wp,-D_FORTIFY_SOURCE=3D2 -fexceptions -fstack-protector = --param=3Dssp-buffer-size=3D4 -m64 -D_GNU_SOURCE = -D_FILE_OFFSET_BITS=3D64 -D_LARGEFILE_SOURCE -fno-strict-aliasing = -fwrapv -Wall -Wmissing-prototypes \ -D_URL_ERROR_PAGE=3D\"https://?error=3Dtrue\" \ = -D_FILEDRAWERS_CONFIG_PATH=3D\"/var/www/filedrawers/lib/filedrawers.conf\= " \ -c tpupload.c tpupload.c:18:19: error: mysql.h: No such file or directory tpupload.c:34: error: expected =91=3D=92, =91,=92, =91;=92, =91asm=92 or = =91__attribute__=92 before =91*=92 token tpupload.c:35: error: expected =91=3D=92, =91,=92, =91;=92, =91asm=92 or = =91__attribute__=92 before =91*=92 token tpupload.c: In function =91upload_progress_db_insert=92: tpupload.c:103: error: =91MYSQL_BIND=92 undeclared (first use in this = function) tpupload.c:103: error: (Each undeclared identifier is reported only once tpupload.c:103: error: for each function it appears in.) tpupload.c:103: error: expected =91;=92 before =91bind=92 tpupload.c:104: error: =91my_bool=92 undeclared (first use in this = function) tpupload.c:104: error: expected =91;=92 before =91is_null=92 tpupload.c:110: error: =91stmt=92 undeclared (first use in this = function) tpupload.c:111: warning: implicit declaration of function = =91mysql_stmt_init=92 tpupload.c:111: error: =91mysql=92 undeclared (first use in this = function) tpupload.c:112: warning: implicit declaration of function = =91mysql_error=92 tpupload.c:112: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c:123: warning: implicit declaration of function = =91mysql_stmt_prepare=92 tpupload.c:124: warning: implicit declaration of function = =91mysql_stmt_error=92 tpupload.c:124: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c:130: error: =91bind=92 undeclared (first use in this = function) tpupload.c:130: warning: left-hand operand of comma expression has no = effect tpupload.c:132: error: =91MYSQL_TYPE_STRING=92 undeclared (first use in = this function) tpupload.c:132: error: expected expression before =91)=92 token tpupload.c:132: error: =91is_null=92 undeclared (first use in this = function) tpupload.c:133: error: expected expression before =91)=92 token tpupload.c:134: error: =91MYSQL_TYPE_LONGLONG=92 undeclared (first use = in this function) tpupload.c:134: error: expected expression before =91)=92 token tpupload.c:135: error: expected expression before =91)=92 token tpupload.c:136: error: =91MYSQL_TYPE_TINY=92 undeclared (first use in = this function) tpupload.c:136: error: expected expression before =91)=92 token tpupload.c:138: warning: implicit declaration of function = =91mysql_stmt_bind_param=92 tpupload.c:139: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c:148: warning: implicit declaration of function = =91mysql_stmt_execute=92 tpupload.c:149: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c: In function =91upload_progress_db_update=92: tpupload.c:160: error: =91MYSQL_BIND=92 undeclared (first use in this = function) tpupload.c:160: error: expected =91;=92 before =91bind=92 tpupload.c:161: error: =91my_bool=92 undeclared (first use in this = function) tpupload.c:161: error: expected =91;=92 before =91is_null=92 tpupload.c:166: error: =91stmt=92 undeclared (first use in this = function) tpupload.c:167: error: =91mysql=92 undeclared (first use in this = function) tpupload.c:168: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c:180: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c:186: error: =91bind=92 undeclared (first use in this = function) tpupload.c:186: warning: left-hand operand of comma expression has no = effect tpupload.c:188: error: =91MYSQL_TYPE_STRING=92 undeclared (first use in = this function) tpupload.c:188: error: expected expression before =91)=92 token tpupload.c:188: error: =91is_null=92 undeclared (first use in this = function) tpupload.c:189: error: =91MYSQL_TYPE_LONGLONG=92 undeclared (first use = in this function) tpupload.c:189: error: expected expression before =91)=92 token tpupload.c:190: error: =91MYSQL_TYPE_TINY=92 undeclared (first use in = this function) tpupload.c:190: error: expected expression before =91)=92 token tpupload.c:191: error: expected expression before =91)=92 token tpupload.c:194: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c:203: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c: In function =91upload_init=92: tpupload.c:317: error: =91mysql=92 undeclared (first use in this = function) tpupload.c:322: error: =91stmt=92 undeclared (first use in this = function) tpupload.c:323: warning: implicit declaration of function = =91mysql_stmt_close=92 tpupload.c:324: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c: In function =91upload_progress=92: tpupload.c:347: error: =91mysql=92 undeclared (first use in this = function) tpupload.c: In function =91upload_complete=92: tpupload.c:373: error: =91mysql=92 undeclared (first use in this = function) tpupload.c: In function =91main=92: tpupload.c:528: error: =91mysql=92 undeclared (first use in this = function) tpupload.c:528: warning: implicit declaration of function = =91mysql_init=92 tpupload.c:530: warning: implicit declaration of function = =91mysql_real_connect=92 tpupload.c:532: warning: format =91%s=92 expects type =91char *=92, but = argument 4 has type =91int=92 tpupload.c:553: error: =91stmt=92 undeclared (first use in this = function) tpupload.c:556: warning: implicit declaration of function = =91mysql_close=92 make[1]: *** [tpupload.o] Error 1 make[1]: Leaving directory = `/build/filedrawers/upload' make: *** [upload/upload.cgi] Error 2 -Daniel -----Original Message----- From: Aaron Knister [mailto:aaronk@umbc.edu] Sent: Mon 8/15/2011 12:32 PM To: Colson, Daniel Cc: openafs-info@openafs.org Subject: Re: [OpenAFS] 64-bit mod_waklog / filedrawers on afs =20 Hi Daniel, While I'm not running filedrawers I am using mod_waklog, albeit with = RHEL6. I had to apply some patches to get it to build correctly. If you send = your build errors I can see if my patches will help. -Aaron On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel = wrote: > ** > > Is anyone else running mod_waklog / filedrawers on afs on a 64-bit = machine? > I've been unable to compile mod_waklog (an essential component of > filedrawers) on 64-bit (RHEL5.6). > --=20 Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu ------_=_NextPart_001_01CC5C56.DBA2EA2D Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable RE: [OpenAFS] 64-bit mod_waklog / filedrawers on afs

Thanks. Any help is appreciated. Here's the error = output from "make".

make[1]: Entering directory = `<path-to-build-dir>/build/filedrawers/libcgi'
gcc -Wall -Wstrict-prototypes -g  -fPIC -I/usr/include/apr-1  = -c -o cgi.o cgi.c
cgi.c: In function ‘post_getline’:
cgi.c:91: warning: function declaration isn’t a prototype
cgi.c: In function ‘cgi_init’:
cgi.c:148: warning: function declaration isn’t a prototype
cgi.c: In function ‘mp_get_file’:
cgi.c:361: warning: function declaration isn’t a prototype
cgi.c:364: warning: function declaration isn’t a prototype
cgi.c:375: warning: function declaration isn’t a prototype
cgi.c:378: warning: function declaration isn’t a prototype
cgi.c:385: warning: function declaration isn’t a prototype
cgi.c:388: warning: function declaration isn’t a prototype
cgi.c:400: warning: function declaration isn’t a prototype
cgi.c:420: warning: function declaration isn’t a prototype
cgi.c:424: warning: function declaration isn’t a prototype
cgi.c: In function ‘cgi_multipart’:
cgi.c:539: warning: function declaration isn’t a prototype
cgi.c:614: warning: function declaration isn’t a prototype
building normal libcgi.a
make[1]: Leaving directory = `<path-to-build-dir>/build/filedrawers/libcgi'
( cd upload; make  )
make[1]: Entering directory = `<path-to-build-dir>/build/filedrawers/upload'
gcc -Wall -Wmissing-prototypes -g -O2 -I../libcgi = -I/usr/include/mysql  -g -pipe -Wp,-D_FORTIFY_SOURCE=3D2 = -fexceptions -fstack-protector --param=3Dssp-buffer-size=3D4 -m64 = -D_GNU_SOURCE -D_FILE_OFFSET_BITS=3D64 -D_LARGEFILE_SOURCE = -fno-strict-aliasing -fwrapv -Wall -Wmissing-prototypes \
            &= nbsp;   -D_URL_ERROR_PAGE=3D\"https://<site>?error=3Dtrue\" \
            &= nbsp;   = -D_FILEDRAWERS_CONFIG_PATH=3D\"/var/www/filedrawers/lib/filedrawers.= conf\" \
            &= nbsp;   -c tpupload.c
tpupload.c:18:19: error: mysql.h: No such file or directory
tpupload.c:34: error: expected ‘=3D’, ‘,’, = ‘;’, ‘asm’ or ‘__attribute__’ before = ‘*’ token
tpupload.c:35: error: expected ‘=3D’, ‘,’, = ‘;’, ‘asm’ or ‘__attribute__’ before = ‘*’ token
tpupload.c: In function ‘upload_progress_db_insert’:
tpupload.c:103: error: ‘MYSQL_BIND’ undeclared (first use in = this function)
tpupload.c:103: error: (Each undeclared identifier is reported only = once
tpupload.c:103: error: for each function it appears in.)
tpupload.c:103: error: expected ‘;’ before = ‘bind’
tpupload.c:104: error: ‘my_bool’ undeclared (first use in this = function)
tpupload.c:104: error: expected ‘;’ before = ‘is_null’
tpupload.c:110: error: ‘stmt’ undeclared (first use in this = function)
tpupload.c:111: warning: implicit declaration of function = ‘mysql_stmt_init’
tpupload.c:111: error: ‘mysql’ undeclared (first use in this = function)
tpupload.c:112: warning: implicit declaration of function = ‘mysql_error’
tpupload.c:112: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c:123: warning: implicit declaration of function = ‘mysql_stmt_prepare’
tpupload.c:124: warning: implicit declaration of function = ‘mysql_stmt_error’
tpupload.c:124: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c:130: error: ‘bind’ undeclared (first use in this = function)
tpupload.c:130: warning: left-hand operand of comma expression has no = effect
tpupload.c:132: error: ‘MYSQL_TYPE_STRING’ undeclared (first = use in this function)
tpupload.c:132: error: expected expression before ‘)’ = token
tpupload.c:132: error: ‘is_null’ undeclared (first use in this = function)
tpupload.c:133: error: expected expression before ‘)’ = token
tpupload.c:134: error: ‘MYSQL_TYPE_LONGLONG’ undeclared (first = use in this function)
tpupload.c:134: error: expected expression before ‘)’ = token
tpupload.c:135: error: expected expression before ‘)’ = token
tpupload.c:136: error: ‘MYSQL_TYPE_TINY’ undeclared (first use = in this function)
tpupload.c:136: error: expected expression before ‘)’ = token
tpupload.c:138: warning: implicit declaration of function = ‘mysql_stmt_bind_param’
tpupload.c:139: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c:148: warning: implicit declaration of function = ‘mysql_stmt_execute’
tpupload.c:149: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c: In function ‘upload_progress_db_update’:
tpupload.c:160: error: ‘MYSQL_BIND’ undeclared (first use in = this function)
tpupload.c:160: error: expected ‘;’ before = ‘bind’
tpupload.c:161: error: ‘my_bool’ undeclared (first use in this = function)
tpupload.c:161: error: expected ‘;’ before = ‘is_null’
tpupload.c:166: error: ‘stmt’ undeclared (first use in this = function)
tpupload.c:167: error: ‘mysql’ undeclared (first use in this = function)
tpupload.c:168: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c:180: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c:186: error: ‘bind’ undeclared (first use in this = function)
tpupload.c:186: warning: left-hand operand of comma expression has no = effect
tpupload.c:188: error: ‘MYSQL_TYPE_STRING’ undeclared (first = use in this function)
tpupload.c:188: error: expected expression before ‘)’ = token
tpupload.c:188: error: ‘is_null’ undeclared (first use in this = function)
tpupload.c:189: error: ‘MYSQL_TYPE_LONGLONG’ undeclared (first = use in this function)
tpupload.c:189: error: expected expression before ‘)’ = token
tpupload.c:190: error: ‘MYSQL_TYPE_TINY’ undeclared (first use = in this function)
tpupload.c:190: error: expected expression before ‘)’ = token
tpupload.c:191: error: expected expression before ‘)’ = token
tpupload.c:194: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c:203: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c: In function ‘upload_init’:
tpupload.c:317: error: ‘mysql’ undeclared (first use in this = function)
tpupload.c:322: error: ‘stmt’ undeclared (first use in this = function)
tpupload.c:323: warning: implicit declaration of function = ‘mysql_stmt_close’
tpupload.c:324: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c: In function ‘upload_progress’:
tpupload.c:347: error: ‘mysql’ undeclared (first use in this = function)
tpupload.c: In function ‘upload_complete’:
tpupload.c:373: error: ‘mysql’ undeclared (first use in this = function)
tpupload.c: In function ‘main’:
tpupload.c:528: error: ‘mysql’ undeclared (first use in this = function)
tpupload.c:528: warning: implicit declaration of function = ‘mysql_init’
tpupload.c:530: warning: implicit declaration of function = ‘mysql_real_connect’
tpupload.c:532: warning: format ‘%s’ expects type ‘char = *’, but argument 4 has type ‘int’
tpupload.c:553: error: ‘stmt’ undeclared (first use in this = function)
tpupload.c:556: warning: implicit declaration of function = ‘mysql_close’
make[1]: *** [tpupload.o] Error 1
make[1]: Leaving directory = `<path-to-build-dir>/build/filedrawers/upload'
make: *** [upload/upload.cgi] Error 2

-Daniel

-----Original Message-----
From: Aaron Knister [mailto:aaronk@umbc.edu]
Sent: Mon 8/15/2011 12:32 PM
To: Colson, Daniel
Cc: openafs-info@openafs.org
Subject: Re: [OpenAFS] 64-bit mod_waklog / filedrawers on afs

Hi Daniel,

While I'm not running filedrawers I am using mod_waklog, albeit with = RHEL6.
I had to apply some patches to get it to build correctly. If you send = your
build errors I can see if my patches will help.

-Aaron

On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel = <dcolson5@uncc.edu> wrote:

> **
>
> Is anyone else running mod_waklog / filedrawers on afs on a 64-bit = machine?
> I've been unable to compile mod_waklog (an essential component = of
> filedrawers) on 64-bit (RHEL5.6).
>



--
Aaron Knister
Systems Administrator
Division of Information Technology
University of Maryland, Baltimore County
aaronk@umbc.edu

------_=_NextPart_001_01CC5C56.DBA2EA2D-- From vittorlima@gmail.com Thu Aug 18 14:48:08 2011 From: vittorlima@gmail.com (vitor lima) Date: Thu, 18 Aug 2011 10:48:08 -0300 Subject: [OpenAFS] Problems to move OpenAFS volumes Message-ID: --20cf30025888ea4b8904aac7dca8 Content-Type: text/plain; charset=ISO-8859-1 Hello! I have the OpenAFS installed and working in the server0. I installed the OpenAFS in the server1 too. I want to move the volumes from server0 to server1, so I execute vos move -id user_x -fromserver server0 -frompartition /vicepb -toserver server1 -topartition /vicepa and the volume is moved with successful. But when the user_x try to log in, the home directory is not find. Can somebody help me? Thank very much! -- Vitor Lima --20cf30025888ea4b8904aac7dca8 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hello!

I have the OpenAFS installed and working in the server0.
I= installed the OpenAFS in the server1 too.

I want to move the volume= s from server0 to server1, so I execute

vos move -id user_x -fromser= ver server0 -frompartition /vicepb -toserver server1 -topartition /vicepa
and the volume is moved with successful.

But when the user_x try= to log in, the home directory=A0 is not find.

Can somebody help me?=




Thank very much!

--
Vitor Lima=
--20cf30025888ea4b8904aac7dca8-- From dcolson5@uncc.edu Mon Aug 22 14:54:16 2011 From: dcolson5@uncc.edu (Colson, Daniel) Date: Mon, 22 Aug 2011 09:54:16 -0400 Subject: [OpenAFS] 64-bit mod_waklog / filedrawers on afs References: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBE2@EXEVS04.its.uncc.edu> Message-ID: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBE3@EXEVS04.its.uncc.edu> This is a multi-part message in MIME format. ------_=_NextPart_001_01CC60D3.867762B1 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Oops. I apologize. I sent the output from the filedrawers build instead = of the mod_waklog build. I'll check on mysql-devel though. Below is the = output from "make" for mod_waklog. `/usr/sbin/apxs -q CC` -I/usr/include/ -I/usr/include/afs = -I/usr/include/httpd -DEAPI -g `/usr/sbin/apxs -q CFLAGS` -c = mod_waklog.c -o mod_waklog.o In file included from /usr/include/httpd/httpd.h:43, from mod_waklog.c:4: /usr/include/httpd/ap_config.h:25:17: error: apr.h: No such file or = directory /usr/include/httpd/ap_config.h:26:23: error: apr_hooks.h: No such file = or directory /usr/include/httpd/ap_config.h:27:32: error: apr_optional_hooks.h: No = such file or directory In file included from /usr/include/httpd/httpd.h:46, from mod_waklog.c:4: /usr/include/httpd/ap_release.h:25:41: error: apr_general.h: No such = file or directory In file included from mod_waklog.c:4: /usr/include/httpd/httpd.h:50:24: error: apr_tables.h: No such file or = directory /usr/include/httpd/httpd.h:51:23: error: apr_pools.h: No such file or = directory /usr/include/httpd/httpd.h:52:22: error: apr_time.h: No such file or = directory /usr/include/httpd/httpd.h:53:28: error: apr_network_io.h: No such file = or directory /usr/include/httpd/httpd.h:54:25: error: apr_buckets.h: No such file or = directory /usr/include/httpd/httpd.h:55:22: error: apr_poll.h: No such file or = directory In file included from /usr/include/httpd/httpd.h:59, from mod_waklog.c:4: /usr/include/httpd/ap_regex.h:87: error: expected = specifier-qualifier-list before =91apr_size_t=92 /usr/include/httpd/ap_regex.h:118: error: expected declaration = specifiers or =91...=92 before =91apr_size_t=92 /usr/include/httpd/ap_regex.h:127: error: expected =91=3D=92, =91,=92, = =91;=92, =91asm=92 or =91__attribute__=92 before =91ap_regerror=92 In file included from mod_waklog.c:4: /usr/include/httpd/httpd.h:433: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:601: error: expected specifier-qualifier-list = before =91apr_int64_t=92 /usr/include/httpd/httpd.h:723:21: error: apr_uri.h: No such file or = directory /usr/include/httpd/httpd.h:730: error: expected specifier-qualifier-list = before =91apr_pool_t=92 /usr/include/httpd/httpd.h:746: error: expected specifier-qualifier-list = before =91apr_pool_t=92 /usr/include/httpd/httpd.h:1008: error: expected = specifier-qualifier-list before =91apr_pool_t=92 /usr/include/httpd/httpd.h:1085: error: expected = specifier-qualifier-list before =91APR_RING_ENTRY=92 /usr/include/httpd/httpd.h:1118: error: expected = specifier-qualifier-list before =91apr_sockaddr_t=92 /usr/include/httpd/httpd.h:1146: error: expected = specifier-qualifier-list before =91apr_port_t=92 /usr/include/httpd/httpd.h:1202: error: expected = specifier-qualifier-list before =91apr_bucket_brigade=92 /usr/include/httpd/httpd.h:1210: error: expected = specifier-qualifier-list before =91apr_bucket_brigade=92 /usr/include/httpd/httpd.h:1216: error: expected = specifier-qualifier-list before =91apr_socket_t=92 /usr/include/httpd/httpd.h:1232: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1242: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1255: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1265: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1274: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1284: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1294: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1306: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1314: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1323: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1334: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1357: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1367: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1379: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1388: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1397: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1439: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1449: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1460: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1468: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1477: error: expected =91=3D=92, =91,=92, = =91;=92, =91asm=92 or =91__attribute__=92 before = =91ap_escape_errorlog_item=92 /usr/include/httpd/httpd.h:1488: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1497: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1526: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1540: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1550: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1602: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1610: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1624: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1632: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1645: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1685: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1697: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1707: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1715: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/httpd.h:1726: error: expected =91)=92 before =91*=92 = token In file included from mod_waklog.c:5: /usr/include/httpd/http_config.h:279: error: expected = specifier-qualifier-list before =91apr_int64_t=92 /usr/include/httpd/http_config.h:357: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_config.h:365: error: expected =91;=92 before = =91void=92 /usr/include/httpd/http_config.h:557: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_config.h:566: error: expected declaration = specifiers or =91...=92 before =91apr_pool_t=92 /usr/include/httpd/http_config.h:582: error: expected declaration = specifiers or =91...=92 before =91apr_pool_t=92 /usr/include/httpd/http_config.h:607: error: expected =91=3D=92, = =91,=92, =91;=92, =91asm=92 or =91__attribute__=92 before = =91ap_pcfg_openfile=92 /usr/include/httpd/http_config.h:619: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_config.h:669: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_config.h:685: error: expected declaration = specifiers or =91...=92 before =91apr_pool_t=92 /usr/include/httpd/http_config.h:686: error: expected declaration = specifiers or =91...=92 before =91apr_pool_t=92 /usr/include/httpd/http_config.h:975: error: expected =91)=92 before = =91int=92 In file included from mod_waklog.c:6: /usr/include/httpd/http_log.h:33:29: error: apr_thread_proc.h: No such = file or directory In file included from /usr/include/syslog.h:1, from /usr/include/httpd/http_log.h:36, from mod_waklog.c:6: /usr/include/sys/syslog.h:200: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 In file included from /usr/include/sys/syslog.h:207, from /usr/include/syslog.h:1, from /usr/include/httpd/http_log.h:36, from mod_waklog.c:6: /usr/include/bits/syslog.h:34: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 In file included from mod_waklog.c:6: /usr/include/httpd/http_log.h:102: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_log.h:109: error: expected =91=3D=92, =91,=92, = =91;=92, =91asm=92 or =91__attribute__=92 before = =91ap_replace_stderr_log=92 /usr/include/httpd/http_log.h:121: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_log.h:172: error: expected declaration = specifiers or =91...=92 before =91apr_status_t=92 /usr/include/httpd/http_log.h:174: error: format string argument not a = string type /usr/include/httpd/http_log.h:196: error: expected declaration = specifiers or =91...=92 before =91apr_status_t=92 /usr/include/httpd/http_log.h:196: error: expected declaration = specifiers or =91...=92 before =91apr_pool_t=92 /usr/include/httpd/http_log.h:198: error: format string argument not a = string type /usr/include/httpd/http_log.h:220: error: expected declaration = specifiers or =91...=92 before =91apr_status_t=92 /usr/include/httpd/http_log.h:222: error: format string argument not a = string type /usr/include/httpd/http_log.h:246: error: expected declaration = specifiers or =91...=92 before =91apr_status_t=92 /usr/include/httpd/http_log.h:248: error: format string argument not a = string type /usr/include/httpd/http_log.h:261: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_log.h:269: error: expected =91=3D=92, =91,=92, = =91;=92, =91asm=92 or =91__attribute__=92 before =91ap_read_pid=92 /usr/include/httpd/http_log.h:282: error: expected = specifier-qualifier-list before =91apr_pool_t=92 /usr/include/httpd/http_log.h:301: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_log.h:335: error: expected =91)=92 before = =91void=92 In file included from mod_waklog.c:7: /usr/include/httpd/http_protocol.h:31:26: error: apr_portable.h: No such = file or directory /usr/include/httpd/http_protocol.h:32:22: error: apr_mmap.h: No such = file or directory In file included from /usr/include/httpd/http_protocol.h:34, from mod_waklog.c:7: /usr/include/httpd/util_filter.h:72: warning: data definition has no = type or storage class /usr/include/httpd/util_filter.h:72: warning: type defaults to =91int=92 = in declaration of =91ap_input_mode_t=92 /usr/include/httpd/util_filter.h:136: error: expected declaration = specifiers or =91...=92 before =91*=92 token /usr/include/httpd/util_filter.h:137: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/util_filter.h:137: warning: type defaults to = =91int=92 in declaration of =91apr_status_t=92 /usr/include/httpd/util_filter.h:137: error: =91apr_status_t=92 declared = as function returning a function /usr/include/httpd/util_filter.h:139: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/util_filter.h:140: error: expected declaration = specifiers or =91...=92 before =91ap_input_mode_t=92 /usr/include/httpd/util_filter.h:141: error: expected declaration = specifiers or =91...=92 before =91apr_read_type_e=92 /usr/include/httpd/util_filter.h:142: error: expected declaration = specifiers or =91...=92 before =91apr_off_t=92 /usr/include/httpd/util_filter.h:142: error: =91ap_in_filter_func=92 = declared as function returning a function /usr/include/httpd/util_filter.h:146: error: expected = specifier-qualifier-list before =91ap_out_filter_func=92 /usr/include/httpd/util_filter.h:299: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/util_filter.h:300: error: expected declaration = specifiers or =91...=92 before =91ap_input_mode_t=92 /usr/include/httpd/util_filter.h:301: error: expected declaration = specifiers or =91...=92 before =91apr_read_type_e=92 /usr/include/httpd/util_filter.h:302: error: expected declaration = specifiers or =91...=92 before =91apr_off_t=92 /usr/include/httpd/util_filter.h:302: error: =91ap_get_brigade=92 = declared as function returning a function /usr/include/httpd/util_filter.h:313: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/util_filter.h:313: error: =91ap_pass_brigade=92 = declared as function returning a function /usr/include/httpd/util_filter.h:351: error: expected declaration = specifiers or =91...=92 before =91ap_out_filter_func=92 /usr/include/httpd/util_filter.h:377: error: expected declaration = specifiers or =91...=92 before =91ap_out_filter_func=92 /usr/include/httpd/util_filter.h:495: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/util_filter.h:496: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/util_filter.h:496: error: expected declaration = specifiers or =91...=92 before =91apr_pool_t=92 /usr/include/httpd/util_filter.h:496: error: =91ap_save_brigade=92 = declared as function returning a function /usr/include/httpd/util_filter.h:506: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/util_filter.h:514: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/util_filter.h:514: error: =91ap_fflush=92 declared as = function returning a function /usr/include/httpd/util_filter.h:551: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/util_filter.h:552: error: =91ap_fputstrs=92 declared = as function returning a function /usr/include/httpd/util_filter.h:562: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/util_filter.h:565: error: =91ap_fprintf=92 declared = as function returning a function /usr/include/httpd/util_filter.h:565: error: format string argument not = a string type In file included from mod_waklog.c:7: /usr/include/httpd/http_protocol.h:45: error: expected =91)=92 before = =91void=92 /usr/include/httpd/http_protocol.h:77: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/http_protocol.h:111: error: expected declaration = specifiers or =91...=92 before =91apr_off_t=92 /usr/include/httpd/http_protocol.h:127: error: expected =91=3D=92, = =91,=92, =91;=92, =91asm=92 or =91__attribute__=92 before = =91ap_rationalize_mtime=92 /usr/include/httpd/http_protocol.h:203: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_protocol.h:228: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_protocol.h:235: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_protocol.h:253: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_protocol.h:356: error: expected declaration = specifiers or =91...=92 before =91va_list=92 /usr/include/httpd/http_protocol.h:426: error: expected declaration = specifiers or =91...=92 before =91apr_size_t=92 /usr/include/httpd/http_protocol.h:534: error: expected declaration = specifiers or =91...=92 before =91apr_size_t=92 /usr/include/httpd/http_protocol.h:535: error: expected declaration = specifiers or =91...=92 before =91apr_size_t=92 /usr/include/httpd/http_protocol.h:537: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/http_protocol.h:537: error: =91ap_rgetline_core=92 = declared as function returning a function /usr/include/httpd/http_protocol.h:554: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_protocol.h:569: error: expected =91)=92 before = =91int=92 /usr/include/httpd/http_protocol.h:608: error: expected = specifier-qualifier-list before =91apr_bucket_refcount=92 /usr/include/httpd/http_protocol.h:616: error: expected =91=3D=92, = =91,=92, =91;=92, =91asm=92 or =91__attribute__=92 before = =91ap_bucket_type_error=92 /usr/include/httpd/http_protocol.h:633: error: expected =91=3D=92, = =91,=92, =91;=92, =91asm=92 or =91__attribute__=92 before =91*=92 token /usr/include/httpd/http_protocol.h:644: error: expected =91=3D=92, = =91,=92, =91;=92, =91asm=92 or =91__attribute__=92 before =91*=92 token /usr/include/httpd/http_protocol.h:648: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/http_protocol.h:648: error: =91ap_byterange_filter=92 = declared as function returning a function /usr/include/httpd/http_protocol.h:649: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/http_protocol.h:649: error: = =91ap_http_header_filter=92 declared as function returning a function /usr/include/httpd/http_protocol.h:651: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/http_protocol.h:651: error: = =91ap_content_length_filter=92 declared as function returning a function /usr/include/httpd/http_protocol.h:652: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/http_protocol.h:652: error: =91ap_old_write_filter=92 = declared as function returning a function In file included from mod_waklog.c:8: /usr/include/httpd/http_request.h:105: warning: type defaults to = =91int=92 in declaration of =91apr_finfo_t=92 /usr/include/httpd/http_request.h:105: error: expected =91;=92, =91,=92 = or =91)=92 before =91*=92 token /usr/include/httpd/http_request.h:131: error: expected declaration = specifiers or =91...=92 before =91apr_bucket_brigade=92 /usr/include/httpd/http_request.h:131: error: = =91ap_sub_req_output_filter=92 declared as function returning a function /usr/include/httpd/http_request.h:198: error: expected declaration = specifiers or =91...=92 before =91apr_time_t=92 /usr/include/httpd/http_request.h:261: error: expected =91)=92 before = =91int=92 In file included from mod_waklog.c:9: /usr/include/httpd/http_core.h:30:22: error: apr_hash.h: No such file or = directory /usr/include/httpd/http_core.h:31:26: error: apr_optional.h: No such = file or directory In file included from mod_waklog.c:9: /usr/include/httpd/http_core.h:221: error: expected =91)=92 before = =91*=92 token /usr/include/httpd/http_core.h:235: error: expected =91=3D=92, =91,=92, = =91;=92, =91asm=92 or =91__attribute__=92 before = =91ap_get_server_port=92 /usr/include/httpd/http_core.h:242: error: expected =91=3D=92, =91,=92, = =91;=92, =91asm=92 or =91__attribute__=92 before = =91ap_get_limit_req_body=92 /usr/include/httpd/http_core.h:291: error: expected = specifier-qualifier-list before =91apr_int64_t=92 /usr/include/httpd/http_core.h:327: error: expected =91=3D=92, =91,=92, = =91;=92, =91asm=92 or =91__attribute__=92 before =91*=92 token /usr/include/httpd/http_core.h:646: error: expected = specifier-qualifier-list before =91apr_hash_t=92 /usr/include/httpd/http_core.h:675: error: expected =91)=92 before = =91int=92 /usr/include/httpd/http_core.h:693: error: expected declaration = specifiers or =91...=92 before =91ap_ident_lookup=92 /usr/include/httpd/http_core.h:694: error: expected declaration = specifiers or =91...=92 before =91(=92 token /usr/include/httpd/http_core.h:694: warning: data definition has no type = or storage class /usr/include/httpd/http_core.h:694: warning: type defaults to =91int=92 = in declaration of =91APR_DECLARE_OPTIONAL_FN=92 mod_waklog.c:49:25: error: apr_strings.h: No such file or directory mod_waklog.c:50:24: error: apr_base64.h: No such file or directory In file included from /usr/include/httpd/unixd.h:30, from mod_waklog.c:57: /usr/include/httpd/ap_listen.h:34: error: expected declaration = specifiers or =91...=92 before =91apr_pool_t=92 /usr/include/httpd/ap_listen.h:34: error: =91accept_function=92 declared = as function returning a function /usr/include/httpd/ap_listen.h:50: error: expected = specifier-qualifier-list before =91apr_socket_t=92 In file included from mod_waklog.c:57: /usr/include/httpd/unixd.h:39:28: error: apr_proc_mutex.h: No such file = or directory /usr/include/httpd/unixd.h:40:30: error: apr_global_mutex.h: No such = file or directory In file included from mod_waklog.c:57: /usr/include/httpd/unixd.h:57: error: expected =91)=92 before = =91ap_unix_identity_t=92 /usr/include/httpd/unixd.h:75: warning: data definition has no type or = storage class /usr/include/httpd/unixd.h:75: warning: type defaults to =91int=92 in = declaration of =91unixd_config_rec=92 /usr/include/httpd/unixd.h:76: error: expected =91=3D=92, =91,=92, = =91;=92, =91asm=92 or =91__attribute__=92 before =91unixd_config=92 /usr/include/httpd/unixd.h:79: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/unixd.h:98: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/unixd.h:99: error: expected =91)=92 before =91*=92 = token /usr/include/httpd/unixd.h:100: error: expected declaration specifiers = or =91...=92 before =91apr_pool_t=92 /usr/include/httpd/unixd.h:100: error: =91unixd_accept=92 declared as = function returning a function mod_waklog.c:58: error: expected =91=3D=92, =91,=92, =91;=92, =91asm=92 = or =91__attribute__=92 before =91unixd_config=92 In file included from /usr/include/et/com_err.h:19, from /usr/include/krb5/krb5.h:2690, from /usr/include/krb5.h:7, from mod_waklog.c:73: /usr/lib/gcc/x86_64-redhat-linux/4.1.2/include/stdarg.h:105: error: = expected =91=3D=92, =91,=92, =91;=92, =91asm=92 or =91__attribute__=92 = before =91va_list=92 In file included from /usr/include/krb5/krb5.h:2690, from /usr/include/krb5.h:7, from mod_waklog.c:73: /usr/include/et/com_err.h:34: error: expected declaration specifiers or = =91...=92 before =91va_list=92 /usr/include/et/com_err.h:38: error: expected declaration specifiers or = =91...=92 before =91va_list=92 /usr/include/et/com_err.h:40: error: expected declaration specifiers or = =91...=92 before =91va_list=92 /usr/include/et/com_err.h:41: error: expected declaration specifiers or = =91...=92 before =91va_list=92 /usr/include/et/com_err.h:43: error: expected declaration specifiers or = =91...=92 before =91va_list=92 In file included from mod_waklog.c:75: /usr/include/afs/venus.h:33: error: field =91server=92 has incomplete = type In file included from /usr/include/stdio.h:72, from /usr/include/rx/rx.h:42, from /usr/include/rx/rxkad_prototypes.h:15, from /usr/include/rx/rxkad.h:102, from /usr/include/afs/auth.h:19, from mod_waklog.c:76: /usr/include/libio.h:482: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/libio.h:484: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 In file included from /usr/include/rx/rx.h:42, from /usr/include/rx/rxkad_prototypes.h:15, from /usr/include/rx/rxkad.h:102, from /usr/include/afs/auth.h:19, from mod_waklog.c:76: /usr/include/stdio.h:343: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/stdio.h:348: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/stdio.h:351: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/stdio.h:362: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/stdio.h:371: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/stdio.h:387: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/stdio.h:418: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/stdio.h:425: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/stdio.h:430: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 In file included from /usr/include/rx/rx.h:42, from /usr/include/rx/rxkad_prototypes.h:15, from /usr/include/rx/rxkad.h:102, from /usr/include/afs/auth.h:19, from mod_waklog.c:76: /usr/include/stdio.h:805: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 In file included from /usr/include/stdio.h:835, from /usr/include/rx/rx.h:42, from /usr/include/rx/rxkad_prototypes.h:15, from /usr/include/rx/rxkad.h:102, from /usr/include/afs/auth.h:19, from mod_waklog.c:76: /usr/include/bits/stdio.h:34: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/bits/stdio.h: In function =91vprintf=92: /usr/include/bits/stdio.h:36: error: =91__arg=92 undeclared (first use = in this function) /usr/include/bits/stdio.h:36: error: (Each undeclared identifier is = reported only once /usr/include/bits/stdio.h:36: error: for each function it appears in.) /usr/include/bits/stdio.h:36: error: too many arguments to function = =91vfprintf=92 In file included from /usr/include/stdio.h:838, from /usr/include/rx/rx.h:42, from /usr/include/rx/rxkad_prototypes.h:15, from /usr/include/rx/rxkad.h:102, from /usr/include/afs/auth.h:19, from mod_waklog.c:76: /usr/include/bits/stdio2.h: At top level: /usr/include/bits/stdio2.h:28: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/bits/stdio2.h:43: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/bits/stdio2.h:60: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 /usr/include/bits/stdio2.h:62: error: expected declaration specifiers or = =91...=92 before =91__gnuc_va_list=92 mod_waklog.c:111: error: expected specifier-qualifier-list before = =91apr_pool_t=92 mod_waklog.c: In function =91log_error=92: mod_waklog.c:184: error: =91va_list=92 undeclared (first use in this = function) mod_waklog.c:184: error: expected =91;=92 before =91ap=92 mod_waklog.c:186: error: =91ap=92 undeclared (first use in this = function) mod_waklog.c:191: warning: passing argument 4 of =91ap_log_error=92 = makes pointer from integer without a cast mod_waklog.c:191: warning: passing argument 5 of =91ap_log_error=92 from = incompatible pointer type mod_waklog.c: In function =91retrieve_config=92: mod_waklog.c:203: error: =91request_rec=92 has no member named = =91main=92 mod_waklog.c:204: error: =91request_rec=92 has no member named = =91main=92 mod_waklog.c:211: error: =91request_rec=92 has no member named = =91per_dir_config=92 mod_waklog.c:214: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c: In function =91set_auth=92: mod_waklog.c:285: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:285: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:298: error: =91server_rec=92 has no member named = =91module_config=92 mod_waklog.c:311: error: =91request_rec=92 has no member named = =91connection=92 mod_waklog.c:311: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:313: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:313: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:317: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:317: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:327: warning: implicit declaration of function = =91apr_table_get=92 mod_waklog.c:327: error: =91request_rec=92 has no member named = =91subprocess_env=92 mod_waklog.c:327: warning: cast to pointer from integer of different = size mod_waklog.c:330: error: =91request_rec=92 has no member named = =91notes=92 mod_waklog.c:330: warning: cast to pointer from integer of different = size mod_waklog.c:339: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:340: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:340: error: =91request_rec=92 has no member named = =91user=92 mod_waklog.c:703: warning: label =91usecachedtoken=92 defined but not = used mod_waklog.c: At top level: mod_waklog.c:749: error: expected =91)=92 before =91*=92 token mod_waklog.c:778: error: expected =91)=92 before =91*=92 token mod_waklog.c:801: error: expected =91)=92 before =91*=92 token mod_waklog.c:831: error: expected =91)=92 before =91*=92 token mod_waklog.c: In function =91set_waklog_enabled=92: mod_waklog.c:870: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:874: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c: At top level: mod_waklog.c:883: error: expected =91)=92 before =91*=92 token mod_waklog.c: In function =91set_waklog_location_principal=92: mod_waklog.c:912: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:914: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:917: warning: implicit declaration of function = =91apr_pstrdup=92 mod_waklog.c:917: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:917: warning: assignment makes pointer from integer without = a cast mod_waklog.c:918: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:918: warning: assignment makes pointer from integer without = a cast mod_waklog.c:920: warning: implicit declaration of function = =91add_to_renewtable=92 mod_waklog.c:920: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c: In function =91set_waklog_afs_cell=92: mod_waklog.c:932: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:934: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:938: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:938: warning: assignment makes pointer from integer without = a cast mod_waklog.c:943: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:943: warning: assignment makes pointer from integer without = a cast mod_waklog.c: In function =91set_waklog_afs_cell_realm=92: mod_waklog.c:954: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:956: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:959: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:959: warning: assignment makes pointer from integer without = a cast mod_waklog.c:962: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:962: warning: assignment makes pointer from integer without = a cast mod_waklog.c: In function =91set_waklog_default_principal=92: mod_waklog.c:971: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:973: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:975: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:978: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:978: warning: assignment makes pointer from integer without = a cast mod_waklog.c:979: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:979: warning: assignment makes pointer from integer without = a cast mod_waklog.c:983: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:983: warning: assignment makes pointer from integer without = a cast mod_waklog.c:984: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c:984: warning: assignment makes pointer from integer without = a cast mod_waklog.c:986: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:990: error: =91cmd_parms=92 has no member named =91pool=92 mod_waklog.c: In function =91set_waklog_use_usertokens=92: mod_waklog.c:1001: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:1007: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c: In function =91set_waklog_disable_token_cache=92: mod_waklog.c:1017: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c:1023: error: =91cmd_parms=92 has no member named = =91server=92 mod_waklog.c: At top level: mod_waklog.c:1032: error: =91waklog_child_exit=92 declared as function = returning a function mod_waklog.c: In function =91waklog_child_exit=92: mod_waklog.c:1053: error: =91APR_SUCCESS=92 undeclared (first use in = this function) mod_waklog.c: At top level: mod_waklog.c:1060: error: expected =91)=92 before =91*=92 token mod_waklog.c: In function =91token_cleanup=92: mod_waklog.c:1154: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c: In function =91waklog_child_routine=92: mod_waklog.c:1173: error: =91server_rec=92 has no member named = =91module_config=92 mod_waklog.c: At top level: mod_waklog.c:1234: error: expected =91)=92 before =91*=92 token mod_waklog.c: In function =91waklog_phase0=92: mod_waklog.c:1489: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1495: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1496: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1498: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1499: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1507: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c: In function =91waklog_phase1=92: mod_waklog.c:1518: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1524: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1525: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1527: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1528: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1530: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c: In function =91waklog_phase3=92: mod_waklog.c:1541: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1547: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1548: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1550: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1551: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1553: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c: In function =91waklog_phase6=92: mod_waklog.c:1564: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1570: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1571: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1573: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1574: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1576: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c: In function =91waklog_phase7=92: mod_waklog.c:1588: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1594: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1595: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1597: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1598: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1600: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1601: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1603: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c: In function =91waklog_phase9=92: mod_waklog.c:1622: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1625: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1628: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1629: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c: In function =91waklog_new_connection=92: mod_waklog.c:1651: error: =91conn_rec=92 has no member named = =91base_server=92 mod_waklog.c:1654: error: =91conn_rec=92 has no member named = =91base_server=92 mod_waklog.c:1657: error: =91conn_rec=92 has no member named = =91base_server=92 mod_waklog.c:1659: error: =91conn_rec=92 has no member named = =91base_server=92 mod_waklog.c: In function =91waklog_phase2=92: mod_waklog.c:1684: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1693: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c:1698: error: =91request_rec=92 has no member named = =91server=92 mod_waklog.c: At top level: mod_waklog.c:1734: error: expected =91)=92 before =91*=92 token mod_waklog.c:1751: error: =91waklog_create_dir_config=92 undeclared here = (not in a function) mod_waklog.c:1751: warning: excess elements in struct initializer mod_waklog.c:1751: warning: (near initialization for = =91waklog_module=92) mod_waklog.c:1752: error: =91waklog_merge_dir_config=92 undeclared here = (not in a function) mod_waklog.c:1752: warning: excess elements in struct initializer mod_waklog.c:1752: warning: (near initialization for = =91waklog_module=92) mod_waklog.c:1753: error: =91waklog_create_server_config=92 undeclared = here (not in a function) mod_waklog.c:1753: warning: excess elements in struct initializer mod_waklog.c:1753: warning: (near initialization for = =91waklog_module=92) mod_waklog.c:1754: warning: excess elements in struct initializer mod_waklog.c:1754: warning: (near initialization for = =91waklog_module=92) mod_waklog.c:1755: warning: excess elements in struct initializer mod_waklog.c:1755: warning: (near initialization for = =91waklog_module=92) mod_waklog.c:1756: error: =91waklog_register_hooks=92 undeclared here = (not in a function) mod_waklog.c:1757: warning: excess elements in struct initializer mod_waklog.c:1757: warning: (near initialization for = =91waklog_module=92) make: *** [mod_waklog.o] Error 1 Thanks, Daniel -----Original Message----- From: Aaron Knister [mailto:aaronk@umbc.edu] Sent: Wed 8/17/2011 10:59 AM To: Colson, Daniel Cc: openafs-info@openafs.org Subject: Re: [OpenAFS] 64-bit mod_waklog / filedrawers on afs =20 Hi Daniel, Well... I've never built filedrawers, but this line: tpupload.c:18:19: error: mysql.h: No such file or directory suggests you're missing the mysql headers. If its redhat try a yum = install mysql-devel. -Aaron On Tue, Aug 16, 2011 at 4:50 PM, Colson, Daniel = wrote: > ** > > Thanks. Any help is appreciated. Here's the error output from "make". > > make[1]: Entering directory = `/build/filedrawers/libcgi' > gcc -Wall -Wstrict-prototypes -g -fPIC -I/usr/include/apr-1 -c -o = cgi.o > cgi.c > cgi.c: In function 'post_getline': > cgi.c:91: warning: function declaration isn't a prototype > cgi.c: In function 'cgi_init': > cgi.c:148: warning: function declaration isn't a prototype > cgi.c: In function 'mp_get_file': > cgi.c:361: warning: function declaration isn't a prototype > cgi.c:364: warning: function declaration isn't a prototype > cgi.c:375: warning: function declaration isn't a prototype > cgi.c:378: warning: function declaration isn't a prototype > cgi.c:385: warning: function declaration isn't a prototype > cgi.c:388: warning: function declaration isn't a prototype > cgi.c:400: warning: function declaration isn't a prototype > cgi.c:420: warning: function declaration isn't a prototype > cgi.c:424: warning: function declaration isn't a prototype > cgi.c: In function 'cgi_multipart': > cgi.c:539: warning: function declaration isn't a prototype > cgi.c:614: warning: function declaration isn't a prototype > building normal libcgi.a > make[1]: Leaving directory = `/build/filedrawers/libcgi' > ( cd upload; make ) > make[1]: Entering directory = `/build/filedrawers/upload' > gcc -Wall -Wmissing-prototypes -g -O2 -I../libcgi -I/usr/include/mysql = -g > -pipe -Wp,-D_FORTIFY_SOURCE=3D2 -fexceptions -fstack-protector > --param=3Dssp-buffer-size=3D4 -m64 -D_GNU_SOURCE = -D_FILE_OFFSET_BITS=3D64 > -D_LARGEFILE_SOURCE -fno-strict-aliasing -fwrapv -Wall = -Wmissing-prototypes > \ > -D_URL_ERROR_PAGE=3D\"https://?error=3Dtrue\" \ > > = -D_FILEDRAWERS_CONFIG_PATH=3D\"/var/www/filedrawers/lib/filedrawers.conf\= " \ > -c tpupload.c > tpupload.c:18:19: error: mysql.h: No such file or directory > tpupload.c:34: error: expected '=3D', ',', ';', 'asm' or = '__attribute__' > before '*' token > tpupload.c:35: error: expected '=3D', ',', ';', 'asm' or = '__attribute__' > before '*' token > tpupload.c: In function 'upload_progress_db_insert': > tpupload.c:103: error: 'MYSQL_BIND' undeclared (first use in this = function) > tpupload.c:103: error: (Each undeclared identifier is reported only = once > tpupload.c:103: error: for each function it appears in.) > tpupload.c:103: error: expected ';' before 'bind' > tpupload.c:104: error: 'my_bool' undeclared (first use in this = function) > tpupload.c:104: error: expected ';' before 'is_null' > tpupload.c:110: error: 'stmt' undeclared (first use in this function) > tpupload.c:111: warning: implicit declaration of function = 'mysql_stmt_init' > tpupload.c:111: error: 'mysql' undeclared (first use in this function) > tpupload.c:112: warning: implicit declaration of function = 'mysql_error' > tpupload.c:112: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c:123: warning: implicit declaration of function > 'mysql_stmt_prepare' > tpupload.c:124: warning: implicit declaration of function > 'mysql_stmt_error' > tpupload.c:124: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c:130: error: 'bind' undeclared (first use in this function) > tpupload.c:130: warning: left-hand operand of comma expression has no > effect > tpupload.c:132: error: 'MYSQL_TYPE_STRING' undeclared (first use in = this > function) > tpupload.c:132: error: expected expression before ')' token > tpupload.c:132: error: 'is_null' undeclared (first use in this = function) > tpupload.c:133: error: expected expression before ')' token > tpupload.c:134: error: 'MYSQL_TYPE_LONGLONG' undeclared (first use in = this > function) > tpupload.c:134: error: expected expression before ')' token > tpupload.c:135: error: expected expression before ')' token > tpupload.c:136: error: 'MYSQL_TYPE_TINY' undeclared (first use in this > function) > tpupload.c:136: error: expected expression before ')' token > tpupload.c:138: warning: implicit declaration of function > 'mysql_stmt_bind_param' > tpupload.c:139: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c:148: warning: implicit declaration of function > 'mysql_stmt_execute' > tpupload.c:149: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c: In function 'upload_progress_db_update': > tpupload.c:160: error: 'MYSQL_BIND' undeclared (first use in this = function) > tpupload.c:160: error: expected ';' before 'bind' > tpupload.c:161: error: 'my_bool' undeclared (first use in this = function) > tpupload.c:161: error: expected ';' before 'is_null' > tpupload.c:166: error: 'stmt' undeclared (first use in this function) > tpupload.c:167: error: 'mysql' undeclared (first use in this function) > tpupload.c:168: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c:180: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c:186: error: 'bind' undeclared (first use in this function) > tpupload.c:186: warning: left-hand operand of comma expression has no > effect > tpupload.c:188: error: 'MYSQL_TYPE_STRING' undeclared (first use in = this > function) > tpupload.c:188: error: expected expression before ')' token > tpupload.c:188: error: 'is_null' undeclared (first use in this = function) > tpupload.c:189: error: 'MYSQL_TYPE_LONGLONG' undeclared (first use in = this > function) > tpupload.c:189: error: expected expression before ')' token > tpupload.c:190: error: 'MYSQL_TYPE_TINY' undeclared (first use in this > function) > tpupload.c:190: error: expected expression before ')' token > tpupload.c:191: error: expected expression before ')' token > tpupload.c:194: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c:203: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c: In function 'upload_init': > tpupload.c:317: error: 'mysql' undeclared (first use in this function) > tpupload.c:322: error: 'stmt' undeclared (first use in this function) > tpupload.c:323: warning: implicit declaration of function > 'mysql_stmt_close' > tpupload.c:324: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c: In function 'upload_progress': > tpupload.c:347: error: 'mysql' undeclared (first use in this function) > tpupload.c: In function 'upload_complete': > tpupload.c:373: error: 'mysql' undeclared (first use in this function) > tpupload.c: In function 'main': > tpupload.c:528: error: 'mysql' undeclared (first use in this function) > tpupload.c:528: warning: implicit declaration of function 'mysql_init' > tpupload.c:530: warning: implicit declaration of function > 'mysql_real_connect' > tpupload.c:532: warning: format '%s' expects type 'char *', but = argument 4 > has type 'int' > tpupload.c:553: error: 'stmt' undeclared (first use in this function) > tpupload.c:556: warning: implicit declaration of function = 'mysql_close' > make[1]: *** [tpupload.o] Error 1 > make[1]: Leaving directory = `/build/filedrawers/upload' > make: *** [upload/upload.cgi] Error 2 > > -Daniel > > > -----Original Message----- > From: Aaron Knister [mailto:aaronk@umbc.edu ] > Sent: Mon 8/15/2011 12:32 PM > To: Colson, Daniel > Cc: openafs-info@openafs.org > Subject: Re: [OpenAFS] 64-bit mod_waklog / filedrawers on afs > > Hi Daniel, > > While I'm not running filedrawers I am using mod_waklog, albeit with = RHEL6. > I had to apply some patches to get it to build correctly. If you send = your > build errors I can see if my patches will help. > > -Aaron > > On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel = wrote: > > > ** > > > > Is anyone else running mod_waklog / filedrawers on afs on a 64-bit > machine? > > I've been unable to compile mod_waklog (an essential component of > > filedrawers) on 64-bit (RHEL5.6). > > > > > > -- > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu > > --=20 Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu ------_=_NextPart_001_01CC60D3.867762B1 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable RE: [OpenAFS] 64-bit mod_waklog / filedrawers on afs

Oops. I apologize. I sent the output from the = filedrawers build instead of the mod_waklog build. I'll check on = mysql-devel though. Below is the output from "make" for = mod_waklog.


`/usr/sbin/apxs -q CC` -I/usr/include/ -I/usr/include/afs = -I/usr/include/httpd -DEAPI -g `/usr/sbin/apxs -q CFLAGS` -c = mod_waklog.c -o mod_waklog.o
In file included from /usr/include/httpd/httpd.h:43,
            &= nbsp;    from mod_waklog.c:4:
/usr/include/httpd/ap_config.h:25:17: error: apr.h: No such file or = directory
/usr/include/httpd/ap_config.h:26:23: error: apr_hooks.h: No such file = or directory
/usr/include/httpd/ap_config.h:27:32: error: apr_optional_hooks.h: No = such file or directory
In file included from /usr/include/httpd/httpd.h:46,
            &= nbsp;    from mod_waklog.c:4:
/usr/include/httpd/ap_release.h:25:41: error: apr_general.h: No such = file or directory
In file included from mod_waklog.c:4:
/usr/include/httpd/httpd.h:50:24: error: apr_tables.h: No such file or = directory
/usr/include/httpd/httpd.h:51:23: error: apr_pools.h: No such file or = directory
/usr/include/httpd/httpd.h:52:22: error: apr_time.h: No such file or = directory
/usr/include/httpd/httpd.h:53:28: error: apr_network_io.h: No such file = or directory
/usr/include/httpd/httpd.h:54:25: error: apr_buckets.h: No such file or = directory
/usr/include/httpd/httpd.h:55:22: error: apr_poll.h: No such file or = directory
In file included from /usr/include/httpd/httpd.h:59,
            &= nbsp;    from mod_waklog.c:4:
/usr/include/httpd/ap_regex.h:87: error: expected = specifier-qualifier-list before ‘apr_size_t’
/usr/include/httpd/ap_regex.h:118: error: expected declaration = specifiers or ‘...’ before ‘apr_size_t’
/usr/include/httpd/ap_regex.h:127: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘ap_regerror’
In file included from mod_waklog.c:4:
/usr/include/httpd/httpd.h:433: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:601: error: expected specifier-qualifier-list = before ‘apr_int64_t’
/usr/include/httpd/httpd.h:723:21: error: apr_uri.h: No such file or = directory
/usr/include/httpd/httpd.h:730: error: expected specifier-qualifier-list = before ‘apr_pool_t’
/usr/include/httpd/httpd.h:746: error: expected specifier-qualifier-list = before ‘apr_pool_t’
/usr/include/httpd/httpd.h:1008: error: expected = specifier-qualifier-list before ‘apr_pool_t’
/usr/include/httpd/httpd.h:1085: error: expected = specifier-qualifier-list before ‘APR_RING_ENTRY’
/usr/include/httpd/httpd.h:1118: error: expected = specifier-qualifier-list before ‘apr_sockaddr_t’
/usr/include/httpd/httpd.h:1146: error: expected = specifier-qualifier-list before ‘apr_port_t’
/usr/include/httpd/httpd.h:1202: error: expected = specifier-qualifier-list before ‘apr_bucket_brigade’
/usr/include/httpd/httpd.h:1210: error: expected = specifier-qualifier-list before ‘apr_bucket_brigade’
/usr/include/httpd/httpd.h:1216: error: expected = specifier-qualifier-list before ‘apr_socket_t’
/usr/include/httpd/httpd.h:1232: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1242: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1255: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1265: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1274: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1284: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1294: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1306: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1314: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1323: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1334: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1357: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1367: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1379: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1388: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1397: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1439: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1449: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1460: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1468: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1477: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘ap_escape_errorlog_item’
/usr/include/httpd/httpd.h:1488: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1497: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1526: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1540: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1550: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1602: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1610: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1624: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1632: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1645: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1685: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1697: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1707: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1715: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/httpd.h:1726: error: expected ‘)’ before = ‘*’ token
In file included from mod_waklog.c:5:
/usr/include/httpd/http_config.h:279: error: expected = specifier-qualifier-list before ‘apr_int64_t’
/usr/include/httpd/http_config.h:357: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/http_config.h:365: error: expected ‘;’ = before ‘void’
/usr/include/httpd/http_config.h:557: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/http_config.h:566: error: expected declaration = specifiers or ‘...’ before ‘apr_pool_t’
/usr/include/httpd/http_config.h:582: error: expected declaration = specifiers or ‘...’ before ‘apr_pool_t’
/usr/include/httpd/http_config.h:607: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘ap_pcfg_openfile’
/usr/include/httpd/http_config.h:619: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/http_config.h:669: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/http_config.h:685: error: expected declaration = specifiers or ‘...’ before ‘apr_pool_t’
/usr/include/httpd/http_config.h:686: error: expected declaration = specifiers or ‘...’ before ‘apr_pool_t’
/usr/include/httpd/http_config.h:975: error: expected ‘)’ = before ‘int’
In file included from mod_waklog.c:6:
/usr/include/httpd/http_log.h:33:29: error: apr_thread_proc.h: No such = file or directory
In file included from /usr/include/syslog.h:1,
            &= nbsp;    from /usr/include/httpd/http_log.h:36,
            &= nbsp;    from mod_waklog.c:6:
/usr/include/sys/syslog.h:200: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
In file included from /usr/include/sys/syslog.h:207,
            &= nbsp;    from /usr/include/syslog.h:1,
            &= nbsp;    from /usr/include/httpd/http_log.h:36,
            &= nbsp;    from mod_waklog.c:6:
/usr/include/bits/syslog.h:34: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
In file included from mod_waklog.c:6:
/usr/include/httpd/http_log.h:102: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/http_log.h:109: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘ap_replace_stderr_log’
/usr/include/httpd/http_log.h:121: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/http_log.h:172: error: expected declaration = specifiers or ‘...’ before ‘apr_status_t’
/usr/include/httpd/http_log.h:174: error: format string argument not a = string type
/usr/include/httpd/http_log.h:196: error: expected declaration = specifiers or ‘...’ before ‘apr_status_t’
/usr/include/httpd/http_log.h:196: error: expected declaration = specifiers or ‘...’ before ‘apr_pool_t’
/usr/include/httpd/http_log.h:198: error: format string argument not a = string type
/usr/include/httpd/http_log.h:220: error: expected declaration = specifiers or ‘...’ before ‘apr_status_t’
/usr/include/httpd/http_log.h:222: error: format string argument not a = string type
/usr/include/httpd/http_log.h:246: error: expected declaration = specifiers or ‘...’ before ‘apr_status_t’
/usr/include/httpd/http_log.h:248: error: format string argument not a = string type
/usr/include/httpd/http_log.h:261: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/http_log.h:269: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘ap_read_pid’
/usr/include/httpd/http_log.h:282: error: expected = specifier-qualifier-list before ‘apr_pool_t’
/usr/include/httpd/http_log.h:301: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/http_log.h:335: error: expected ‘)’ before = ‘void’
In file included from mod_waklog.c:7:
/usr/include/httpd/http_protocol.h:31:26: error: apr_portable.h: No such = file or directory
/usr/include/httpd/http_protocol.h:32:22: error: apr_mmap.h: No such = file or directory
In file included from /usr/include/httpd/http_protocol.h:34,
            &= nbsp;    from mod_waklog.c:7:
/usr/include/httpd/util_filter.h:72: warning: data definition has no = type or storage class
/usr/include/httpd/util_filter.h:72: warning: type defaults to = ‘int’ in declaration of ‘ap_input_mode_t’
/usr/include/httpd/util_filter.h:136: error: expected declaration = specifiers or ‘...’ before ‘*’ token
/usr/include/httpd/util_filter.h:137: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/util_filter.h:137: warning: type defaults to = ‘int’ in declaration of ‘apr_status_t’
/usr/include/httpd/util_filter.h:137: error: ‘apr_status_t’ = declared as function returning a function
/usr/include/httpd/util_filter.h:139: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/util_filter.h:140: error: expected declaration = specifiers or ‘...’ before ‘ap_input_mode_t’
/usr/include/httpd/util_filter.h:141: error: expected declaration = specifiers or ‘...’ before ‘apr_read_type_e’
/usr/include/httpd/util_filter.h:142: error: expected declaration = specifiers or ‘...’ before ‘apr_off_t’
/usr/include/httpd/util_filter.h:142: error: = ‘ap_in_filter_func’ declared as function returning a = function
/usr/include/httpd/util_filter.h:146: error: expected = specifier-qualifier-list before ‘ap_out_filter_func’
/usr/include/httpd/util_filter.h:299: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/util_filter.h:300: error: expected declaration = specifiers or ‘...’ before ‘ap_input_mode_t’
/usr/include/httpd/util_filter.h:301: error: expected declaration = specifiers or ‘...’ before ‘apr_read_type_e’
/usr/include/httpd/util_filter.h:302: error: expected declaration = specifiers or ‘...’ before ‘apr_off_t’
/usr/include/httpd/util_filter.h:302: error: ‘ap_get_brigade’ = declared as function returning a function
/usr/include/httpd/util_filter.h:313: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/util_filter.h:313: error: ‘ap_pass_brigade’ = declared as function returning a function
/usr/include/httpd/util_filter.h:351: error: expected declaration = specifiers or ‘...’ before ‘ap_out_filter_func’
/usr/include/httpd/util_filter.h:377: error: expected declaration = specifiers or ‘...’ before ‘ap_out_filter_func’
/usr/include/httpd/util_filter.h:495: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/util_filter.h:496: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/util_filter.h:496: error: expected declaration = specifiers or ‘...’ before ‘apr_pool_t’
/usr/include/httpd/util_filter.h:496: error: ‘ap_save_brigade’ = declared as function returning a function
/usr/include/httpd/util_filter.h:506: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/util_filter.h:514: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/util_filter.h:514: error: ‘ap_fflush’ = declared as function returning a function
/usr/include/httpd/util_filter.h:551: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/util_filter.h:552: error: ‘ap_fputstrs’ = declared as function returning a function
/usr/include/httpd/util_filter.h:562: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/util_filter.h:565: error: ‘ap_fprintf’ = declared as function returning a function
/usr/include/httpd/util_filter.h:565: error: format string argument not = a string type
In file included from mod_waklog.c:7:
/usr/include/httpd/http_protocol.h:45: error: expected ‘)’ = before ‘void’
/usr/include/httpd/http_protocol.h:77: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/http_protocol.h:111: error: expected declaration = specifiers or ‘...’ before ‘apr_off_t’
/usr/include/httpd/http_protocol.h:127: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘ap_rationalize_mtime’
/usr/include/httpd/http_protocol.h:203: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/http_protocol.h:228: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/http_protocol.h:235: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/http_protocol.h:253: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/http_protocol.h:356: error: expected declaration = specifiers or ‘...’ before ‘va_list’
/usr/include/httpd/http_protocol.h:426: error: expected declaration = specifiers or ‘...’ before ‘apr_size_t’
/usr/include/httpd/http_protocol.h:534: error: expected declaration = specifiers or ‘...’ before ‘apr_size_t’
/usr/include/httpd/http_protocol.h:535: error: expected declaration = specifiers or ‘...’ before ‘apr_size_t’
/usr/include/httpd/http_protocol.h:537: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/http_protocol.h:537: error: = ‘ap_rgetline_core’ declared as function returning a = function
/usr/include/httpd/http_protocol.h:554: error: expected ‘)’ = before ‘*’ token
/usr/include/httpd/http_protocol.h:569: error: expected ‘)’ = before ‘int’
/usr/include/httpd/http_protocol.h:608: error: expected = specifier-qualifier-list before ‘apr_bucket_refcount’
/usr/include/httpd/http_protocol.h:616: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘ap_bucket_type_error’
/usr/include/httpd/http_protocol.h:633: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘*’ token
/usr/include/httpd/http_protocol.h:644: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘*’ token
/usr/include/httpd/http_protocol.h:648: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/http_protocol.h:648: error: = ‘ap_byterange_filter’ declared as function returning a = function
/usr/include/httpd/http_protocol.h:649: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/http_protocol.h:649: error: = ‘ap_http_header_filter’ declared as function returning a = function
/usr/include/httpd/http_protocol.h:651: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/http_protocol.h:651: error: = ‘ap_content_length_filter’ declared as function returning a = function
/usr/include/httpd/http_protocol.h:652: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/http_protocol.h:652: error: = ‘ap_old_write_filter’ declared as function returning a = function
In file included from mod_waklog.c:8:
/usr/include/httpd/http_request.h:105: warning: type defaults to = ‘int’ in declaration of ‘apr_finfo_t’
/usr/include/httpd/http_request.h:105: error: expected ‘;’, = ‘,’ or ‘)’ before ‘*’ token
/usr/include/httpd/http_request.h:131: error: expected declaration = specifiers or ‘...’ before ‘apr_bucket_brigade’
/usr/include/httpd/http_request.h:131: error: = ‘ap_sub_req_output_filter’ declared as function returning a = function
/usr/include/httpd/http_request.h:198: error: expected declaration = specifiers or ‘...’ before ‘apr_time_t’
/usr/include/httpd/http_request.h:261: error: expected ‘)’ = before ‘int’
In file included from mod_waklog.c:9:
/usr/include/httpd/http_core.h:30:22: error: apr_hash.h: No such file or = directory
/usr/include/httpd/http_core.h:31:26: error: apr_optional.h: No such = file or directory
In file included from mod_waklog.c:9:
/usr/include/httpd/http_core.h:221: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/http_core.h:235: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘ap_get_server_port’
/usr/include/httpd/http_core.h:242: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘ap_get_limit_req_body’
/usr/include/httpd/http_core.h:291: error: expected = specifier-qualifier-list before ‘apr_int64_t’
/usr/include/httpd/http_core.h:327: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘*’ token
/usr/include/httpd/http_core.h:646: error: expected = specifier-qualifier-list before ‘apr_hash_t’
/usr/include/httpd/http_core.h:675: error: expected ‘)’ before = ‘int’
/usr/include/httpd/http_core.h:693: error: expected declaration = specifiers or ‘...’ before ‘ap_ident_lookup’
/usr/include/httpd/http_core.h:694: error: expected declaration = specifiers or ‘...’ before ‘(’ token
/usr/include/httpd/http_core.h:694: warning: data definition has no type = or storage class
/usr/include/httpd/http_core.h:694: warning: type defaults to = ‘int’ in declaration of = ‘APR_DECLARE_OPTIONAL_FN’
mod_waklog.c:49:25: error: apr_strings.h: No such file or directory
mod_waklog.c:50:24: error: apr_base64.h: No such file or directory
In file included from /usr/include/httpd/unixd.h:30,
            &= nbsp;    from mod_waklog.c:57:
/usr/include/httpd/ap_listen.h:34: error: expected declaration = specifiers or ‘...’ before ‘apr_pool_t’
/usr/include/httpd/ap_listen.h:34: error: ‘accept_function’ = declared as function returning a function
/usr/include/httpd/ap_listen.h:50: error: expected = specifier-qualifier-list before ‘apr_socket_t’
In file included from mod_waklog.c:57:
/usr/include/httpd/unixd.h:39:28: error: apr_proc_mutex.h: No such file = or directory
/usr/include/httpd/unixd.h:40:30: error: apr_global_mutex.h: No such = file or directory
In file included from mod_waklog.c:57:
/usr/include/httpd/unixd.h:57: error: expected ‘)’ before = ‘ap_unix_identity_t’
/usr/include/httpd/unixd.h:75: warning: data definition has no type or = storage class
/usr/include/httpd/unixd.h:75: warning: type defaults to ‘int’ = in declaration of ‘unixd_config_rec’
/usr/include/httpd/unixd.h:76: error: expected ‘=3D’, = ‘,’, ‘;’, ‘asm’ or = ‘__attribute__’ before ‘unixd_config’
/usr/include/httpd/unixd.h:79: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/unixd.h:98: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/unixd.h:99: error: expected ‘)’ before = ‘*’ token
/usr/include/httpd/unixd.h:100: error: expected declaration specifiers = or ‘...’ before ‘apr_pool_t’
/usr/include/httpd/unixd.h:100: error: ‘unixd_accept’ declared = as function returning a function
mod_waklog.c:58: error: expected ‘=3D’, ‘,’, = ‘;’, ‘asm’ or ‘__attribute__’ before = ‘unixd_config’
In file included from /usr/include/et/com_err.h:19,
            &= nbsp;    from /usr/include/krb5/krb5.h:2690,
            &= nbsp;    from /usr/include/krb5.h:7,
            &= nbsp;    from mod_waklog.c:73:
/usr/lib/gcc/x86_64-redhat-linux/4.1.2/include/stdarg.h:105: error: = expected ‘=3D’, ‘,’, ‘;’, ‘asm’ = or ‘__attribute__’ before ‘va_list’
In file included from /usr/include/krb5/krb5.h:2690,
            &= nbsp;    from /usr/include/krb5.h:7,
            &= nbsp;    from mod_waklog.c:73:
/usr/include/et/com_err.h:34: error: expected declaration specifiers or = ‘...’ before ‘va_list’
/usr/include/et/com_err.h:38: error: expected declaration specifiers or = ‘...’ before ‘va_list’
/usr/include/et/com_err.h:40: error: expected declaration specifiers or = ‘...’ before ‘va_list’
/usr/include/et/com_err.h:41: error: expected declaration specifiers or = ‘...’ before ‘va_list’
/usr/include/et/com_err.h:43: error: expected declaration specifiers or = ‘...’ before ‘va_list’
In file included from mod_waklog.c:75:
/usr/include/afs/venus.h:33: error: field ‘server’ has = incomplete type
In file included from /usr/include/stdio.h:72,
            &= nbsp;    from /usr/include/rx/rx.h:42,
            &= nbsp;    from /usr/include/rx/rxkad_prototypes.h:15,
            &= nbsp;    from /usr/include/rx/rxkad.h:102,
            &= nbsp;    from /usr/include/afs/auth.h:19,
            &= nbsp;    from mod_waklog.c:76:
/usr/include/libio.h:482: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/libio.h:484: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
In file included from /usr/include/rx/rx.h:42,
            &= nbsp;    from /usr/include/rx/rxkad_prototypes.h:15,
            &= nbsp;    from /usr/include/rx/rxkad.h:102,
            &= nbsp;    from /usr/include/afs/auth.h:19,
            &= nbsp;    from mod_waklog.c:76:
/usr/include/stdio.h:343: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/stdio.h:348: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/stdio.h:351: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/stdio.h:362: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/stdio.h:371: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/stdio.h:387: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/stdio.h:418: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/stdio.h:425: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/stdio.h:430: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
In file included from /usr/include/rx/rx.h:42,
            &= nbsp;    from /usr/include/rx/rxkad_prototypes.h:15,
            &= nbsp;    from /usr/include/rx/rxkad.h:102,
            &= nbsp;    from /usr/include/afs/auth.h:19,
            &= nbsp;    from mod_waklog.c:76:
/usr/include/stdio.h:805: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
In file included from /usr/include/stdio.h:835,
            &= nbsp;    from /usr/include/rx/rx.h:42,
            &= nbsp;    from /usr/include/rx/rxkad_prototypes.h:15,
            &= nbsp;    from /usr/include/rx/rxkad.h:102,
            &= nbsp;    from /usr/include/afs/auth.h:19,
            &= nbsp;    from mod_waklog.c:76:
/usr/include/bits/stdio.h:34: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/bits/stdio.h: In function ‘vprintf’:
/usr/include/bits/stdio.h:36: error: ‘__arg’ undeclared (first = use in this function)
/usr/include/bits/stdio.h:36: error: (Each undeclared identifier is = reported only once
/usr/include/bits/stdio.h:36: error: for each function it appears = in.)
/usr/include/bits/stdio.h:36: error: too many arguments to function = ‘vfprintf’
In file included from /usr/include/stdio.h:838,
            &= nbsp;    from /usr/include/rx/rx.h:42,
            &= nbsp;    from /usr/include/rx/rxkad_prototypes.h:15,
            &= nbsp;    from /usr/include/rx/rxkad.h:102,
            &= nbsp;    from /usr/include/afs/auth.h:19,
            &= nbsp;    from mod_waklog.c:76:
/usr/include/bits/stdio2.h: At top level:
/usr/include/bits/stdio2.h:28: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/bits/stdio2.h:43: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/bits/stdio2.h:60: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
/usr/include/bits/stdio2.h:62: error: expected declaration specifiers or = ‘...’ before ‘__gnuc_va_list’
mod_waklog.c:111: error: expected specifier-qualifier-list before = ‘apr_pool_t’
mod_waklog.c: In function ‘log_error’:
mod_waklog.c:184: error: ‘va_list’ undeclared (first use in = this function)
mod_waklog.c:184: error: expected ‘;’ before = ‘ap’
mod_waklog.c:186: error: ‘ap’ undeclared (first use in this = function)
mod_waklog.c:191: warning: passing argument 4 of = ‘ap_log_error’ makes pointer from integer without a cast
mod_waklog.c:191: warning: passing argument 5 of = ‘ap_log_error’ from incompatible pointer type
mod_waklog.c: In function ‘retrieve_config’:
mod_waklog.c:203: error: ‘request_rec’ has no member named = ‘main’
mod_waklog.c:204: error: ‘request_rec’ has no member named = ‘main’
mod_waklog.c:211: error: ‘request_rec’ has no member named = ‘per_dir_config’
mod_waklog.c:214: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c: In function ‘set_auth’:
mod_waklog.c:285: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:285: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:298: error: ‘server_rec’ has no member named = ‘module_config’
mod_waklog.c:311: error: ‘request_rec’ has no member named = ‘connection’
mod_waklog.c:311: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:313: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:313: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:317: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:317: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:327: warning: implicit declaration of function = ‘apr_table_get’
mod_waklog.c:327: error: ‘request_rec’ has no member named = ‘subprocess_env’
mod_waklog.c:327: warning: cast to pointer from integer of different = size
mod_waklog.c:330: error: ‘request_rec’ has no member named = ‘notes’
mod_waklog.c:330: warning: cast to pointer from integer of different = size
mod_waklog.c:339: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:340: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:340: error: ‘request_rec’ has no member named = ‘user’
mod_waklog.c:703: warning: label ‘usecachedtoken’ defined but = not used
mod_waklog.c: At top level:
mod_waklog.c:749: error: expected ‘)’ before ‘*’ = token
mod_waklog.c:778: error: expected ‘)’ before ‘*’ = token
mod_waklog.c:801: error: expected ‘)’ before ‘*’ = token
mod_waklog.c:831: error: expected ‘)’ before ‘*’ = token
mod_waklog.c: In function ‘set_waklog_enabled’:
mod_waklog.c:870: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:874: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c: At top level:
mod_waklog.c:883: error: expected ‘)’ before ‘*’ = token
mod_waklog.c: In function ‘set_waklog_location_principal’:
mod_waklog.c:912: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:914: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:917: warning: implicit declaration of function = ‘apr_pstrdup’
mod_waklog.c:917: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:917: warning: assignment makes pointer from integer without = a cast
mod_waklog.c:918: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:918: warning: assignment makes pointer from integer without = a cast
mod_waklog.c:920: warning: implicit declaration of function = ‘add_to_renewtable’
mod_waklog.c:920: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c: In function ‘set_waklog_afs_cell’:
mod_waklog.c:932: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:934: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:938: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:938: warning: assignment makes pointer from integer without = a cast
mod_waklog.c:943: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:943: warning: assignment makes pointer from integer without = a cast
mod_waklog.c: In function ‘set_waklog_afs_cell_realm’:
mod_waklog.c:954: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:956: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:959: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:959: warning: assignment makes pointer from integer without = a cast
mod_waklog.c:962: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:962: warning: assignment makes pointer from integer without = a cast
mod_waklog.c: In function ‘set_waklog_default_principal’:
mod_waklog.c:971: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:973: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:975: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:978: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:978: warning: assignment makes pointer from integer without = a cast
mod_waklog.c:979: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:979: warning: assignment makes pointer from integer without = a cast
mod_waklog.c:983: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:983: warning: assignment makes pointer from integer without = a cast
mod_waklog.c:984: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c:984: warning: assignment makes pointer from integer without = a cast
mod_waklog.c:986: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:990: error: ‘cmd_parms’ has no member named = ‘pool’
mod_waklog.c: In function ‘set_waklog_use_usertokens’:
mod_waklog.c:1001: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:1007: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c: In function = ‘set_waklog_disable_token_cache’:
mod_waklog.c:1017: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c:1023: error: ‘cmd_parms’ has no member named = ‘server’
mod_waklog.c: At top level:
mod_waklog.c:1032: error: ‘waklog_child_exit’ declared as = function returning a function
mod_waklog.c: In function ‘waklog_child_exit’:
mod_waklog.c:1053: error: ‘APR_SUCCESS’ undeclared (first use = in this function)
mod_waklog.c: At top level:
mod_waklog.c:1060: error: expected ‘)’ before ‘*’ = token
mod_waklog.c: In function ‘token_cleanup’:
mod_waklog.c:1154: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c: In function ‘waklog_child_routine’:
mod_waklog.c:1173: error: ‘server_rec’ has no member named = ‘module_config’
mod_waklog.c: At top level:
mod_waklog.c:1234: error: expected ‘)’ before ‘*’ = token
mod_waklog.c: In function ‘waklog_phase0’:
mod_waklog.c:1489: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1495: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1496: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1498: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1499: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1507: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c: In function ‘waklog_phase1’:
mod_waklog.c:1518: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1524: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1525: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1527: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1528: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1530: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c: In function ‘waklog_phase3’:
mod_waklog.c:1541: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1547: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1548: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1550: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1551: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1553: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c: In function ‘waklog_phase6’:
mod_waklog.c:1564: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1570: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1571: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1573: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1574: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1576: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c: In function ‘waklog_phase7’:
mod_waklog.c:1588: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1594: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1595: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1597: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1598: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1600: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1601: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1603: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c: In function ‘waklog_phase9’:
mod_waklog.c:1622: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1625: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1628: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1629: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c: In function ‘waklog_new_connection’:
mod_waklog.c:1651: error: ‘conn_rec’ has no member named = ‘base_server’
mod_waklog.c:1654: error: ‘conn_rec’ has no member named = ‘base_server’
mod_waklog.c:1657: error: ‘conn_rec’ has no member named = ‘base_server’
mod_waklog.c:1659: error: ‘conn_rec’ has no member named = ‘base_server’
mod_waklog.c: In function ‘waklog_phase2’:
mod_waklog.c:1684: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1693: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c:1698: error: ‘request_rec’ has no member named = ‘server’
mod_waklog.c: At top level:
mod_waklog.c:1734: error: expected ‘)’ before ‘*’ = token
mod_waklog.c:1751: error: ‘waklog_create_dir_config’ = undeclared here (not in a function)
mod_waklog.c:1751: warning: excess elements in struct initializer
mod_waklog.c:1751: warning: (near initialization for = ‘waklog_module’)
mod_waklog.c:1752: error: ‘waklog_merge_dir_config’ undeclared = here (not in a function)
mod_waklog.c:1752: warning: excess elements in struct initializer
mod_waklog.c:1752: warning: (near initialization for = ‘waklog_module’)
mod_waklog.c:1753: error: ‘waklog_create_server_config’ = undeclared here (not in a function)
mod_waklog.c:1753: warning: excess elements in struct initializer
mod_waklog.c:1753: warning: (near initialization for = ‘waklog_module’)
mod_waklog.c:1754: warning: excess elements in struct initializer
mod_waklog.c:1754: warning: (near initialization for = ‘waklog_module’)
mod_waklog.c:1755: warning: excess elements in struct initializer
mod_waklog.c:1755: warning: (near initialization for = ‘waklog_module’)
mod_waklog.c:1756: error: ‘waklog_register_hooks’ undeclared = here (not in a function)
mod_waklog.c:1757: warning: excess elements in struct initializer
mod_waklog.c:1757: warning: (near initialization for = ‘waklog_module’)
make: *** [mod_waklog.o] Error 1


Thanks,
Daniel


-----Original Message-----
From: Aaron Knister [mailto:aaronk@umbc.edu]
Sent: Wed 8/17/2011 10:59 AM
To: Colson, Daniel
Cc: openafs-info@openafs.org
Subject: Re: [OpenAFS] 64-bit mod_waklog / filedrawers on afs

Hi Daniel,

Well... I've never built filedrawers, but this line:

tpupload.c:18:19: error: mysql.h: No such file or directory

suggests you're missing the mysql headers. If its redhat try a yum = install
mysql-devel.

-Aaron

On Tue, Aug 16, 2011 at 4:50 PM, Colson, Daniel = <dcolson5@uncc.edu> wrote:

> **
>
> Thanks. Any help is appreciated. Here's the error output from = "make".
>
> make[1]: Entering directory = `<path-to-build-dir>/build/filedrawers/libcgi'
> gcc -Wall -Wstrict-prototypes -g  -fPIC = -I/usr/include/apr-1  -c -o cgi.o
> cgi.c
> cgi.c: In function 'post_getline':
> cgi.c:91: warning: function declaration isn't a prototype
> cgi.c: In function 'cgi_init':
> cgi.c:148: warning: function declaration isn't a prototype
> cgi.c: In function 'mp_get_file':
> cgi.c:361: warning: function declaration isn't a prototype
> cgi.c:364: warning: function declaration isn't a prototype
> cgi.c:375: warning: function declaration isn't a prototype
> cgi.c:378: warning: function declaration isn't a prototype
> cgi.c:385: warning: function declaration isn't a prototype
> cgi.c:388: warning: function declaration isn't a prototype
> cgi.c:400: warning: function declaration isn't a prototype
> cgi.c:420: warning: function declaration isn't a prototype
> cgi.c:424: warning: function declaration isn't a prototype
> cgi.c: In function 'cgi_multipart':
> cgi.c:539: warning: function declaration isn't a prototype
> cgi.c:614: warning: function declaration isn't a prototype
> building normal libcgi.a
> make[1]: Leaving directory = `<path-to-build-dir>/build/filedrawers/libcgi'
> ( cd upload; make  )
> make[1]: Entering directory = `<path-to-build-dir>/build/filedrawers/upload'
> gcc -Wall -Wmissing-prototypes -g -O2 -I../libcgi = -I/usr/include/mysql  -g
> -pipe -Wp,-D_FORTIFY_SOURCE=3D2 -fexceptions -fstack-protector
> --param=3Dssp-buffer-size=3D4 -m64 -D_GNU_SOURCE = -D_FILE_OFFSET_BITS=3D64
> -D_LARGEFILE_SOURCE -fno-strict-aliasing -fwrapv -Wall = -Wmissing-prototypes
> \
>           &nb= sp;     -D_URL_ERROR_PAGE=3D\"https://<site>?error=3Dtrue\" \
>
> = -D_FILEDRAWERS_CONFIG_PATH=3D\"/var/www/filedrawers/lib/filedrawers.= conf\" \
>           &nb= sp;     -c tpupload.c
> tpupload.c:18:19: error: mysql.h: No such file or directory
> tpupload.c:34: error: expected '=3D', ',', ';', 'asm' or = '__attribute__'
> before '*' token
> tpupload.c:35: error: expected '=3D', ',', ';', 'asm' or = '__attribute__'
> before '*' token
> tpupload.c: In function 'upload_progress_db_insert':
> tpupload.c:103: error: 'MYSQL_BIND' undeclared (first use in this = function)
> tpupload.c:103: error: (Each undeclared identifier is reported only = once
> tpupload.c:103: error: for each function it appears in.)
> tpupload.c:103: error: expected ';' before 'bind'
> tpupload.c:104: error: 'my_bool' undeclared (first use in this = function)
> tpupload.c:104: error: expected ';' before 'is_null'
> tpupload.c:110: error: 'stmt' undeclared (first use in this = function)
> tpupload.c:111: warning: implicit declaration of function = 'mysql_stmt_init'
> tpupload.c:111: error: 'mysql' undeclared (first use in this = function)
> tpupload.c:112: warning: implicit declaration of function = 'mysql_error'
> tpupload.c:112: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c:123: warning: implicit declaration of function
> 'mysql_stmt_prepare'
> tpupload.c:124: warning: implicit declaration of function
> 'mysql_stmt_error'
> tpupload.c:124: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c:130: error: 'bind' undeclared (first use in this = function)
> tpupload.c:130: warning: left-hand operand of comma expression has = no
> effect
> tpupload.c:132: error: 'MYSQL_TYPE_STRING' undeclared (first use in = this
> function)
> tpupload.c:132: error: expected expression before ')' token
> tpupload.c:132: error: 'is_null' undeclared (first use in this = function)
> tpupload.c:133: error: expected expression before ')' token
> tpupload.c:134: error: 'MYSQL_TYPE_LONGLONG' undeclared (first use = in this
> function)
> tpupload.c:134: error: expected expression before ')' token
> tpupload.c:135: error: expected expression before ')' token
> tpupload.c:136: error: 'MYSQL_TYPE_TINY' undeclared (first use in = this
> function)
> tpupload.c:136: error: expected expression before ')' token
> tpupload.c:138: warning: implicit declaration of function
> 'mysql_stmt_bind_param'
> tpupload.c:139: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c:148: warning: implicit declaration of function
> 'mysql_stmt_execute'
> tpupload.c:149: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c: In function 'upload_progress_db_update':
> tpupload.c:160: error: 'MYSQL_BIND' undeclared (first use in this = function)
> tpupload.c:160: error: expected ';' before 'bind'
> tpupload.c:161: error: 'my_bool' undeclared (first use in this = function)
> tpupload.c:161: error: expected ';' before 'is_null'
> tpupload.c:166: error: 'stmt' undeclared (first use in this = function)
> tpupload.c:167: error: 'mysql' undeclared (first use in this = function)
> tpupload.c:168: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c:180: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c:186: error: 'bind' undeclared (first use in this = function)
> tpupload.c:186: warning: left-hand operand of comma expression has = no
> effect
> tpupload.c:188: error: 'MYSQL_TYPE_STRING' undeclared (first use in = this
> function)
> tpupload.c:188: error: expected expression before ')' token
> tpupload.c:188: error: 'is_null' undeclared (first use in this = function)
> tpupload.c:189: error: 'MYSQL_TYPE_LONGLONG' undeclared (first use = in this
> function)
> tpupload.c:189: error: expected expression before ')' token
> tpupload.c:190: error: 'MYSQL_TYPE_TINY' undeclared (first use in = this
> function)
> tpupload.c:190: error: expected expression before ')' token
> tpupload.c:191: error: expected expression before ')' token
> tpupload.c:194: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c:203: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c: In function 'upload_init':
> tpupload.c:317: error: 'mysql' undeclared (first use in this = function)
> tpupload.c:322: error: 'stmt' undeclared (first use in this = function)
> tpupload.c:323: warning: implicit declaration of function
> 'mysql_stmt_close'
> tpupload.c:324: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c: In function 'upload_progress':
> tpupload.c:347: error: 'mysql' undeclared (first use in this = function)
> tpupload.c: In function 'upload_complete':
> tpupload.c:373: error: 'mysql' undeclared (first use in this = function)
> tpupload.c: In function 'main':
> tpupload.c:528: error: 'mysql' undeclared (first use in this = function)
> tpupload.c:528: warning: implicit declaration of function = 'mysql_init'
> tpupload.c:530: warning: implicit declaration of function
> 'mysql_real_connect'
> tpupload.c:532: warning: format '%s' expects type 'char *', but = argument 4
> has type 'int'
> tpupload.c:553: error: 'stmt' undeclared (first use in this = function)
> tpupload.c:556: warning: implicit declaration of function = 'mysql_close'
> make[1]: *** [tpupload.o] Error 1
> make[1]: Leaving directory = `<path-to-build-dir>/build/filedrawers/upload'
> make: *** [upload/upload.cgi] Error 2
>
> -Daniel
>
>
> -----Original Message-----
> From: Aaron Knister [mailto:aaronk@umbc.edu = <aaronk@umbc.edu>]
> Sent: Mon 8/15/2011 12:32 PM
> To: Colson, Daniel
> Cc: openafs-info@openafs.org
> Subject: Re: [OpenAFS] 64-bit mod_waklog / filedrawers on afs
>
> Hi Daniel,
>
> While I'm not running filedrawers I am using mod_waklog, albeit = with RHEL6.
> I had to apply some patches to get it to build correctly. If you = send your
> build errors I can see if my patches will help.
>
> -Aaron
>
> On Wed, Aug 10, 2011 at 3:16 PM, Colson, Daniel = <dcolson5@uncc.edu> wrote:
>
> > **
> >
> > Is anyone else running mod_waklog / filedrawers on afs on a = 64-bit
> machine?
> > I've been unable to compile mod_waklog (an essential component = of
> > filedrawers) on 64-bit (RHEL5.6).
> >
>
>
>
> --
> Aaron Knister
> Systems Administrator
> Division of Information Technology
> University of Maryland, Baltimore County
> aaronk@umbc.edu
>
>


--
Aaron Knister
Systems Administrator
Division of Information Technology
University of Maryland, Baltimore County
aaronk@umbc.edu

------_=_NextPart_001_01CC60D3.867762B1-- From jblaine@kickflop.net Wed Aug 24 15:20:54 2011 From: jblaine@kickflop.net (Jeff Blaine) Date: Wed, 24 Aug 2011 10:20:54 -0400 Subject: [OpenAFS] Re: Performance issues In-Reply-To: <4E54FDE1.9030803@secure-endpoints.com> References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> <20110824050320.15532dpqf4whury8@bitis.umrk.nl> <20110824151734.2108154adgcyaxgg@bitis.umrk.nl> <4E54FDE1.9030803@secure-endpoints.com> Message-ID: <4E5508C6.1090404@kickflop.net> > For read/write data, if the cache is too small, the cache manager is > required to flush data to the file server sooner than it would prefer. > Since many files used today are in the GB range, it is not unusual to > have caches sizes of 10GB to 20GB. The local disk is cheap; network > bandwidth is not. http://wiki.openafs.org/ "This page is in Indonesion Would you like to translate it?" No? Continuing... [ Page displays in what I assume to be Indonesian ] Search: cache Click 1st link "ConfiguringTheCache" http://openafs-wiki.stanford.edu/AFSLore/ConfiguringtheCache/ "Machines serving multiple users usually perform better with a cache of at least 60 to 70 MB." From sxw@inf.ed.ac.uk Wed Aug 24 15:29:53 2011 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Wed, 24 Aug 2011 15:29:53 +0100 Subject: [OpenAFS] Re: Performance issues In-Reply-To: <4E5508C6.1090404@kickflop.net> References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> <20110824050320.15532dpqf4whury8@bitis.umrk.nl> <20110824151734.2108154adgcyaxgg@bitis.umrk.nl> <4E54FDE1.9030803@secure-endpoints.com> <4E5508C6.1090404@kickflop.net> Message-ID: <9DF33F3F-8BCC-4283-B27F-FA4FE7B0330F@inf.ed.ac.uk> It's a wiki. You can edit it. S. Sent from my iPhone On 24 Aug 2011, at 15:20, Jeff Blaine wrote: >> For read/write data, if the cache is too small, the cache manager is >> required to flush data to the file server sooner than it would prefer. >> Since many files used today are in the GB range, it is not unusual to >> have caches sizes of 10GB to 20GB. The local disk is cheap; network >> bandwidth is not. > > http://wiki.openafs.org/ > > "This page is in Indonesion Would you like to translate it?" > > No? > > Continuing... > > [ Page displays in what I assume to be Indonesian ] > > Search: cache > > Click 1st link "ConfiguringTheCache" > > http://openafs-wiki.stanford.edu/AFSLore/ConfiguringtheCache/ > > "Machines serving multiple users usually perform better with > a cache of at least 60 to 70 MB." > _______________________________________________ > OpenAFS-info mailing list > OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > From adeason@sinenomine.net Wed Aug 24 15:55:03 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Wed, 24 Aug 2011 09:55:03 -0500 Subject: [OpenAFS] Re: Install OpenAFS on ppc64? References: <1313988591.6154.17.camel@yentj.infowrap.com.tw> <20110822130833.6109e6ec@thirdoffive.cmf.nrl.navy.mil> <1314060971.6543.10.camel@yentj.infowrap.com.tw> <20110823145025.50a629db@thirdoffive.cmf.nrl.navy.mil> <4E54A6D4.2050108@infowrap.com.tw> Message-ID: <20110824095503.e97cd8ed.adeason@sinenomine.net> On Wed, 24 Aug 2011 15:23:00 +0800 Ting-jen Yen wrote: > Hello, > > Yes, I can see the file /proc/fs/openafs/afs_ioctl. > For some reason, the system became unstable after I browsed > this directory. (Could be coincident. I don't know.) Can you 'strace -f' afsd, and show what happens when it tries to call ioctl() on that path? (you should see an open for that path immediately followed by an ioctl) Can you give the compiler errors that occurred before you modified the source? (Give the full command run followed by the errors.) There might be some kind of discrepancy with 64/32-bitness involved, but even if that's not the problem, obviously we shouldn't fail to compile. -- Andrew Deason adeason@sinenomine.net From adeason@sinenomine.net Wed Aug 24 15:56:10 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Wed, 24 Aug 2011 09:56:10 -0500 Subject: [OpenAFS] Re: Problems to move OpenAFS volumes References: Message-ID: <20110824095610.b056354a.adeason@sinenomine.net> On Thu, 18 Aug 2011 10:48:08 -0300 vitor lima wrote: > I want to move the volumes from server0 to server1, so I execute > > vos move -id user_x -fromserver server0 -frompartition /vicepb -toserver > server1 -topartition /vicepa > > and the volume is moved with successful. What does "vos" say? What is the error message given? You may want to run with -verbose to get more details. -- Andrew Deason adeason@sinenomine.net From adeason@sinenomine.net Wed Aug 24 16:34:59 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Wed, 24 Aug 2011 10:34:59 -0500 Subject: [OpenAFS] Re: 64-bit mod_waklog / filedrawers on afs References: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBE2@EXEVS04.its.uncc.edu> <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBE3@EXEVS04.its.uncc.edu> Message-ID: <20110824103459.8b0a3867.adeason@sinenomine.net> On Mon, 22 Aug 2011 09:54:16 -0400 "Colson, Daniel" wrote: > Oops. I apologize. I sent the output from the filedrawers build > instead of the mod_waklog build. I'll check on mysql-devel though. > Below is the output from "make" for mod_waklog. > > > `/usr/sbin/apxs -q CC` -I/usr/include/ -I/usr/include/afs -I/usr/include/httpd -DEAPI -g `/usr/sbin/apxs -q CFLAGS` -c mod_waklog.c -o mod_waklog.o Did you try using the patches Aaron mentioned earlier in the thread? I'm not too familiar with mod_waklog, but this looks like a build line without Aaron's fix_build_scripts.patch. -- Andrew Deason adeason@sinenomine.net From adeason@sinenomine.net Wed Aug 24 16:43:58 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Wed, 24 Aug 2011 10:43:58 -0500 Subject: [OpenAFS] Re: Performance issues References: <20110817230344.13043n7pa6p469c0@bitis.umrk.nl> <20110822233311.d7186dd9834e91a70f6dc556@sinenomine.net> <20110824011004.14312cbdnpqhew5c@bitis.umrk.nl> Message-ID: <20110824104358.e57447ba.adeason@sinenomine.net> On Wed, 24 Aug 2011 01:10:04 +0200 Jaap Winius wrote: > > ... I would guess that this has more to do with the bandwidth and > > latency, but unless there are more dropped or reordered packets or > > something, I'm not sure either way what we would see from the > > OpenAFS fileserver point of view besides worse bandwidth/latency. > > After both the fs and db servers were moved from the virtual hosts > (and from behind the firewall and NAT) to the bare metal, we saw a > marked improvement in local performance. For example, certain > applications, particularly the Iceweasel (Firefox) browser, only > became usable after the fs server was moved. But as I said, after the > changes we've seen no performance improvement for when the user > volumes are accessed remotely. Yes... and that suggests a change in bandwidth and/or latency (or packets getting dropped, getting reordered, etc etc) Whether that is caused by the number of firewalls or whatever or by the characteristics of the connection... to OpenAFS it is the same. What I mean is, this isn't OpenAFS specific. You can just measure the differences in throughput and delays of UDP packets to see if there's any differences in the setup. For the purposes of sustained throughput, we have no idea what routers, firewalls, NATs, etc that we're going through, so it doesn't make a difference to us. -- Andrew Deason adeason@sinenomine.net From mmeffie@sinenomine.net Wed Aug 24 17:04:20 2011 From: mmeffie@sinenomine.net (Michael Meffie) Date: Wed, 24 Aug 2011 12:04:20 -0400 Subject: [OpenAFS] Home not found In-Reply-To: References: Message-ID: <4E552104.5050003@sinenomine.net> vitor lima wrote: > Hi! > > I have a problem with my OpenAFS server: > > When a user do the "log in", the home directory is not found and the > user logs using the root (the / ). > > What can be this? > Somebody can help me? Hello Vitor, Is this a new OpenAFS installation? Has this worked before? You are going to need to provide some more information before we can give you more guidance. Firstly, is the OpenAFS client running? What is the output of the mount command? If /afs is mounted, does the user have a token? How are you configuring the client's cell information; are you using DNS or a local CellServDB? Mike -- From vittorlima@gmail.com Wed Aug 24 18:16:40 2011 From: vittorlima@gmail.com (vitor lima) Date: Wed, 24 Aug 2011 14:16:40 -0300 Subject: [OpenAFS] Home not found In-Reply-To: <4E552104.5050003@sinenomine.net> References: <4E552104.5050003@sinenomine.net> Message-ID: --0016e64c1af6b7acf404ab4379d7 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hello! Sorry for the absence of information. This OpenAFS installation is running in the server1 and I have a OpenAFS installation working well in the server0. I thought the server1's OpenAFS was working, but when I stop the bosserver of the server0, that problem begins to occur. Oh, the OpenAFS client is running and I am using the CellServDB. 2011/8/24 Michael Meffie > vitor lima wrote: > >> Hi! >> >> I have a problem with my OpenAFS server: >> >> When a user do the "log in", the home directory is not found and the use= r >> logs using the root (the / ). >> >> What can be this? >> Somebody can help me? >> > > Hello Vitor, > > Is this a new OpenAFS installation? Has this worked before? You are > going to need to provide some more information before we can give > you more guidance. Firstly, is the OpenAFS client running? What > is the output of the mount command? If /afs is mounted, does the > user have a token? How are you configuring the client's cell information= ; > are you using DNS or a local CellServDB? > > Mike -- > --=20 Vitor Lima Bacharelando em Ci=EAncia da Computa=E7=E3o Instituto de Matem=E1tica e Estat=EDstica (IME) da Universidade de S=E3o Paulo (USP) --0016e64c1af6b7acf404ab4379d7 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hello!

Sorry for the absence of information.

This OpenAFS ins= tallation is running in the server1 and I have a OpenAFS installation worki= ng well in the server0.
I=A0 thought the server1's OpenAFS was worki= ng, but when I stop the bosserver of the server0, that problem begins to oc= cur.

Oh, the OpenAFS client is running and I am using the CellServDB.





2011/8/24 Michael Meffie <mmeffie@sinenomin= e.net>
vitor lima wrote:
Hi!

I have a problem with my OpenAFS server:

When a user do the "log in", the home directory is not found and = the user logs using the root (the / ).

What can be this?
Somebody can help me?

Hello Vitor,

Is this a new OpenAFS installation? Has this worked before? =A0You are
going to need to provide some more information before we can give
you more guidance. =A0Firstly, is the OpenAFS client running? What
is the output of the mount command? =A0If /afs is mounted, does the
user have a token? =A0How are you configuring the client's cell informa= tion;
are you using DNS or a local CellServDB?

Mike --



--
Vitor Lima
Bacharela= ndo em Ci=EAncia da Computa=E7=E3o
Instituto de Matem=E1tica e Estat=EDs= tica (IME)
da Universidade de S=E3o Paulo (USP)
--0016e64c1af6b7acf404ab4379d7-- From vittorlima@gmail.com Wed Aug 24 18:33:01 2011 From: vittorlima@gmail.com (vitor lima) Date: Wed, 24 Aug 2011 14:33:01 -0300 Subject: [OpenAFS] Home not found In-Reply-To: References: <4E552104.5050003@sinenomine.net> Message-ID: --0016e6d6455938428f04ab43b4bb Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Theses are the messages that I receive when I try the log in: " pc03 login: vitor Password: Linux 2.6.33.4 Last Login: Wed Aug 24 14:27:27 2011 tty6 *No directory, logging in HOME=3D/* " 2011/8/24 vitor lima > Hello! > > Sorry for the absence of information. > > This OpenAFS installation is running in the server1 and I have a OpenAFS > installation working well in the server0. > I thought the server1's OpenAFS was working, but when I stop the bosserv= er > of the server0, that problem begins to occur. > > Oh, the OpenAFS client is running and I am using the CellServDB. > > > > > > > 2011/8/24 Michael Meffie > >> vitor lima wrote: >> >>> Hi! >>> >>> I have a problem with my OpenAFS server: >>> >>> When a user do the "log in", the home directory is not found and the us= er >>> logs using the root (the / ). >>> >>> What can be this? >>> Somebody can help me? >>> >> >> Hello Vitor, >> >> Is this a new OpenAFS installation? Has this worked before? You are >> going to need to provide some more information before we can give >> you more guidance. Firstly, is the OpenAFS client running? What >> is the output of the mount command? If /afs is mounted, does the >> user have a token? How are you configuring the client's cell informatio= n; >> are you using DNS or a local CellServDB? >> >> Mike -- >> > > > > -- > Vitor Lima > Bacharelando em Ci=EAncia da Computa=E7=E3o > Instituto de Matem=E1tica e Estat=EDstica (IME) > da Universidade de S=E3o Paulo (USP) > --=20 Vitor Lima Bacharelando em Ci=EAncia da Computa=E7=E3o Instituto de Matem=E1tica e Estat=EDstica (IME) da Universidade de S=E3o Paulo (USP) --0016e6d6455938428f04ab43b4bb Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Theses are the messages that I receive when I try the log in:

"=
pc03 login: vitor
Password:
Linux 2.6.33.4
Last Login: Wed=A0 = Aug 24=A0 14:27:27 2011 tty6
No directory, logging in HOME=3D/"



2011/8/24 vitor lima <vittorlima@gmail.com>=
Hello!

Sorry for the absence of information.

This OpenAFS ins= tallation is running in the server1 and I have a OpenAFS installation worki= ng well in the server0.
I=A0 thought the server1's OpenAFS was worki= ng, but when I stop the bosserver of the server0, that problem begins to oc= cur.

Oh, the OpenAFS client is running and I am using the CellServDB.





<= br>2011/8/24 Michael Meffie <mmeffie@sinenomine.net>
vitor lima w= rote:
Hi!

I have a problem with my OpenAFS server:

When a user do the "log in", the home directory is not found and = the user logs using the root (the / ).

What can be this?
Somebody can help me?

Hello Vitor,

Is this a new OpenAFS installation? Has this worked before? =A0You are
going to need to provide some more information before we can give
you more guidance. =A0Firstly, is the OpenAFS client running? What
is the output of the mount command? =A0If /afs is mounted, does the
user have a token? =A0How are you configuring the client's cell informa= tion;
are you using DNS or a local CellServDB?

Mike --



--
Vitor Lima
Bacharelando em Ci=EAncia da Computa= =E7=E3o
Instituto de Matem=E1tica e Estat=EDstica (IME)
da Universida= de de S=E3o Paulo (USP)



--
Vitor Lima<= br>Bacharelando em Ci=EAncia da Computa=E7=E3o
Instituto de Matem=E1tica= e Estat=EDstica (IME)
da Universidade de S=E3o Paulo (USP)
--0016e6d6455938428f04ab43b4bb-- From vittorlima@gmail.com Wed Aug 24 18:37:06 2011 From: vittorlima@gmail.com (vitor lima) Date: Wed, 24 Aug 2011 14:37:06 -0300 Subject: [OpenAFS] Home not found In-Reply-To: References: <4E552104.5050003@sinenomine.net> Message-ID: --0016e6d64559d520d504ab43c2eb Content-Type: text/plain; charset=ISO-8859-1 Sorry, It's just "/" instead "3D/" --0016e6d64559d520d504ab43c2eb Content-Type: text/html; charset=ISO-8859-1 Sorry, It's just "/" instead "3D/"

--0016e6d64559d520d504ab43c2eb-- From vittorlima@gmail.com Wed Aug 24 18:50:35 2011 From: vittorlima@gmail.com (vitor lima) Date: Wed, 24 Aug 2011 14:50:35 -0300 Subject: [OpenAFS] Home not found In-Reply-To: <20110824133628.12036o32xv1ug3a4@mail.msu.edu> References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> Message-ID: --0016364d1fc50c262b04ab43f318 Content-Type: text/plain; charset=ISO-8859-1 I am using OpenLDAP. All the lines of the exiting of the comand getent passwd are similar to userX:*:1012:110:User user:/afs/my.cell.name/usr/userX:/bin/bash So, I think that configuration is well. Do I get this token with the comand kinit or with the aklog ?? Thanks. --0016364d1fc50c262b04ab43f318 Content-Type: text/html; charset=ISO-8859-1 I am using OpenLDAP.
All the lines of the exiting of the comand getent passwd are similar to

userX:*:1012:110:User user:/afs/my.cell.name/usr/userX:/bin/bash

So, I think that configuration is well.


Do I get this token with the comand
kinit <user_name>
or with the
aklog
??


Thanks.
--0016364d1fc50c262b04ab43f318-- From adeason@sinenomine.net Wed Aug 24 20:10:04 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Wed, 24 Aug 2011 14:10:04 -0500 Subject: [OpenAFS] Re: Home not found References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> Message-ID: <20110824141004.6b0b31e2.adeason@sinenomine.net> On Wed, 24 Aug 2011 14:50:35 -0300 vitor lima wrote: > Do I get this token with the comand > kinit > or with the > aklog > ?? Use both. Run 'kinit ' and then run 'aklog'. -- Andrew Deason adeason@sinenomine.net From adeason@sinenomine.net Wed Aug 24 20:12:24 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Wed, 24 Aug 2011 14:12:24 -0500 Subject: [OpenAFS] Re: Home not found References: <4E552104.5050003@sinenomine.net> Message-ID: <20110824141224.a775349e.adeason@sinenomine.net> On Wed, 24 Aug 2011 14:16:40 -0300 vitor lima wrote: > This OpenAFS installation is running in the server1 and I have a > OpenAFS installation working well in the server0. I thought the > server1's OpenAFS was working, but when I stop the bosserver of the > server0, that problem begins to occur. What server processes are running on each of these servers? ('fileserver', 'vlserver', or 'ptserver', etc.) If you only have 1 or 2 database servers (the servers that run 'vlserver' and 'ptserver'), you will probably need server0 to be up for most things to function. -- Andrew Deason adeason@sinenomine.net From shadow@gmail.com Thu Aug 25 01:09:22 2011 From: shadow@gmail.com (Derrick Brashear) Date: Wed, 24 Aug 2011 20:09:22 -0400 Subject: [OpenAFS] Re: Home not found In-Reply-To: <20110824141224.a775349e.adeason@sinenomine.net> References: <4E552104.5050003@sinenomine.net> <20110824141224.a775349e.adeason@sinenomine.net> Message-ID: On Wed, Aug 24, 2011 at 3:12 PM, Andrew Deason wro= te: > On Wed, 24 Aug 2011 14:16:40 -0300 > vitor lima wrote: > >> This OpenAFS installation is running in the server1 and I have a >> OpenAFS installation working well in the server0. =A0I =A0thought the >> server1's OpenAFS was working, but when I stop the bosserver of the >> server0, that problem begins to occur. > > What server processes are running on each of these servers? > ('fileserver', 'vlserver', or 'ptserver', etc.) If you only have 1 or 2 > database servers (the servers that run 'vlserver' and 'ptserver'), you > will probably need server0 to be up for most things to function. notably, if server0 has the lowest ip and is the only or just one of 2 servers naed in CellServDB, you have to leave it up. --=20 Derrick From vittorlima@gmail.com Thu Aug 25 14:03:31 2011 From: vittorlima@gmail.com (vitor lima) Date: Thu, 25 Aug 2011 10:03:31 -0300 Subject: [OpenAFS] Home not found In-Reply-To: References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> Message-ID: --0016e6d6455942440604ab540e43 Content-Type: text/plain; charset=ISO-8859-1 Andrew Deason: "notably, if server0 has the lowest ip and is the only or just one of 2 servers naed in CellServDB, you have to leave it up." --- Sorry, but I did not understand. In the file CellServDB I have the lines in the next order: 192.168.1.103 #server0 192.168.1.104 #server1 And these are the only servers that I have. You want that I trade the two lines like that 192.168.1.104 #server1 192.168.1.103 #server0 ?? I made it in the client and the problem continued... --0016e6d6455942440604ab540e43 Content-Type: text/html; charset=ISO-8859-1 
Andrew Deason:
"notably, if server0 has the lowest ip and is the only or just one of 2
servers naed in CellServDB, you have to leave it up."

---
Sorry, but I did not understand.


In the file CellServDB I have the lines in the next order:

192.168.1.103    #server0
192.168.1.104    #server1

And these are the only servers that I have.
You want that I trade the two lines like that


192.168.1.104    #server1
192.168.1.103    #server0
??

I made it in the client and the problem continued...



--0016e6d6455942440604ab540e43-- From adeason@sinenomine.net Thu Aug 25 15:10:21 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Thu, 25 Aug 2011 09:10:21 -0500 Subject: [OpenAFS] Re: Home not found References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> Message-ID: <20110825091021.cb062951.adeason@sinenomine.net> On Thu, 25 Aug 2011 10:03:31 -0300 vitor lima wrote: > Sorry, but I did not understand. > > In the file CellServDB I have the lines in the next order: > > 192.168.1.103 #server0 > 192.168.1.104 #server1 > > And these are the only servers that I have. This is correct, but server0 must stay up for everything to work. You may still be able to access files without server0, but it may be slower and it is not guaranteed. Are both server0 and server1 running the 'vlserver' and 'ptserver' processes? -- Andrew Deason adeason@sinenomine.net From deengert@anl.gov Thu Aug 25 16:38:40 2011 From: deengert@anl.gov (Douglas E. Engert) Date: Thu, 25 Aug 2011 10:38:40 -0500 Subject: [OpenAFS] Home not found In-Reply-To: References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> Message-ID: <4E566C80.3040000@anl.gov> On 8/25/2011 8:03 AM, vitor lima wrote: > Andrew Deason: > > "notably, if server0 has the lowest ip and is the only or just one of 2 > > servers naed in CellServDB, you have to leave it up." > > --- > Sorry, but I did not understand. > > > In the file CellServDB I have the lines in the next order: > > 192.168.1.103 #server0 > 192.168.1.104 #server1 > > And these are the only servers that I have. > You want that I trade the two lines like that > > > 192.168.1.104 #server1 > 192.168.1.103 #server0 > ?? > > I made it in the client and the problem continued... The database servers vote to see if there is a quorum. In case of ties, the server with the lowest IP gets the tie breaker vote. So if only server1 is running, it does not know the state of server0, and has to assume it is running, and server0 represents the cell. If server1 goes down, server0 still represents the cell. So you would be best with 3 database servers. With 2 servers all you really get is a backup of the database. server0 must still be running for full operations. > > > > > > > > -- Douglas E. Engert Argonne National Laboratory 9700 South Cass Avenue Argonne, Illinois 60439 (630) 252-5444 From vittorlima@gmail.com Thu Aug 25 17:59:46 2011 From: vittorlima@gmail.com (vitor lima) Date: Thu, 25 Aug 2011 13:59:46 -0300 Subject: [OpenAFS] Re: Home not found In-Reply-To: References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> <20110825092228.7e623909.adeason@sinenomine.net> Message-ID: --0016e6d6455924838a04ab575b06 Content-Type: text/plain; charset=ISO-8859-1 Andrew: " This is correct, but server0 must stay up for everything to work. You may still be able to access files without server0, but it may be slower and it is not guaranteed. Are both server0 and server1 running the 'vlserver' and 'ptserver' processes? " The OpenAFS's process in the server0: bosserver; fileserver; volserver; ptserver; vlserver. The server1 has the same process running. So, I need to configure the server1 to work without the server0, because I will need to turn it off. --0016e6d6455924838a04ab575b06 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Andrew:
"
This is correct, but server0 must stay up for everythi= ng to work.
=A0You may still be able to access files without server0, bu= t it may be slower
and it is not guaranteed. Are both server0 and serve= r1 running
the 'vlserver' and 'ptserver' processes?
"

<= br>The OpenAFS's process in the server0:

bosserver;
fileserve= r;
volserver;
ptserver;
vlserver.

The server1 has the same = process running.

So, I need to configure the server1 to work without the server0, becaus= e I will need to turn it off.






--0016e6d6455924838a04ab575b06-- From adeason@sinenomine.net Thu Aug 25 18:14:52 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Thu, 25 Aug 2011 12:14:52 -0500 Subject: [OpenAFS] Re: Home not found References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> <20110825092228.7e623909.adeason@sinenomine.net> Message-ID: <20110825121452.28c0d5ca.adeason@sinenomine.net> On Thu, 25 Aug 2011 13:59:46 -0300 vitor lima wrote: > So, I need to configure the server1 to work without the server0, > because I will need to turn it off. If you need the cell to function if _any_ server is turned off, you need 3 servers. A 2-server setup cannot be fully redundant in the way that you want (if I understand you correctly). If server1 can stay on all the time, and you just need to turn off server0, then give server1 a lower IP address than server0, and that will "sort of" work as long as server1 stays on. But if you have 3 servers, that is better and is a more normal setup. -- Andrew Deason adeason@sinenomine.net From vittorlima@gmail.com Thu Aug 25 20:06:07 2011 From: vittorlima@gmail.com (vitor lima) Date: Thu, 25 Aug 2011 16:06:07 -0300 Subject: [OpenAFS] Re: Home not found In-Reply-To: References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> <20110825092228.7e623909.adeason@sinenomine.net> <20110825121813.2b83940c.adeason@sinenomine.net> Message-ID: --0016364d1fc508eba504ab591f2a Content-Type: text/plain; charset=ISO-8859-1 Andrew wrote: >If you need the cell to function if _any_ server is turned off, you need >3 servers. A 2-server setup cannot be fully redundant in the way that >you want (if I understand you correctly). > >If server1 can stay on all the time, and you just need to turn off >server0, then give server1 a lower IP address than server0, and that >will "sort of" work as long as server1 stays on. But if you have 3 >servers, that is better and is a more normal setup. I need the server1 working without the server0. I want to move all the volumes from server0 to server1 and then turn off the server0. So, after, I will restart the server0 and leave both turned on. --0016364d1fc508eba504ab591f2a Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable 
Andrew wrote:
>If you need the cell to function if _any_ server =
is turned off, you need
>3 servers. A 2-server setup cannot be fully =
redundant in the way that
>you want (if I understand you correctly).<=
br>
>
>If server1 can stay on all the time, and you just need to turn =
off
>server0, then give server1 a lower IP address than server0, and =
that
>will "sort of" work as long as server1 stays on. But =
if you have 3

>servers, that is better and is a more normal setup.


I need t=
he server1 working without the server0.
I want to move all the volumes f=
rom server0 to server1 and then turn off the server0.

So, after, I w=
ill restart the server0 and leave both turned on.


--0016364d1fc508eba504ab591f2a-- From shadow@gmail.com Thu Aug 25 21:09:39 2011 From: shadow@gmail.com (Derrick Brashear) Date: Thu, 25 Aug 2011 16:09:39 -0400 Subject: [OpenAFS] Re: Home not found In-Reply-To: References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> <20110825092228.7e623909.adeason@sinenomine.net> <20110825121813.2b83940c.adeason@sinenomine.net> Message-ID: On Thu, Aug 25, 2011 at 3:06 PM, vitor lima wrote: > Andrew wrote: >>If you need the cell to function if _any_ server is turned off, you need >>3 servers. A 2-server setup cannot be fully redundant in the way that >>you want (if I understand you correctly). > >> >>If server1 can stay on all the time, and you just need to turn off >>server0, then give server1 a lower IP address than server0, and that >>will "sort of" work as long as server1 stays on. But if you have 3 > >>servers, that is better and is a more normal setup. > > > I need the server1 working without the server0. > I want to move all the volumes from server0 to server1 and then turn off the > server0. > > So, after, I will restart the server0 and leave both turned on. you can do that, and then decommission server0, if you wish, by removing it from all the CellServDB files (and making sure to restart the servers on server1). but then server0 will be not a server at all, ever. basically, the issue is server0 if it has the lower (numeric) IP address cannot be a server sometimes. either you're using it, or you're not. -- Derrick From haba@kth.se Fri Aug 26 07:43:46 2011 From: haba@kth.se (Harald Barth) Date: Fri, 26 Aug 2011 08:43:46 +0200 (CEST) Subject: [OpenAFS] Re: Debugging a network performance problem that affects AFS In-Reply-To: <4D30A035.6070506@btv.ibm.com> References: <4D30535A.2030103@btv.ibm.com> <20110114093523.511c6fb1.adeason@sinenomine.net> <4D30A035.6070506@btv.ibm.com> Message-ID: <20110826.084346.679382604055780848.haba@habanero> > In this particular case, I am running xstat_cm_test on the client, > against that same client. So the kind of error you cite shouldn't > exist. I suspect it's a problem with the building of the package. I > need to reboot to Gentoo, and I'm pretty sure I've run xstat_cm_test > there, though with different options. Now I have stumbled over the same problem. On all machines I have tried (well, they are all 64 bit Linux) I get # xstat_cm_test localhost 2 -onceonly -debug RunTheTest: Allocating socket array for 1 Cache Manager(s) Allocating 1 long(s) for coll ID CollID at index 0 is 2 Starting up the xstat_cm service, debugging enabled, one-shot operation [xstat_cm_Init] Asking for 1 collection(s): 2 [xstat_cm_Init] Initializing Rx on port 0 [xstat_cm_Init] Rx initialized on port 0 [xstat_cm_Init] Probe LWP client security object created [xstat_cm_Init] Copying in the following socket info: [xstat_cm_Init] IP addr 0s, port 1231122960 [xstat_cm_Init] Host name for server index 0 is localhost [xstat_cm_Init] Connecting to srv idx 0, IP addr 127.0.0.1, port 7001, service 1 [xstat_cm_Init] New connection at 0x14e8ee0 [xstat_cm_Init] Creating the probe LWP [xstat_cm_Init] Probe LWP process structure located at 0x14fb670 [RunTheTest] Calling LWP_WaitProcess() on event 0x62f180 [xstat_cm_LWP] Waking up, getting data from 1 server(s) [xstat_cm_LWP] Getting collections from Cache Manager 'localhost' [xstat_cm_LWP] Connection OK, calling RXAFSCB_GetXStats [xstat_cm_LWP] Asking for data collection 2 xstat_cm_LWP: Calling RXAFSCB_GetXStats, conn=0x14e8ee0, clientVersionNumber=2, collectionNumber=2, srvVersionNumberP=0x6203cfac, timeP=0x6311a4, dataP=0x6311b8 xstat_cm_LWP: [bufflen=2048, buffer at 0x62f1a0] [xstat_cm_LWP] Calling handler routine. ----------------------------------------------------------- ** Data size mismatch in performance collection!** Expecting 1064, got 759 ** Version mismatch with Cache Manager [xstat_cm_LWP] Polling complete for probe round 1. [xstat_cm_LWP] Signalling main process at 0x62f180 [RunTheTest] Returned from LWP_WaitProcess() Yawn, main thread just woke up. Cleaning things out... # This is against localhost. So this is not version mismatch but bug. Any ideas? I have not started searching around like trying on other OSes yet. Harald. PS: Or is this still the same bug as 2006? http://www.mail-archive.com/openafs-devel@openafs.org/msg08382.html From mmeffie@sinenomine.net Fri Aug 26 13:02:57 2011 From: mmeffie@sinenomine.net (Michael Meffie) Date: Fri, 26 Aug 2011 08:02:57 -0400 Subject: [OpenAFS] Re: Debugging a network performance problem that affects AFS In-Reply-To: <20110826.084346.679382604055780848.haba@habanero> References: <4D30535A.2030103@btv.ibm.com> <20110114093523.511c6fb1.adeason@sinenomine.net> <4D30A035.6070506@btv.ibm.com> <20110826.084346.679382604055780848.haba@habanero> Message-ID: <4E578B71.4060105@sinenomine.net> Harald Barth wrote: > ----------------------------------------------------------- > ** Data size mismatch in performance collection!** Expecting 1064, got 759 > ** Version mismatch with Cache Manager > [xstat_cm_LWP] Polling complete for probe round 1. > [xstat_cm_LWP] Signalling main process at 0x62f180 > [RunTheTest] Returned from LWP_WaitProcess() > > Yawn, main thread just woke up. Cleaning things out... > # > > This is against localhost. So this is not version mismatch but bug. > > Any ideas? I have not started searching around like trying on other OSes yet. > > Harald. Hello Harald, There is a candidate fix for this in gerrit. Can you try it? No need to reinstall the cache manager, this just fixes xstat_cm_test. http://gerrit.openafs.org/#change,5237 Thank, Mike From vittorlima@gmail.com Sat Aug 27 02:14:42 2011 From: vittorlima@gmail.com (vitor lima) Date: Fri, 26 Aug 2011 22:14:42 -0300 Subject: [OpenAFS] Re: Home not found In-Reply-To: References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> <20110825092228.7e623909.adeason@sinenomine.net> <20110825121813.2b83940c.adeason@sinenomine.net> Message-ID: --0016e64c1af606b40804ab72630a Content-Type: text/plain; charset=ISO-8859-1 I removed the line 192.168.1.103 # server0 from CellServDB of the client and of the server1. So, I turned off the server0 and I restarted the client and the server1. The problem continued. The user had a ticket (I checked this with the comand tokens), but the volume was not mounted. I ran the comand root@server1:~# vos listvol -server server1 Total number of volumes on server server1 partition /vicepa: 4 root.afs 536870912 RW 4 K On-line root.cell 536870915 RW 5 K On-line user.vitor 536871177 RW 502356 K On-line user.vitor.backup 536871179 BK 497186 K On-line Total volumes onLine 4 ; Total volumes offLine 0 ; Total busy 0 Total number of volumes on server cow01 partition /vicepb: 0 Total volumes onLine 0 ; Total volumes offLine 0 ; Total busy 0 Total number of volumes on server cow01 partition /vicepc: 0 Total volumes onLine 0 ; Total volumes offLine 0 ; Total busy 0 Total number of volumes on server cow01 partition /vicepd: 0 Total volumes onLine 0 ; Total volumes offLine 0 ; Total busy 0 But when I tried to mount the volume user.vitor, I received a message talking "volume don't exist in this cell". Then, I ran root@server1:~# vos listvldb VLDB entries for all servers root.afs RWrite: 536870912 number of sites -> 1 server cow01.labmap.ime.usp.br partition /vicepa RW Site root.cell RWrite: 536870915 number of sites -> 1 server cow01.labmap.ime.usp.br partition /vicepa RW Site Total entries: 2 The user.vitor is not in the VLDB server O.o This volume was moved from server0 to server1 with the comand vos move -id user.vitor -fromserver server0 -frompartition /vicepb -toserver server1 -topartition /vicepa What can I do now? --0016e64c1af606b40804ab72630a Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: base64 SSByZW1vdmVkIHRoZSBsaW5lIDxicj4xOTIuMTY4LjEuMTAzIKCgICMgc2VydmVyMDxicj5mcm9t IENlbGxTZXJ2REIgb2YgdGhlIGNsaWVudCBhbmQgb2YgdGhlIHNlcnZlcjEuPGJyPjxicj5Tbywg SSB0dXJuZWQgb2ZmIHRoZSBzZXJ2ZXIwIGFuZCBJIHJlc3RhcnRlZCB0aGUgY2xpZW50IGFuZCB0 aGUgc2VydmVyMS48YnI+PGJyPlRoZSBwcm9ibGVtIGNvbnRpbnVlZC48YnI+PGJyPlRoZSB1c2Vy IGhhZCBhIHRpY2tldCAoSSBjaGVja2VkIHRoaXMgd2l0aCB0aGUgY29tYW5kIHRva2VucyksIGJ1 dCB0aGUgdm9sdW1lIHdhcyBub3QgbW91bnRlZC48YnI+CkkgcmFuIHRoZSBjb21hbmQ8YnI+cm9v dEBzZXJ2ZXIxOn4jIHZvcyBsaXN0dm9sIC1zZXJ2ZXIgc2VydmVyMaAgPGJyPlRvdGFsIG51bWJl ciBvZiB2b2x1bWVzIG9uIHNlcnZlciBzZXJ2ZXIxIHBhcnRpdGlvbiAvdmljZXBhOiA0PGJyPnJv b3QuYWZzoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoCA1MzY4NzA5MTIgUlegoKCgoKCgoKAgNCBL IE9uLWxpbmWgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgIDxicj4Kcm9vdC5jZWxsoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgIDUzNjg3MDkxNSBS V6CgoKCgoKCgoCA1IEsgT24tbGluZaCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKAgPGJyPgp1c2VyLnZpdG9yoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKAgNTM2ODcxMTc3IFJXoKCgoCA1MDIzNTYgSyBPbi1saW5loKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoCA8YnI+CnVzZXIudml0b3IuYmFja3Vw oKCgoKCgoKCgoKCgoKCgoCA1MzY4NzExNzkgQkugoKCgIDQ5NzE4NiBLIE9uLWxpbmWgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgIDxicj4KoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKAgPGJyPgpUb3RhbCB2b2x1bWVzIG9uTGluZSA0IDsgVG90YWwgdm9sdW1lcyBvZmZMaW5l IDAgOyBUb3RhbCBidXN5IDCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgIDxicj4KoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKAgPGJyPgpUb3RhbCBudW1iZXIgb2Ygdm9s dW1lcyBvbiBzZXJ2ZXIgY293MDEgcGFydGl0aW9uIC92aWNlcGI6IDCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoCA8YnI+CqCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgIDxicj4KVG90YWwgdm9sdW1lcyBvbkxpbmUgMCA7IFRvdGFsIHZvbHVtZXMgb2ZmTGluZSAw IDsgVG90YWwgYnVzeSAwoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKAgPGJyPgqgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoCA8YnI+ClRvdGFsIG51bWJlciBvZiB2b2x1 bWVzIG9uIHNlcnZlciBjb3cwMSBwYXJ0aXRpb24gL3ZpY2VwYzogMKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgIDxicj4KoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKAgPGJyPgpUb3RhbCB2b2x1bWVzIG9uTGluZSAwIDsgVG90YWwgdm9sdW1lcyBvZmZMaW5lIDAg OyBUb3RhbCBidXN5IDCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCg oKCgoKCgoKCgoKCgoKCgoCA8YnI+Cjxicj5Ub3RhbCBudW1iZXIgb2Ygdm9sdW1lcyBvbiBzZXJ2 ZXIgY293MDEgcGFydGl0aW9uIC92aWNlcGQ6IDAgPGJyPjxicj5Ub3RhbCB2b2x1bWVzIG9uTGlu ZSAwIDsgVG90YWwgdm9sdW1lcyBvZmZMaW5lIDAgOyBUb3RhbCBidXN5IDA8YnI+PGJyPjxicj5C dXQgd2hlbiBJIHRyaWVkIHRvIG1vdW50IHRoZSB2b2x1bWUgdXNlci52aXRvciwgSSByZWNlaXZl ZCBhIG1lc3NhZ2UgdGFsa2luZyAmcXVvdDt2b2x1bWUgZG9uJiMzOTt0IGV4aXN0IGluIHRoaXMg Y2VsbCZxdW90Oy48YnI+ClRoZW4sIEkgcmFuPGJyPjxicj5yb290QHNlcnZlcjE6fiMgdm9zIGxp c3R2bGRiPGJyPlZMREIgZW50cmllcyBmb3IgYWxsIHNlcnZlcnMgPGJyPjxicj5yb290LmFmcyA8 YnI+oKCgIFJXcml0ZTogNTM2ODcwOTEyIDxicj6goKAgbnVtYmVyIG9mIHNpdGVzIC0mZ3Q7IDE8 YnI+oKCgoKCgIHNlcnZlciA8YSBocmVmPSJodHRwOi8vY293MDEubGFibWFwLmltZS51c3AuYnIi PmNvdzAxLmxhYm1hcC5pbWUudXNwLmJyPC9hPiBwYXJ0aXRpb24gL3ZpY2VwYSBSVyBTaXRlIDxi cj4KPGJyPnJvb3QuY2VsbCA8YnI+oKCgIFJXcml0ZTogNTM2ODcwOTE1IDxicj6goKAgbnVtYmVy IG9mIHNpdGVzIC0mZ3Q7IDE8YnI+oKCgoKCgIHNlcnZlciA8YSBocmVmPSJodHRwOi8vY293MDEu bGFibWFwLmltZS51c3AuYnIiPmNvdzAxLmxhYm1hcC5pbWUudXNwLmJyPC9hPiBwYXJ0aXRpb24g L3ZpY2VwYSBSVyBTaXRlIDxicj48YnI+VG90YWwgZW50cmllczogMjxicj48YnI+PGJyPlRoZSB1 c2VyLnZpdG9yIGlzIG5vdCBpbiB0aGUgVkxEQiBzZXJ2ZXKgoKCgoKAgTy5vPGJyPgpUaGlzIHZv bHVtZSB3YXMgbW92ZWQgZnJvbSBzZXJ2ZXIwIHRvIHNlcnZlcjEgd2l0aCB0aGUgY29tYW5kPGJy PnZvcyBtb3ZlIC1pZCB1c2VyLnZpdG9yIC1mcm9tc2VydmVyIHNlcnZlcjAgLWZyb21wYXJ0aXRp b24gL3ZpY2VwYiAtdG9zZXJ2ZXIgc2VydmVyMSAtdG9wYXJ0aXRpb24gL3ZpY2VwYTxicj48YnI+ V2hhdCBjYW4gSSBkbyBub3c/PGJyPgo= --0016e64c1af606b40804ab72630a-- From geza@kzsdabas.hu Sat Aug 27 08:31:39 2011 From: geza@kzsdabas.hu (=?ISO-8859-1?Q?G=E9mes_G=E9za?=) Date: Sat, 27 Aug 2011 09:31:39 +0200 Subject: [OpenAFS] Re: Home not found In-Reply-To: References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> <20110825092228.7e623909.adeason@sinenomine.net> <20110825121813.2b83940c.adeason@sinenomine.net> Message-ID: <4E589D5B.601@kzsdabas.hu> This is a multi-part message in MIME format. --------------060200000909040701080303 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit 2011-08-27 03:14 keltezéssel, vitor lima írta: > I removed the line > 192.168.1.103 # server0 > from CellServDB of the client and of the server1. > > So, I turned off the server0 and I restarted the client and the server1. > > The problem continued. > > The user had a ticket (I checked this with the comand tokens), but the > volume was not mounted. > I ran the comand > root@server1:~# vos listvol -server server1 > Total number of volumes on server server1 partition /vicepa: 4 > root.afs 536870912 RW 4 K > On-line > > root.cell 536870915 RW 5 K > On-line > > user.vitor 536871177 RW 502356 K > On-line > > user.vitor.backup 536871179 BK 497186 K > On-line > > > > Total volumes onLine 4 ; Total volumes offLine 0 ; Total busy > 0 > > > > Total number of volumes on server cow01 partition /vicepb: > 0 > > > > Total volumes onLine 0 ; Total volumes offLine 0 ; Total busy > 0 > > > > Total number of volumes on server cow01 partition /vicepc: > 0 > > > > Total volumes onLine 0 ; Total volumes offLine 0 ; Total busy > 0 > > > Total number of volumes on server cow01 partition /vicepd: 0 > > Total volumes onLine 0 ; Total volumes offLine 0 ; Total busy 0 > > > But when I tried to mount the volume user.vitor, I received a message > talking "volume don't exist in this cell". > Then, I ran > > root@server1:~# vos listvldb > VLDB entries for all servers > > root.afs > RWrite: 536870912 > number of sites -> 1 > server cow01.labmap.ime.usp.br > partition /vicepa RW Site > > root.cell > RWrite: 536870915 > number of sites -> 1 > server cow01.labmap.ime.usp.br > partition /vicepa RW Site > > Total entries: 2 > > > The user.vitor is not in the VLDB server O.o > This volume was moved from server0 to server1 with the comand > vos move -id user.vitor -fromserver server0 -frompartition /vicepb > -toserver server1 -topartition /vicepa > > What can I do now? My suggestion is to turn on server0, and start to investigate: Does vos listvldb shows user.vitor now? Does the server (not the same as the client CellservDB which contain info about each known cell) CellservDB (/etc/openafs/server/CellServDB in my debian install) contain both server0 and server1 on server0 and server1 and all other volume and file server machines in your cell? Good Luck! Geza --------------060200000909040701080303 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit 2011-08-27 03:14 keltezéssel, vitor lima írta:
I removed the line
192.168.1.103    # server0
from CellServDB of the client and of the server1.

So, I turned off the server0 and I restarted the client and the server1.

The problem continued.

The user had a ticket (I checked this with the comand tokens), but the volume was not mounted.
I ran the comand
root@server1:~# vos listvol -server server1 
Total number of volumes on server server1 partition /vicepa: 4
root.afs                          536870912 RW          4 K On-line                                                                                                                                                                    ;      
root.cell                         536870915 RW          5 K On-line                                                                                                                                                                    ;      
user.vitor                        536871177 RW     502356 K On-line                                                                                                                                                                    ;      
user.vitor.backup                 536871179 BK     497186 K On-line                                                                                                                                                                    ;      
                                                                                                                                                                                                                                             
Total volumes onLine 4 ; Total volumes offLine 0 ; Total busy 0                                                                                                                                                                     ;        
                                                                                                                                                                                                                                             
Total number of volumes on server cow01 partition /vicepb: 0                                                                                                                                                                     ;            
                                                                                                                                                                                                                                             
Total volumes onLine 0 ; Total volumes offLine 0 ; Total busy 0                                                                                                                                                                     ;         
                                                                                                                                                                                                                                             
Total number of volumes on server cow01 partition /vicepc: 0                                                                                                                                                                     ;            
                                                                                                                                                                                                                                             
Total volumes onLine 0 ; Total volumes offLine 0 ; Total busy 0                                                                                                                                                                     ;         

Total number of volumes on server cow01 partition /vicepd: 0

Total volumes onLine 0 ; Total volumes offLine 0 ; Total busy 0


But when I tried to mount the volume user.vitor, I received a message talking "volume don't exist in this cell".
Then, I ran

root@server1:~# vos listvldb
VLDB entries for all servers

root.afs
    RWrite: 536870912
    number of sites -> 1
       server cow01.labmap.ime.usp.br partition /vicepa RW Site

root.cell
    RWrite: 536870915
    number of sites -> 1
       server cow01.labmap.ime.usp.br partition /vicepa RW Site

Total entries: 2


The user.vitor is not in the VLDB server       O.o
This volume was moved from server0 to server1 with the comand
vos move -id user.vitor -fromserver server0 -frompartition /vicepb -toserver server1 -topartition /vicepa

What can I do now?
My suggestion is to turn on server0, and start to investigate:
Does vos listvldb shows user.vitor now?
Does the server (not the same as the client CellservDB which contain info about each known cell) CellservDB (/etc/openafs/server/CellServDB in my debian install) contain both server0 and server1 on server0 and server1 and all other volume and file server machines in your cell?

Good Luck!

Geza
--------------060200000909040701080303-- From jm130794@gmail.com Sun Aug 28 15:30:00 2011 From: jm130794@gmail.com (jm130794) Date: Sun, 28 Aug 2011 16:30:00 +0200 Subject: [OpenAFS] Remove a mount point Message-ID: --20cf300fb01b0b47ae04ab919df2 Content-Type: text/plain; charset=ISO-8859-1 Hello, I made a error when I created a mount point :( # vos create afs1 a cell.dir1 # fs mkmount dir1 cell.dir1 # vos remove -id cell.dir1 On afs1 fileserver, I don't see dir1 but, on unix and windows client, it still appears. What should I do ? Thanks, JM --20cf300fb01b0b47ae04ab919df2 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hello,

I made a error when I created a mount point :(

# vos c= reate afs1 a cell.dir1
# fs mkmount dir1 cell.dir1
# vos remove -id c= ell.dir1

On afs1 fileserver, I don't see dir1 but, on unix and w= indows client, it still appears.

What should I do ?

Thanks,

JM


--20cf300fb01b0b47ae04ab919df2-- From aaronk@umbc.edu Sun Aug 28 15:36:47 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Sun, 28 Aug 2011 10:36:47 -0400 Subject: [OpenAFS] Remove a mount point In-Reply-To: References: Message-ID: --485b395db1967cc93c04ab91b6a9 Content-Type: text/plain; charset=ISO-8859-1 That in theory is an easy fix! fs rmmount dir1 On Sun, Aug 28, 2011 at 10:30 AM, jm130794 wrote: > Hello, > > I made a error when I created a mount point :( > > # vos create afs1 a cell.dir1 > # fs mkmount dir1 cell.dir1 > # vos remove -id cell.dir1 > > On afs1 fileserver, I don't see dir1 but, on unix and windows client, it > still appears. > > What should I do ? > > Thanks, > > JM > > > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --485b395db1967cc93c04ab91b6a9 Content-Type: text/html; charset=ISO-8859-1 That in theory is an easy fix!

fs rmmount dir1

On Sun, Aug 28, 2011 at 10:30 AM, jm130794 <jm130794@gmail.com> wrote:
Hello,

I made a error when I created a mount point :(

# vos create afs1 a cell.dir1
# fs mkmount dir1 cell.dir1
# vos remove -id cell.dir1

On afs1 fileserver, I don't see dir1 but, on unix and windows client, it still appears.

What should I do ?

Thanks,

JM





--
Aaron Knister
Systems Administrator
Division of Information Technology
University of Maryland, Baltimore County
aaronk@umbc.edu
--485b395db1967cc93c04ab91b6a9-- From jm130794@gmail.com Sun Aug 28 16:41:58 2011 From: jm130794@gmail.com (jm130794) Date: Sun, 28 Aug 2011 17:41:58 +0200 Subject: [OpenAFS] Remove a mount point In-Reply-To: References: Message-ID: --20cf303b411d68359a04ab929e68 Content-Type: text/plain; charset=ISO-8859-1 Hello, I did that on the file server ... not from a client. I did that from a Linux client and it worked! What do you think about it ? JM 2011/8/28 Aaron Knister > That in theory is an easy fix! > > fs rmmount dir1 > > > On Sun, Aug 28, 2011 at 10:30 AM, jm130794 wrote: > >> Hello, >> >> I made a error when I created a mount point :( >> >> # vos create afs1 a cell.dir1 >> # fs mkmount dir1 cell.dir1 >> # vos remove -id cell.dir1 >> >> On afs1 fileserver, I don't see dir1 but, on unix and windows client, it >> still appears. >> >> What should I do ? >> >> Thanks, >> >> JM >> >> >> > > > -- > Aaron Knister > Systems Administrator > Division of Information Technology > University of Maryland, Baltimore County > aaronk@umbc.edu > --20cf303b411d68359a04ab929e68 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hello,

I did that on the file server ... not from a client.

I= did that from a Linux client and it worked!

What do you think about= it ?

JM

2011/8/28 Aaron Knister <= span dir=3D"ltr"><aaronk@umbc.edu= >
That in theory is= an easy fix!

fs rmmount dir1

<= br>
On Sun, Aug 28, 2011 at 10:30 AM, jm130794 <jm= 130794@gmail.com> wrote:
Hello,

I m= ade a error when I created a mount point :(

# vos create afs1 a cell= .dir1
# fs mkmount dir1 cell.dir1
# vos remove -id cell.dir1

On afs1 fileserver, I don't see dir1 = but, on unix and windows client, it still appears.

What should I do ?

Thanks,

JM





--
Aaron Knister
Systems Administrator
Division of Informat= ion Technology
University of Maryland, Baltimore County
aaronk@umbc.edu

--20cf303b411d68359a04ab929e68-- From jason@rampaginggeek.com Sun Aug 28 22:00:23 2011 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Sun, 28 Aug 2011 17:00:23 -0400 Subject: [OpenAFS] improving cache partition performance Message-ID: <4E5AAC67.2000308@rampaginggeek.com> hi everyone, I'm looking at some performance problems on our web servers, and I think that hard drive performance is a bottleneck. All filesystems are on a RAID0 disk array with two 7200RPM disks (no battery-backed cache). The cache is on a dedicated ext3 partition. At some points, my await times on the disk are up to 1000ms. How can I improve this? Would setting commit time larger than 5 seconds be OK? Would setting the ext3 journal type to "data=writeback" be OK? platform: RHEL5.5 x86_64 Thanks, Jason From pontius@btv.ibm.com Mon Aug 29 12:54:40 2011 From: pontius@btv.ibm.com (Dale Pontius) Date: Mon, 29 Aug 2011 07:54:40 -0400 Subject: [OpenAFS] improving cache partition performance In-Reply-To: <4E5AAC67.2000308@rampaginggeek.com> References: <4E5AAC67.2000308@rampaginggeek.com> Message-ID: <4E5B7E00.1000400@btv.ibm.com> On 08/28/2011 05:00 PM, Jason Edgecombe wrote: > hi everyone, > > I'm looking at some performance problems on our web servers, and I > think that hard drive performance is a bottleneck. All filesystems are > on a RAID0 disk array with two 7200RPM disks (no battery-backed > cache). The cache is on a dedicated ext3 partition. At some points, my > await times on the disk are up to 1000ms. How can I improve this? > Would setting commit time larger than 5 seconds be OK? Would setting > the ext3 journal type to "data=writeback" be OK? > > platform: RHEL5.5 x86_64 > > For starters, I'd use ext2 instead of ext3, or perhaps use ext4 with journaling turned off. I seem to remember once hearing that having journaling on for the afs cache partition was even bad, as opposed to merely slightly slower. It's just a cache, it's never all that big, and if fsck were to have problems, it's easier to just plain wipe and start over. Dale Pontius -- Dale Pontius Senior Engineer IBM Corporation Phone: (802) 769-6850 Tie-Line: 446-6850 email: pontius@us.ibm.com This e-mail and its attachments, if any, may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply e-mail and delete all copies of this message from your system without copying it and notify sender of the misdirection by reply e-mail. From haba@kth.se Mon Aug 29 13:13:07 2011 From: haba@kth.se (Harald Barth) Date: Mon, 29 Aug 2011 14:13:07 +0200 (CEST) Subject: [OpenAFS] improving cache partition performance In-Reply-To: <4E5B7E00.1000400@btv.ibm.com> References: <4E5AAC67.2000308@rampaginggeek.com> <4E5B7E00.1000400@btv.ibm.com> Message-ID: <20110829.141307.769867324886520758.haba@habanero> > For starters, I'd use ext2 instead of ext3, or perhaps use ext4 with > journaling turned off. I'd second the use of ext2 for AFS cache because you can allways wipe and start over with an empty cache partition. During a crash, the transactions that were not written to the server yet are lost anyway. How big a cache do you need for your web server? Maybe stripe/mirror would do much better than RAID5. Harald. From haba@kth.se Mon Aug 29 15:16:19 2011 From: haba@kth.se (Harald Barth) Date: Mon, 29 Aug 2011 16:16:19 +0200 (CEST) Subject: [OpenAFS] Re: Debugging a network performance problem that affects AFS In-Reply-To: <4E578B71.4060105@sinenomine.net> References: <4D30A035.6070506@btv.ibm.com> <20110826.084346.679382604055780848.haba@habanero> <4E578B71.4060105@sinenomine.net> Message-ID: <20110829.161619.273237189756731587.haba@habanero> > There is a candidate fix for this in gerrit. Can you try it? No need > to reinstall the cache manager, this just fixes xstat_cm_test. > > http://gerrit.openafs.org/#change,5237 Adapted patch for 1.6.0 (the #elif defined(AFS_SUN57_ENV) in osi.h does prevent a clean apply) and tested it. I get a lot of numbers which to my untrained eye look reasonable. Now I "only" need to learn what to look for and then I will swiftly tune my cache sizes (ok, one can dream ;-) Harald. From jwedgeco@uncc.edu Mon Aug 29 15:22:12 2011 From: jwedgeco@uncc.edu (Edgecombe, Jason) Date: Mon, 29 Aug 2011 14:22:12 +0000 Subject: [OpenAFS] improving cache partition performance In-Reply-To: <20110829.141307.769867324886520758.haba@habanero> References: <4E5AAC67.2000308@rampaginggeek.com> <4E5B7E00.1000400@btv.ibm.com> <20110829.141307.769867324886520758.haba@habanero> Message-ID: <4DE25DED585CD840A23808AFDE3DA2060701D4@RPITSEXMS1.its.uncc.edu> % fs getcacheparms AFS using 145647975 of the cache's available 212484585 1K byte blocks. I'm already using RAID 0, so I'm already striping. I'll have to do some ben= chmarking with ext2. Our web servers handle PHP web apps and large movie files. Jason --------------------------------------------------------------------------- Jason Edgecombe | Linux and Solaris Administrator UNC Charlotte | The William States Lee College of Engineering 9201 University City Blvd. | Charlotte, NC 28223-0001 Phone: 704-687-3514 jwedgeco@uncc.edu | http://coe.uncc.edu | =A0Facebook --------------------------------------------------------------------------- If you are not the intended recipient of this transmission or a person resp= onsible for delivering it to the intended recipient, any disclosure, copyin= g, distribution, or other use of any of the information in this transmissio= n is strictly prohibited. If you have received this transmission in error, = please notify me immediately by reply e-mail or by telephone at 704-687-351= 4.=A0 Thank you. -----Original Message----- From: openafs-info-admin@openafs.org [mailto:openafs-info-admin@openafs.org= ] On Behalf Of Harald Barth Sent: Monday, August 29, 2011 8:13 AM To: openafs-info@openafs.org Subject: Re: [OpenAFS] improving cache partition performance > For starters, I'd use ext2 instead of ext3, or perhaps use ext4 with > journaling turned off. I'd second the use of ext2 for AFS cache because you can allways wipe and start over with an empty cache partition. During a crash, the transactions that were not written to the server yet are lost anyway. How big a cache do you need for your web server? Maybe stripe/mirror would do much better than RAID5. Harald. _______________________________________________ OpenAFS-info mailing list OpenAFS-info@openafs.org https://lists.openafs.org/mailman/listinfo/openafs-info From sxw@inf.ed.ac.uk Mon Aug 29 15:47:32 2011 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Mon, 29 Aug 2011 15:47:32 +0100 Subject: [OpenAFS] improving cache partition performance In-Reply-To: <4DE25DED585CD840A23808AFDE3DA2060701D4@RPITSEXMS1.its.uncc.edu> References: <4E5AAC67.2000308@rampaginggeek.com> <4E5B7E00.1000400@btv.ibm.com> <20110829.141307.769867324886520758.haba@habanero> <4DE25DED585CD840A23808AFDE3DA2060701D4@RPITSEXMS1.its.uncc.edu> Message-ID: <6CA3D994-3E10-4421-9921-AC26F69C3347@inf.ed.ac.uk> To paraphrase a famous AFS saying: The fastest disk access is the one = you don't do. If your disks are having performance issues (I'd also be interested in = seeing utilisation figures, and the difference between the await and = svctm times), then the easiest way to fix that is to have less requests = hit the disks. There are a few ways you can do this a) Turn off journalling - you don't need a journalled file system for an = AFS cache, and ext3 is known to be slower than ext2 b) Use a 1.6.0 AFS client. The Linux VM layer in the 1.6.0 codebase has = been rewritten, so that we make much better use of the page cache, and = satisfy more results from memory rather than going to the underlying = disk c) Add more memory to your servers. Linux will use any free memory to = cache disk data, so this can significantly reduce your IO overhead. S. From adeason@sinenomine.net Mon Aug 29 15:58:19 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 29 Aug 2011 09:58:19 -0500 Subject: [OpenAFS] Re: Home not found References: <4E552104.5050003@sinenomine.net> <20110824133628.12036o32xv1ug3a4@mail.msu.edu> <20110825092228.7e623909.adeason@sinenomine.net> <20110825121813.2b83940c.adeason@sinenomine.net> Message-ID: <20110829095819.6ecf2626.adeason@sinenomine.net> On Fri, 26 Aug 2011 22:14:42 -0300 vitor lima wrote: > The user.vitor is not in the VLDB server O.o > This volume was moved from server0 to server1 with the comand > vos move -id user.vitor -fromserver server0 -frompartition /vicepb -toserver > server1 -topartition /vicepa > > What can I do now? If you just want to get things working, you can run: vos syncvldb server1 which should create the vldb entry for user.vitor. However, you _really_ should not need to do that. If you want to look into why that happened, you should run: udebug server0 7003 -long udebug server1 7003 -long when both server0 and server1 are up. -- Andrew Deason adeason@sinenomine.net From pontius@btv.ibm.com Mon Aug 29 17:08:48 2011 From: pontius@btv.ibm.com (Dale Pontius) Date: Mon, 29 Aug 2011 12:08:48 -0400 Subject: [OpenAFS] improving cache partition performance In-Reply-To: <6CA3D994-3E10-4421-9921-AC26F69C3347@inf.ed.ac.uk> References: <4E5AAC67.2000308@rampaginggeek.com> <4E5B7E00.1000400@btv.ibm.com> <20110829.141307.769867324886520758.haba@habanero> <4DE25DED585CD840A23808AFDE3DA2060701D4@RPITSEXMS1.its.uncc.edu> <6CA3D994-3E10-4421-9921-AC26F69C3347@inf.ed.ac.uk> Message-ID: <4E5BB990.3090906@btv.ibm.com> On 08/29/2011 10:47 AM, Simon Wilkinson wrote: > To paraphrase a famous AFS saying: The fastest disk access is the one you don't do. > > If your disks are having performance issues (I'd also be interested in seeing utilisation figures, and the difference between the await and svctm times), then the easiest way to fix that is to have less requests hit the disks. > > There are a few ways you can do this > a) Turn off journalling - you don't need a journalled file system for an AFS cache, and ext3 is known to be slower than ext2 > b) Use a 1.6.0 AFS client. The Linux VM layer in the 1.6.0 codebase has been rewritten, so that we make much better use of the page cache, and satisfy more results from memory rather than going to the underlying disk > c) Add more memory to your servers. Linux will use any free memory to cache disk data, so this can significantly reduce your IO overhead. > It's possible to turn journaling off in ext4. Ext4 has much more fundamental improvements over ext3, where ext3 basically added journaling to ext2. Does anyone have a feel for running the afs cache this way? This commit seems to indicate that it's a good idea: https://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=0390131ba84fd3f726f9e24fc4553828125700bb Dale Pontius -- Dale Pontius Senior Engineer IBM Corporation Phone: (802) 769-6850 Tie-Line: 446-6850 email: pontius@us.ibm.com This e-mail and its attachments, if any, may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply e-mail and delete all copies of this message from your system without copying it and notify sender of the misdirection by reply e-mail. From haba@kth.se Mon Aug 29 17:33:44 2011 From: haba@kth.se (Harald Barth) Date: Mon, 29 Aug 2011 18:33:44 +0200 (CEST) Subject: [OpenAFS] improving cache partition performance In-Reply-To: <4DE25DED585CD840A23808AFDE3DA2060701D4@RPITSEXMS1.its.uncc.edu> References: <4E5B7E00.1000400@btv.ibm.com> <20110829.141307.769867324886520758.haba@habanero> <4DE25DED585CD840A23808AFDE3DA2060701D4@RPITSEXMS1.its.uncc.edu> Message-ID: <20110829.183344.152792587061568918.haba@habanero> > % fs getcacheparms > AFS using 145647975 of the cache's available 212484585 1K byte blocks. So why are you "only" at 145 of 212 GB? Either your cache really is bigger than your working set since boot or you have hit the number of cache entries before the cache size is exhausted. > I'm already using RAID 0, so I'm already striping. Ah, sorry, when I read RAID, I thought RAID5. > I'll have to do some benchmarking with ext2. I will be faster that ext3. The question is how much, which probably depends on your specific load. I have not made any recent experiments with other file systems (for example ext4, xfs) as cache yet, I just know that reiserfs killed my files back in the days... I use xfs as server (/vicep*/) file system. Harald. PS: Some years ago I did some benchmarking of AFS cache performance. I was astonished how much slower ext2 on ramdisk was compared to memcache. So the question is, how fast could a AFS disk cache be if it did not needed to bother with a file system and could use swap? From sxw@inf.ed.ac.uk Mon Aug 29 19:27:05 2011 From: sxw@inf.ed.ac.uk (Simon Wilkinson) Date: Mon, 29 Aug 2011 19:27:05 +0100 Subject: [OpenAFS] improving cache partition performance In-Reply-To: <20110829.183344.152792587061568918.haba@habanero> References: <4E5B7E00.1000400@btv.ibm.com> <20110829.141307.769867324886520758.haba@habanero> <4DE25DED585CD840A23808AFDE3DA2060701D4@RPITSEXMS1.its.uncc.edu> <20110829.183344.152792587061568918.haba@habanero> Message-ID: <5C323B12-D5F0-4CA3-BA0E-B9005C5D6124@inf.ed.ac.uk> On 29 Aug 2011, at 17:33, Harald Barth wrote: >=20 > PS: Some years ago I did some benchmarking of AFS cache performance. I > was astonished how much slower ext2 on ramdisk was compared to > memcache. So the question is, how fast could a AFS disk cache be if it > did not needed to bother with a file system and could use swap? We've made significant improvements here with 1.6.0 - the performance of = the AFS cache is now similar to the performance of the file system which = underlies the cache. It's still slower than memcache (which has = significantly less overhead), but does have the advantage that you don't = have to permanently devote a chunk of your operating system's memory to = the disk cache. Creating a block-device based cache is often discussed. The idea here is = that you interface with the underlying storage at a block-based, rather = than file-based level. You'd split your device into chunk sized records, = and simply ship (portions of) those records to and from disk as = required. Memory caching would be performed by the operating system = caching AFS pages, so this would also avoid the double caching problem. = It's a very interesting research project for anyone that has the time to = actually implement it.=20 Cheers, Simon. From boyland@cs.uwm.edu Mon Aug 29 20:18:14 2011 From: boyland@cs.uwm.edu (John Tang Boyland) Date: Mon, 29 Aug 2011 14:18:14 -0500 Subject: [OpenAFS] can neither delete nor use backup volume (DA server 1.6.0pre2) Message-ID: <488.1314645494@pabst.cs.uwm.edu> I accidentally moved a backed-up volume from partition a to b. I moved it back, but now the backup volume is misbehaving. (The process is REPEATABLE!) I can't seem to remove all traces of it, but neither can I use it. I can get root access on the (Linux) file server and start hacking files directly, but perhaps someone on this list can help me avoid making a terrible blunder. Here's a partial history. Notice that after the "zap", vos backup gives a different error once, but then it's back to the same set of errors again. I've been able to create backups of other volumes on the same partition today. So it seems that vos zap doesn't actually zap everything. Something's hanging around which is preventing vos backup from working. (And no, salvaging doesn't help either.) Any advice? John P.S. If you cc me in a response, I will get it quicker. ------------------------------------------------------------------ admin 79 % rxdebug localhost -port 7001 -version Trying 127.0.0.1 (port 7001): AFS version: OpenAFS 1.4.11 built 2009-07-13 admin 80 % rxdebug jeremiah -port 7000 -version Trying 129.89.143.70 (port 7000): AFS version: OpenAFS 1.6.0pre2 built 2011-02-17 (91.pre2.sl6@fnal.gov) admin 82 % vos examine fa11.cs351.backup **** Could not attach volume 536875958 **** RWrite: 536875956 Backup: 536875958 number of sites -> 1 server jeremiah.cs.uwm.edu partition /vicepa RW Site admin 83 % vos examine fa11.cs351 fa11.cs351 536875956 RW 1679 K On-line jeremiah.cs.uwm.edu /vicepa RWrite 536875956 ROnly 0 Backup 536875958 MaxQuota 100000 K Creation Thu Jun 23 20:15:28 2011 Copy Mon Aug 29 11:13:07 2011 Backup Mon Aug 29 13:43:23 2011 Last Access Mon Aug 29 13:33:06 2011 Last Update Mon Aug 29 13:28:10 2011 294 accesses in the past day (i.e., vnode references) RWrite: 536875956 Backup: 536875958 number of sites -> 1 server jeremiah.cs.uwm.edu partition /vicepa RW Site admin 84 % vos remove fa11.cs351.backup vos: Missing required parameter '-id' admin 85 % vos remove jeremiah a fa11.cs351.backup WARNING: Volume 536875958 did not exist on the partition Volume 536875958 on partition /vicepa server jeremiah.cs.uwm.edu deleted admin 86 % vos examine fa11.cs351.backup **** Could not attach volume 536875958 **** Volume fa11.cs351.backup does not exist in VLDB RWrite: 536875956 number of sites -> 1 server jeremiah.cs.uwm.edu partition /vicepa RW Site admin 87 % vos zap jeremiah a fa11.cs351.backup -force -verbose vos: forcibly removing all traces of volume 536875958, please wait...done. admin 88 % vos examine fa11.cs351.backup Could not fetch the information about volume 536875958 from the server : No such device Volume fa11.cs351.backup does not exist Dump only information from VLDB fa11.cs351 RWrite: 536875956 number of sites -> 1 server jeremiah.cs.uwm.edu partition /vicepa RW Site admin 89 % vos backup fa11.cs351 Failed to clone the volume 536875956 Volume not attached, does not exist, or not on line Error in vos backup command. Volume not attached, does not exist, or not on line admin 90 % vos examine fa11.cs351.backup Could not fetch the information about volume 536875958 from the server : No such device Volume fa11.cs351.backup does not exist Dump only information from VLDB fa11.cs351 RWrite: 536875956 number of sites -> 1 server jeremiah.cs.uwm.edu partition /vicepa RW Site admin 91 % vos backup fa11.cs351 Failed to start a transaction on the backup volume 536875958 Volume not attached, does not exist, or not on line Error in vos backup command. Volume not attached, does not exist, or not on line admin 92 % vos examine fa11.cs351.backup **** Could not attach volume 536875958 **** RWrite: 536875956 Backup: 536875958 number of sites -> 1 server jeremiah.cs.uwm.edu partition /vicepa RW Site admin 93 % vos create jeremiah a fa11.test Volume 536876132 created on partition /vicepa of jeremiah admin 94 % vos backup fa11.test Created backup volume for fa11.test admin 95 % vos examine fa11.test.backup fa11.test.backup 536876134 BK 2 K On-line jeremiah.cs.uwm.edu /vicepa RWrite 536876132 ROnly 0 Backup 0 MaxQuota 5000 K Creation Mon Aug 29 13:59:13 2011 Copy Mon Aug 29 13:59:13 2011 Backup Mon Aug 29 13:59:13 2011 Last Update Mon Aug 29 13:59:05 2011 0 accesses in the past day (i.e., vnode references) RWrite: 536876132 Backup: 536876134 number of sites -> 1 server jeremiah.cs.uwm.edu partition /vicepa RW Site [SOMEWHAT LATER, I had the bright idea to repeat my error, and to my surprise and delight(?) the same error happens!] admin 107 % vos move fa11.test jeremiah a jeremiah b Volume 536876132 moved from jeremiah /vicepa to jeremiah /vicepb admin 108 % vos move fa11.test jeremiah b jeremiah a Volume 536876132 moved from jeremiah /vicepb to jeremiah /vicepa admin 109 % vos backup fa11.test Failed to start a transaction on the backup volume 536876134 Volume not attached, does not exist, or not on line Error in vos backup command. Volume not attached, does not exist, or not on line admin 110 % vos examine fa11.test.backup **** Could not attach volume 536876134 **** RWrite: 536876132 Backup: 536876134 number of sites -> 1 server jeremiah.cs.uwm.edu partition /vicepa RW Site --------------------------------- NB: FileLog says: Mon Aug 29 13:57:10 2011 FSYNC_com_VolOff: failed to get heavyweight reference to volume 536875958 (state=20, flags=0x18) Salvaleg logs (I tried to salvage fa11.cs351.backup after the series of commands here, which said it was successful but the logs show: $ cat SalsrvLog 08/29/2011 13:43:00 dispatching child to salvage volume 536875956... 08/29/2011 13:43:00 2 nVolumesInInodeFile 56 08/29/2011 13:43:00 CHECKING CLONED VOLUME 536875958. 08/29/2011 13:43:00 SALVAGING VOLUME 536875956. 08/29/2011 13:43:00 fa11.cs351 (536875956) updated 08/29/2011 13:28 08/29/2011 13:43:00 totalInodes 513 08/29/2011 13:43:00 Salvaged fa11.cs351 (536875956): 506 files, 1679 blocks 08/29/2011 13:56:19 dispatching child to salvage volume 536875956... 08/29/2011 13:56:19 2 nVolumesInInodeFile 56 08/29/2011 13:56:19 CHECKING CLONED VOLUME 536875958. 08/29/2011 13:56:19 SALVAGING VOLUME 536875956. 08/29/2011 13:56:19 fa11.cs351 (536875956) updated 08/29/2011 13:28 08/29/2011 13:56:19 totalInodes 513 08/29/2011 13:56:20 Salvaged fa11.cs351 (536875956): 506 files, 1679 blocks and $ cat SalvageLog @(#) OpenAFS 1.6.0pre2 built 2011-02-17 (91.pre2.sl6@fnal.gov) 08/29/2011 14:01:49 STARTING AFS SALVAGER 2.4 (/usr/afs/bin/salvager /vicepa 536875958) 08/29/2011 14:01:49 _VLockFd: conflicting lock held on fd 4, offset 0 by pid 2177 (locktype=2) (process 2177 is /usr/afs/bin/salvageserver) From adeason@sinenomine.net Mon Aug 29 20:32:46 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 29 Aug 2011 14:32:46 -0500 Subject: [OpenAFS] Re: can neither delete nor use backup volume (DA server 1.6.0pre2) In-Reply-To: <488.1314645494@pabst.cs.uwm.edu> References: <488.1314645494@pabst.cs.uwm.edu> Message-ID: <20110829143247.f69f6d0b.adeason@sinenomine.net> On Mon, 29 Aug 2011 14:18:14 -0500 John Tang Boyland wrote: > NB: FileLog says: > Mon Aug 29 13:57:10 2011 FSYNC_com_VolOff: failed to get heavyweight reference to volume 536875958 (state=20, flags=0x18) 20 is VOL_STATE_DELETED. This issue was fixed in gerrit 4261, ee2811b0, in 1.6.0pre4. It affects a lot of things when you move volumes around, and then move them back to where they were originally. We're up to pre7 now, btw, with 1.6.0 imminent. If you want to be running prereleases, you really don't want to lag behind in them; there are many known issues in the older pres. -- Andrew Deason adeason@sinenomine.net From adeason@sinenomine.net Mon Aug 29 20:38:41 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Mon, 29 Aug 2011 14:38:41 -0500 Subject: [OpenAFS] Re: can neither delete nor use backup volume (DA server 1.6.0pre2) References: <488.1314645494@pabst.cs.uwm.edu> Message-ID: <20110829143841.e5d8bdc7.adeason@sinenomine.net> On Mon, 29 Aug 2011 14:18:14 -0500 John Tang Boyland wrote: > admin 87 % vos zap jeremiah a fa11.cs351.backup -force -verbose > vos: forcibly removing all traces of volume 536875958, please wait...done. Also just by the way, 'vos zap -force' is not a very smart tool, and it usually doesn't make a lot of sense for non-RW volumes. For RO and BK volumes, all it does is remove a few header files without touching any of the data, which is not what you want to do. If you have a troublesome non-RW volume, and 'vos remove/zap' and salvaging does not help, my general advice is just to stop right there (don't do 'vos zap -force'). Those two should handle every possible case, so if they are not handling it then there's a bug. Trying more things might make things worse or lose valuable debugging information. -- Andrew Deason adeason@sinenomine.net From aaronk@umbc.edu Mon Aug 29 21:03:42 2011 From: aaronk@umbc.edu (Aaron Knister) Date: Mon, 29 Aug 2011 16:03:42 -0400 Subject: [OpenAFS] Remove a mount point In-Reply-To: References: Message-ID: --20cf3078151099100104abaa659b Content-Type: text/plain; charset=ISO-8859-1 Hmm...I'm not sure what you're asking. Those commands should work on any system with the afs client installed, even if it also has the afs file server processes installed and running. If you're noticing inconsistent views between clients of directories/files within AFS, I'd venture to say that it's likely an issue with cache manager callback communication with the file servers. Double check that your clients are able to talk to your file server(s) on port 7001/udp (please correct me if that port/protocol isn't right). -Aaron On Sun, Aug 28, 2011 at 11:41 AM, jm130794 wrote: > Hello, > > I did that on the file server ... not from a client. > > I did that from a Linux client and it worked! > > What do you think about it ? > > JM > > > 2011/8/28 Aaron Knister > >> That in theory is an easy fix! >> >> fs rmmount dir1 >> >> >> On Sun, Aug 28, 2011 at 10:30 AM, jm130794 wrote: >> >>> Hello, >>> >>> I made a error when I created a mount point :( >>> >>> # vos create afs1 a cell.dir1 >>> # fs mkmount dir1 cell.dir1 >>> # vos remove -id cell.dir1 >>> >>> On afs1 fileserver, I don't see dir1 but, on unix and windows client, it >>> still appears. >>> >>> What should I do ? >>> >>> Thanks, >>> >>> JM >>> >>> >>> >> >> >> -- >> Aaron Knister >> Systems Administrator >> Division of Information Technology >> University of Maryland, Baltimore County >> aaronk@umbc.edu >> > > -- Aaron Knister Systems Administrator Division of Information Technology University of Maryland, Baltimore County aaronk@umbc.edu --20cf3078151099100104abaa659b Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hmm...I'm not sure what you're asking. Those commands should work o= n any system with the afs client installed, even if it also has the afs fil= e server processes installed and running. If you're noticing inconsiste= nt views between clients of directories/files within AFS, I'd venture t= o say that it's likely an issue with cache manager callback communicati= on with the file servers. Double check that your clients are able to talk t= o your file server(s) on port 7001/udp (please correct me if that port/prot= ocol isn't right).

-Aaron

On Sun, Aug 28, 2011 at 11:41 = AM, jm130794 <jm= 130794@gmail.com> wrote:
Hello,

I did that on the file server ... not from a client.

I= did that from a Linux client and it worked!

What do you think about= it ?

JM


2011/8/28 Aaron Knister <aaronk@umbc.edu>
That in theory is an ea= sy fix!

fs rmmount dir1


On Sun, Aug 28, 2011 at 10:30 AM, jm130794 <jm= 130794@gmail.com> wrote:
Hello,

I made a = error when I created a mount point :(

# vos create afs1 a cell.dir1<= br> # fs mkmount dir1 cell.dir1
# vos remove -id cell.dir1

On afs1 fileserver, I don't see dir1 = but, on unix and windows client, it still appears.

What should I do ?

Thanks,

JM





--
Aaron Knister
Systems Administrator
Division of Informat= ion Technology
University of Maryland, Baltimore County
aaronk@umbc.edu




--
Aaron Knist= er
Systems Administrator
Division of Information Technology
Univer= sity of Maryland, Baltimore County
aaronk@umbc.edu
--20cf3078151099100104abaa659b-- From jason@rampaginggeek.com Mon Aug 29 23:42:37 2011 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Mon, 29 Aug 2011 18:42:37 -0400 Subject: [OpenAFS] improving cache partition performance In-Reply-To: <740335677.7221.1314586837835.JavaMail.root@mailstore01.gopc.net> References: <740335677.7221.1314586837835.JavaMail.root@mailstore01.gopc.net> Message-ID: <4E5C15DD.8000503@rampaginggeek.com> Thanks Chris, Both of my servers have 16GB of RAM, and yes, I'm serving PHP content. I'm not sure if I can add more spindles in these boxes with my RAID controller. (1U Dell PowerEdge R210 with H200 RAID card.) I disabled APC cache because of a segfault every ~5000 hits, which happened to hit during an important demo by my boss. I was asking about commit time to see if there was a way to combine disk transactions by using more RAM, and to see how dangerous that might be. Jason On 08/28/2011 11:00 PM, Chris Hoy Poy wrote: > Hi Jason, > > RAM - cache more. > > Otherwise - add more spindles. > > Improve the local AFS cache for each webserver - is that local or disk ? > > I wouldn't mess around with the EXT3 settings, due to the way OpenAFS works. In my experience, messing with these settings - you might get a touch more output, but the time associated with it, compared with the cost of adding more spindles or adding more RAM, is typically not worth it. You aren't likely to get more then a few ms or a few more IOPS - either way. Best to push it out of RAM altogether. > > > Other things to check - are your operations READ bottlenecked or WRITE bottlenecked? Whats your proportion of read to write operations? If you are heavy on reads - definitely get more RAM into the system. If you are write bottlenecked - you need more spindles (though if your size is ~50-100GB, have a look at the SSD options - you can get some excellent raided SSDs fairly cheap these days). > > If the webserver is serving PHP content, look at APC cache to get (again) get the reads off the disk. > > > //chris > > ----- Original Message ----- > From: "Jason Edgecombe" > To: openafs-info@openafs.org > Sent: Monday, 29 August, 2011 5:00:23 AM > Subject: [OpenAFS] improving cache partition performance > > hi everyone, > > I'm looking at some performance problems on our web servers, and I think > that hard drive performance is a bottleneck. All filesystems are on a > RAID0 disk array with two 7200RPM disks (no battery-backed cache). The > cache is on a dedicated ext3 partition. At some points, my await times > on the disk are up to 1000ms. How can I improve this? Would setting > commit time larger than 5 seconds be OK? Would setting the ext3 journal > type to "data=writeback" be OK? > > platform: RHEL5.5 x86_64 > > Thanks, > Jason > _______________________________________________ OpenAFS-info mailing > list OpenAFS-info@openafs.org > https://lists.openafs.org/mailman/listinfo/openafs-info > From phalenor@gmail.com Tue Aug 30 00:10:56 2011 From: phalenor@gmail.com (Andy Cobaugh) Date: Mon, 29 Aug 2011 19:10:56 -0400 (EDT) Subject: [OpenAFS] improving cache partition performance In-Reply-To: <4E5C15DD.8000503@rampaginggeek.com> References: <740335677.7221.1314586837835.JavaMail.root@mailstore01.gopc.net> <4E5C15DD.8000503@rampaginggeek.com> Message-ID: Just to add some more datapoints to the discussion... Our webservers are HP DL360G5's, 14GB RAM. Pair of 36GB 15K 2.5" SAS drives in RAID-1. /usr/vice/cache is ext2 with noatime,nodiratime. These machines run dovecot IMAP, apache with lots of php applications, RT, and vsftpd serving anon and private ftp accounts. Serving content that isn't in the cache yet, we can get about 70-80MB/s depending on which fileserver it's coming from, and after it's cached, the gigabit network becomes the bottleneck. The cache partition is ~34GB in size, and we're running with these options: -dynroot -dynroot-sparse -fakestat -afsdb -nosettime -daemons 20 -stat 48000 -volumes 2048 -chunksize 19 -rxpck 2048 With those cache manager settings, cache partition utilization is sitting at about 92%. I can get even better numbers with memcache, and indeed most of our other machines are running with 2GB of memcache, I like seeing read performance in GB/s, and when most of your machines have 32GB or more (we have 3 with 256GB), a couple GB here and there won't have a noticeable impact. Jason: do you know in particular what kind of workload is causing issues for you? You mentioned your wait times are on the order of seconds, are you sure that's caused by the underlying disk? At the very least, I would try mounting your cache partition as ext2, as has already been suggested. Turning off atime and diratime shouldn't hurt, and if your disks are having issues with seaks, this should help some. Also, you really want to run 1.6.0pre7, or 1.6.0 when it shows up. Nothing wrong with 1.4, but if you're trying to get the most performance out of afs on modern hardware, switching to 1.6 gives you some real cheap gains. There are huge performance improvements on Linux going from 1.4 to 1.6, and all of my new installations are 1.6.0pre7 for that reason. Especially with disk-based caches, as Simon mentioned. 1.6.0pre7 gets write performance for disk caches almost on par with memcache, though read performance is still lacking, as memory will almost always be faster than disk, but disk will always be 'cheaper' than memory. Worth a try at least, and pre7 has been very stable in our environment. --andy From jason@rampaginggeek.com Tue Aug 30 00:39:14 2011 From: jason@rampaginggeek.com (Jason Edgecombe) Date: Mon, 29 Aug 2011 19:39:14 -0400 Subject: [OpenAFS] improving cache partition performance In-Reply-To: References: <740335677.7221.1314586837835.JavaMail.root@mailstore01.gopc.net> <4E5C15DD.8000503@rampaginggeek.com> Message-ID: <4E5C2322.30106@rampaginggeek.com> On 08/29/2011 07:10 PM, Andy Cobaugh wrote: > > Just to add some more datapoints to the discussion... > > Our webservers are HP DL360G5's, 14GB RAM. Pair of 36GB 15K 2.5" SAS > drives in RAID-1. /usr/vice/cache is ext2 with noatime,nodiratime. > These machines run dovecot IMAP, apache with lots of php applications, > RT, and vsftpd serving anon and private ftp accounts. > > Serving content that isn't in the cache yet, we can get about > 70-80MB/s depending on which fileserver it's coming from, and after > it's cached, the gigabit network becomes the bottleneck. The cache > partition is ~34GB in size, and we're running with these options: > > -dynroot -dynroot-sparse -fakestat -afsdb -nosettime -daemons 20 > -stat 48000 -volumes 2048 -chunksize 19 -rxpck 2048 > > With those cache manager settings, cache partition utilization is > sitting at about 92%. I can get even better numbers with memcache, and > indeed most of our other machines are running with 2GB of memcache, I > like seeing read performance in GB/s, and when most of your machines > have 32GB or more (we have 3 with 256GB), a couple GB here and there > won't have a noticeable impact. > > Jason: do you know in particular what kind of workload is causing > issues for you? You mentioned your wait times are on the order of > seconds, are you sure that's caused by the underlying disk? At the > very least, I would try mounting your cache partition as ext2, as has > already been suggested. Turning off atime and diratime shouldn't hurt, > and if your disks are having issues with seaks, this should help some. > > Also, you really want to run 1.6.0pre7, or 1.6.0 when it shows up. > Nothing wrong with 1.4, but if you're trying to get the most > performance out of afs on modern hardware, switching to 1.6 gives you > some real cheap gains. There are huge performance improvements on > Linux going from 1.4 to 1.6, and all of my new installations are > 1.6.0pre7 for that reason. Especially with disk-based caches, as Simon > mentioned. 1.6.0pre7 gets write performance for disk caches almost on > par with memcache, though read performance is still lacking, as memory > will almost always be faster than disk, but disk will always be > 'cheaper' than memory. > > Worth a try at least, and pre7 has been very stable in our environment. I was told that noatime is bad for an AFS cache partition because AFS uses the atime to know when the cache entry was last accessed. Is that the case? Jason From phalenor@gmail.com Tue Aug 30 00:44:02 2011 From: phalenor@gmail.com (Andy Cobaugh) Date: Mon, 29 Aug 2011 19:44:02 -0400 (EDT) Subject: [OpenAFS] improving cache partition performance In-Reply-To: <4E5C2322.30106@rampaginggeek.com> References: <740335677.7221.1314586837835.JavaMail.root@mailstore01.gopc.net> <4E5C15DD.8000503@rampaginggeek.com> <4E5C2322.30106@rampaginggeek.com> Message-ID: On 2011-08-29 at 19:39, Jason Edgecombe ( jason@rampaginggeek.com ) said: > I was told that noatime is bad for an AFS cache partition because AFS uses > the atime to know when the cache entry was last accessed. Oops, looks like you're right, unless someone more knowledgeable says otherwise. --andy From boyland@cs.uwm.edu Tue Aug 30 02:00:37 2011 From: boyland@cs.uwm.edu (John Tang Boyland) Date: Mon, 29 Aug 2011 20:00:37 -0500 Subject: [OpenAFS] Re: can neither delete nor use backup volume (DA server 1.6.0pre2) Message-ID: <1009.1314666037@pabst.cs.uwm.edu> Andrew Deason wrote: ] On Mon, 29 Aug 2011 14:18:14 -0500 ] John Tang Boyland wrote: ] ] > NB: FileLog says: ] > Mon Aug 29 13:57:10 2011 FSYNC_com_VolOff: failed to get heavyweight reference to volume 536875958 (state=20, flags=0x18) ] ] 20 is VOL_STATE_DELETED. This issue was fixed in gerrit 4261, ee2811b0, ] in 1.6.0pre4. It affects a lot of things when you move volumes around, ] and then move them back to where they were originally. ] ] We're up to pre7 now, btw, with 1.6.0 imminent. If you want to be ] running prereleases, you really don't want to lag behind in them; there ] are many known issues in the older pres. Good point. I'm running the latest openafs-server from Scientific Linux 6. Hopefully they will release a later RPM soon. ] > admin 87 % vos zap jeremiah a fa11.cs351.backup -force -verbose ] > vos: forcibly removing all traces of volume 536875958, please wait...done. ] ] Also just by the way, 'vos zap -force' is not a very smart tool, and it ] usually doesn't make a lot of sense for non-RW volumes. For RO and BK ] volumes, all it does is remove a few header files without touching any ] of the data, which is not what you want to do. ] ] If you have a troublesome non-RW volume, and 'vos remove/zap' and ] salvaging does not help, my general advice is just to stop right there ] (don't do 'vos zap -force'). Those two should handle every possible ] case, so if they are not handling it then there's a bug. Trying more ] things might make things worse or lose valuable debugging information. OK. That's useful for the future. I'll stay away from zap -force. (An earlier "zap" had "error 4".) Even if SL6 releases a later RPM for openafs-server soon, I suppose the volume is irretrievably broken. Time to create a new volume, I guess. I suppose then I can try to remove/zap the whole RW volume and maybe get back into a clean state. Thanks, John From yentj@infowrap.com.tw Tue Aug 30 06:28:20 2011 From: yentj@infowrap.com.tw (Ting-jen Yen) Date: Tue, 30 Aug 2011 13:28:20 +0800 Subject: [OpenAFS] Re: 64-bit mod_waklog / filedrawers on afs In-Reply-To: <20110824103459.8b0a3867.adeason@sinenomine.net> References: <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBDF@EXEVS04.its.uncc.edu> <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBE2@EXEVS04.its.uncc.edu> <61FBF8EDE8FE9A4D9A52D5C47531320D19D9EBE3@EXEVS04.its.uncc.edu> <20110824103459.8b0a3867.adeason@sinenomine.net> Message-ID: <1314682100.6133.16.camel@yentj.infowrap.com.tw> On Web=A1A24 Aug 2011 10:56:00 -0400 Andrew Deason wrote: > On Mon, 22 Aug 2011 09:54:16 -0400 > "Colson, Daniel" wrote: >=20 > > Oops. I apologize. I sent the output from the filedrawers build > > instead of the mod_waklog build. I'll check on mysql-devel though. > > Below is the output from "make" for mod_waklog. > >=20 > >=20 > > `/usr/sbin/apxs -q CC` -I/usr/include/ -I/usr/include/afs -I/usr/incl= ude/httpd -DEAPI -g `/usr/sbin/apxs -q CFLAGS` -c mod_waklog.c -o mod_wak= log.o >=20 > Did you try using the patches Aaron mentioned earlier in the thread? I'= m > not too familiar with mod_waklog, but this looks like a build line > without Aaron's fix_build_scripts.patch. >=20 Hello, The output after calling ioctl() of "strace -f afsd" is in the following, (compiler errors are after this output message.) --------- strace -f afsd line 155----------------------- open("/proc/fs/openafs/afs_ioctl", O_RDWR) =3D 3 ioctl(3, CAPI_REGISTER or SNDCTL_COPR_LOAD, 0xffa4c320) =3D -1 EINVAL (Invalid argument) close(3) =3D 0 clone(Process 19813 attached (waiting for parent) Process 19813 resumed (parent 19812 ready) child_stack=3D0, flags=3DCLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=3D0xfff1e48) =3D 19813 [pid 19812] clone(Process 19814 attached (waiting for parent) Process 19814 resumed (parent 19812 ready) child_stack=3D0, flags=3DCLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=3D0xfff1e48) =3D 19814 [pid 19812] clone(Process 19815 attached (waiting for parent) Process 19815 resumed (parent 19812 ready) child_stack=3D0, flags=3DCLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=3D0xfff1e48) =3D 19815 [pid 19812] open("/proc/fs/openafs/afs_ioctl", O_RDWR) =3D 3 [pid 19813] setpriority(PRIO_PROCESS, 0, -10 [pid 19815] setpriority(PRIO_PROCESS, 0, -10 [pid 19814] open("/proc/fs/openafs/afs_ioctl", O_RDWR [pid 19815] <... setpriority resumed> ) =3D 0 [pid 19813] <... setpriority resumed> ) =3D 0 [pid 19815] open("/proc/fs/openafs/afs_ioctl", O_RDWR [pid 19814] <... open resumed> ) =3D 3 [pid 19815] <... open resumed> ) =3D 3 [pid 19813] open("/proc/fs/openafs/afs_ioctl", O_RDWR [pid 19815] ioctl(3, CAPI_REGISTER or SNDCTL_COPR_LOAD [pid 19814] ioctl(3, CAPI_REGISTER or SNDCTL_COPR_LOAD [pid 19812] ioctl(3, CAPI_REGISTER or SNDCTL_COPR_LOAD [pid 19815] <... ioctl resumed> , 0xffa4c290) =3D -1 EINVAL (Invalid argument) [pid 19813] <... open resumed> ) =3D 3 [pid 19815] close(3 [pid 19813] ioctl(3, CAPI_REGISTER or SNDCTL_COPR_LOAD [pid 19815] <... close resumed> ) =3D 0 [pid 19813] <... ioctl resumed> , 0xffa4c290) =3D -1 EINVAL (Invalid argument) [pid 19813] close(3 [pid 19815] exit_group(1) =3D ? Process 19815 detached [pid 19813] <... close resumed> ) =3D 0 [pid 19813] exit_group(1) =3D ? Process 19813 detached [pid 19812] <... ioctl resumed> , 0xffa4c320) =3D -1 EINVAL (Invalid argument) [pid 19812] --- SIGCHLD (Child exited) @ 0 (0) --- [pid 19812] close(3) =3D 0 [pid 19812] fstat64(1, {st_mode=3DS_IFCHR|0620, st_rdev=3Dmakedev(136, 0), ...}) =3D 0 [pid 19814] <... ioctl resumed> , 0xffa4c290) =3D -1 EINVAL (Invalid argument) [pid 19812] mmap(NULL, 65536, PROT_READ|PROT_WRITE, MAP_PRIVATE| MAP_ANONYMOUS, -1, 0) =3D 0xf7fe0000 [pid 19812] write(1, "afsd: Error -1 in basic initiali"..., 40afsd: Error -1 in basic initialization. ) =3D 40 [pid 19812] exit_group(1) =3D ? [pid 19814] close(3) =3D 0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D For compiler errors, if I simple run rpmbuild -bb openafs.spec, it=20 shows: (using SRPM of openafs-1.4.14.1-1.1.src.rpm) ---------------------------------------------------- [root@r602 SPECS]# rpmbuild -bb openafs.spec error: parse error in expression error: /usr/src/redhat/SPECS/openafs.spec:337: parseExpressionBoolean returns -1 Checking for unpackaged file(s): /usr/lib/rpm/check-files /var/tmp/openafs-1.4.14.1-root =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D If I specified build_userspace, build_modules, and build_authlibs in rpmbuild, the compiling will go ahead, and stop at: ------------------------------------------------ cc -o aklog -DALLOW_REGISTER -I/usr/src/redhat/BUILD/openafs-1.4.14.1/src/config -I. -I. -I/usr/src/redhat/BUILD/openafs-1.4.14.1/include -I/usr/src/redhat/BUILD/openafs-1.4.14.1/include/afs -I/usr/src/redhat/BUILD/openafs-1.4.14.1/include/rx .... (line to long, so I deleted the rest of it) /usr/bin/ld: cannot find -lcom_err collect2: ld returned 1 exit status =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D I did notice I have /usr/lib64/libcom_err.so which is symlink to /lib64/libcom_err.so.2, though there is no /usr/lib/libcom_err.so. I supposed this is part of 64/32-bitness. I did manually add a link /usr/lib/libcom_err.so to /lib/libcom_err.so.2. The compiling would now go past the previous part, and no error until it began to compile kernel module. It would dump a lot of warning/error messages, like: ------------------------------------------------- /usr/src/redhat/BUILD/openafs-1.4.14.1/src/libafs/MODLOAD-2.6.18-194.el5-= SP/afs_ vnop_flock.c:617: warning: 'struct flock64' declared inside parameter list /usr/src/redhat/BUILD/openafs-1.4.14.1/src/libafs/MODLOAD-2.6.18-194.el5-= SP/afs_ vnop_flock.c:620: error: storage size of 'flock' isn't known /usr/src/redhat/BUILD/openafs-1.4.14.1/src/libafs/MODLOAD-2.6.18-194.el5-= SP/afs_ vnop_flock.c:632: error: dereferencing pointer to incomplete type =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D I try to get a preprocessed code output (using -E) of the=20 afs_vnop_flock.c, and noticed that struct flock64 is never defined. So I went to edit src/afs/afs.h, and changed the only occurance of 'flock64' to 'flock', and the whole compiling would then succeed. Though the service cannot be started. Because whole compiler output is too large even after compressed, I only included the most significant parts, (in my opinion anyway), if it is really necessary to show more information, please let me know, I will see what I can do then. -- Ting-jen From pontius@btv.ibm.com Tue Aug 30 12:32:30 2011 From: pontius@btv.ibm.com (Dale Pontius) Date: Tue, 30 Aug 2011 07:32:30 -0400 Subject: [OpenAFS] improving cache partition performance In-Reply-To: References: <740335677.7221.1314586837835.JavaMail.root@mailstore01.gopc.net> <4E5C15DD.8000503@rampaginggeek.com> <4E5C2322.30106@rampaginggeek.com> Message-ID: <4E5CCA4E.5090907@btv.ibm.com> On 08/29/2011 07:44 PM, Andy Cobaugh wrote: > On 2011-08-29 at 19:39, Jason Edgecombe ( jason@rampaginggeek.com ) said: >> I was told that noatime is bad for an AFS cache partition because AFS >> uses the atime to know when the cache entry was last accessed. > > Oops, looks like you're right, unless someone more knowledgeable says > otherwise. > Is "relatime" sufficient for AFS usage? It's now the default, and may not be as fine-grained as old atime. --- Dale Pontius Senior Engineer IBM Corporation Phone: (802) 769-6850 Tie-Line: 446-6850 email: pontius@us.ibm.com This e-mail and its attachments, if any, may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply e-mail and delete all copies of this message from your system without copying it and notify sender of the misdirection by reply e-mail. From marc.c.dionne@gmail.com Tue Aug 30 16:27:52 2011 From: marc.c.dionne@gmail.com (Marc Dionne) Date: Tue, 30 Aug 2011 11:27:52 -0400 Subject: [OpenAFS] improving cache partition performance In-Reply-To: References: <740335677.7221.1314586837835.JavaMail.root@mailstore01.gopc.net> <4E5C15DD.8000503@rampaginggeek.com> <4E5C2322.30106@rampaginggeek.com> Message-ID: On Mon, Aug 29, 2011 at 19:44, Andy Cobaugh wrote: > On 2011-08-29 at 19:39, Jason Edgecombe ( jason@rampaginggeek.com ) said: >> >> I was told that noatime is bad for an AFS cache partition because AFS uses >> the atime to know when the cache entry was last accessed. > > Oops, looks like you're right, unless someone more knowledgeable says > otherwise. > > --andy Someone can correct me, but reading the code it looks to me like the cache files' atime is only used at startup to initialize the cache manager's LRU mechanism. After that, the cache manager takes care of maintaining counters in an internal structure whenever cache files are used, without relying on the atime of the underlying filesystem. If that's correct, noatime might cause you not to do an ideal (LRU) usage of files soon after startup, but probably has little impact in the long run. Marc From adeason@sinenomine.net Tue Aug 30 16:30:46 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Tue, 30 Aug 2011 10:30:46 -0500 Subject: [OpenAFS] Re: improving cache partition performance References: <740335677.7221.1314586837835.JavaMail.root@mailstore01.gopc.net> <4E5C15DD.8000503@rampaginggeek.com> <4E5C2322.30106@rampaginggeek.com> Message-ID: <20110830103046.e909ea9b.adeason@sinenomine.net> On Mon, 29 Aug 2011 19:39:14 -0400 Jason Edgecombe wrote: > I was told that noatime is bad for an AFS cache partition because AFS > uses the atime to know when the cache entry was last accessed. > > Is that the case? atime is only read when the client starts up and scans the extant cache files. And the only thing we use it for is when determining which cache entries to kick out when we get full. It's not going to break if you turn it off, and the atime accuracy is a little dubious anyway, since we don't issue a read for the cache fs on every read from afs. I wouldn't think any benefit from the cache eviction decisions would outweigh the penalties for atime updates in the long run. -- Andrew Deason adeason@sinenomine.net From adeason@sinenomine.net Tue Aug 30 16:33:03 2011 From: adeason@sinenomine.net (Andrew Deason) Date: Tue, 30 Aug 2011 10:33:03 -0500 Subject: [OpenAFS] Re: can neither delete nor use backup volume (DA server 1.6.0pre2) In-Reply-To: <1009.1314666037@pabst.cs.uwm.edu> References: <1009.1314666037@pabst.cs.uwm.edu> Message-ID: <20110830103303.eaec1c13.adeason@sinenomine.net> On Mon, 29 Aug 2011 20:00:37 -0500 John Tang Boyland wrote: > Even if SL6 releases a later RPM for openafs-server soon, I suppose > the volume is irretrievably broken. Time to create a new volume, > I guess. I suppose then I can try to remove/zap the whole RW volume > and maybe get back into a clean state. No sorry, I didn't mean to say that at all. If you run with a newer server, the volume should be usable, and a salvage should clear up any lingering problems if any exist. -- Andrew Deason adeason@sinenomine.net From Ken@Elkabany.com Wed Aug 31 09:07:45 2011 From: Ken@Elkabany.com (Ken Elkabany) Date: Wed, 31 Aug 2011 01:07:45 -0700 Subject: [OpenAFS] Tips for increasing throughput Message-ID: --20cf301d4272b9c38d04abc8a0a0 Content-Type: text/plain; charset=ISO-8859-1 Hi, On a network capable of scp-ing files between machines at 60MB/sec, we are only able to achieve 2-3MB/sec of throughput when using AFS. We've been conducting tests on 300MB files, by copying them from the /afs/* mount to the local filesystem with memcache enabled. In production, we will not be using memcache, but we wanted to eliminate the overhead of the on-disk cache manager. We've tried tweaking our options as follows: "-verbose -nosettime -memcache -chunksize 20 -stat 2800 -daemons 5 -volumes 128". The most important option is the chunksize, which did not have a measurable effect. Any other tips? Thanks! Ken --20cf301d4272b9c38d04abc8a0a0 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi,

On a network capable of scp-ing files between machin= es at 60MB/sec, we are only able to achieve 2-3MB/sec of throughput when us= ing AFS. We've been conducting tests on 300MB files, by copying them fr= om the /afs/* mount to the local filesystem with memcache enabled. In produ= ction, we will not be using memcache, but we wanted to eliminate the overhe= ad of the on-disk cache manager. We've tried tweaking our options as fo= llows: "-verbose -nosettime -memcache -chunksize 20 -stat 2800 -daemon= s 5 -volumes 128". The most important option is the chunksize, which d= id not have a measurable effect. Any other tips?

Thanks!
Ken
--20cf301d4272b9c38d04abc8a0a0-- From jaltman@secure-endpoints.com Wed Aug 31 13:28:18 2011 From: jaltman@secure-endpoints.com (Jeffrey Altman) Date: Wed, 31 Aug 2011 08:28:18 -0400 Subject: [OpenAFS] Tips for increasing throughput In-Reply-To: References: Message-ID: <4E5E28E2.6000809@secure-endpoints.com> This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigAA09A791B31B291831A6BB13 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On 8/31/2011 4:07 AM, Ken Elkabany wrote: > Hi, >=20 > On a network capable of scp-ing files between machines at 60MB/sec, we > are only able to achieve 2-3MB/sec of throughput when using AFS. We've > been conducting tests on 300MB files, by copying them from the /afs/* > mount to the local filesystem with memcache enabled. In production, we > will not be using memcache, but we wanted to eliminate the overhead of > the on-disk cache manager. We've tried tweaking our options as follows:= > "-verbose -nosettime -memcache -chunksize 20 -stat 2800 -daemons 5 > -volumes 128". The most important option is the chunksize, which did no= t > have a measurable effect. Any other tips? >=20 > Thanks! > Ken Using what versions of OpenAFS on the client and file server? --------------enigAA09A791B31B291831A6BB13 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (MingW32) iQEcBAEBAgAGBQJOXijiAAoJENxm1CNJffh4qTQH/jBOjuYOILwB349krDfe2tYc VB/CfLymnwVVs63lwCE9qID48ekBavg5QwLZSmmlPerQHy7I8kaKfv/Y+r5/dENK Q+gLwyZyhr5HcLgkLXJIy7cLpGvQ4JjnWg8ml43RBTd+XVBOcNifwuk6d8bd163t KsdRID1DwO0Q92SS45OnbekJUr/1DhyOBHcKkPt9lHZiWtfH1t8hO27sHVLijeOg HEp7LqvObB9zMU7vYSRywFX4P/syu43w/D06tqBnCHVEWQhgm9YxK96hNDvKFm4z du5E0wHd1I398v6UHdXlAOVH7RI2rp4heGE0KpcfQaG9ySWrbzNMnRDrT2W4V70= =xts1 -----END PGP SIGNATURE----- --------------enigAA09A791B31B291831A6BB13-- From matt@linuxbox.com Wed Aug 31 14:01:29 2011 From: matt@linuxbox.com (Matt W. Benjamin) Date: Wed, 31 Aug 2011 09:01:29 -0400 (EDT) Subject: [OpenAFS] Tips for increasing throughput In-Reply-To: Message-ID: <887429386.6.1314795689199.JavaMail.root@thunderbeast.private.linuxbox.com> Hi, I think some more detail on the network/network path would be helpful. Regards, Matt ----- "Ken Elkabany" wrote: > Hi, > > > On a network capable of scp-ing files between machines at 60MB/sec, we > are only able to achieve 2-3MB/sec of throughput when using AFS. We've > been conducting tests on 300MB files, by copying them from the /afs/* > mount to the local filesystem with memcache enabled. In production, we > will not be using memcache, but we wanted to eliminate the overhead of > the on-disk cache manager. We've tried tweaking our options as > follows: "-verbose -nosettime -memcache -chunksize 20 -stat 2800 > -daemons 5 -volumes 128". The most important option is the chunksize, > which did not have a measurable effect. Any other tips? > > > Thanks! > Ken -- Matt Benjamin The Linux Box 206 South Fifth Ave. Suite 150 Ann Arbor, MI 48104 http://linuxbox.com tel. 734-761-4689 fax. 734-769-8938 cel. 734-216-5309 From Ken@Elkabany.com Wed Aug 31 21:43:04 2011 From: Ken@Elkabany.com (Ken Elkabany) Date: Wed, 31 Aug 2011 13:43:04 -0700 Subject: [OpenAFS] Tips for increasing throughput In-Reply-To: <887429386.6.1314795689199.JavaMail.root@thunderbeast.private.linuxbox.com> References: <887429386.6.1314795689199.JavaMail.root@thunderbeast.private.linuxbox.com> Message-ID: --0016e64aff62f20d2404abd32d38 Content-Type: text/plain; charset=ISO-8859-1 We're using 1.4.14 on the file servers (Ubuntu Natty) and 1.4.12 on the clients (Ubuntu Lucid). We don't know many details about the network topology as our servers are sitting on EC2, though as I stated previously, our scp benchmarks are significantly faster. The fileserver configuration is the default from the repository: /usr/lib/openafs/fileserver -p 23 -busyat 600 -rxpck 400 -s 1200 -l 1200 -cb 65535 -b 240 -vc 1200 On Wed, Aug 31, 2011 at 6:01 AM, Matt W. Benjamin wrote: > Hi, > > I think some more detail on the network/network path would be helpful. > > Regards, > > Matt > > ----- "Ken Elkabany" wrote: > > > Hi, > > > > > > On a network capable of scp-ing files between machines at 60MB/sec, we > > are only able to achieve 2-3MB/sec of throughput when using AFS. We've > > been conducting tests on 300MB files, by copying them from the /afs/* > > mount to the local filesystem with memcache enabled. In production, we > > will not be using memcache, but we wanted to eliminate the overhead of > > the on-disk cache manager. We've tried tweaking our options as > > follows: "-verbose -nosettime -memcache -chunksize 20 -stat 2800 > > -daemons 5 -volumes 128". The most important option is the chunksize, > > which did not have a measurable effect. Any other tips? > > > > > > Thanks! > > Ken > > -- > > Matt Benjamin > > The Linux Box > 206 South Fifth Ave. Suite 150 > Ann Arbor, MI 48104 > > http://linuxbox.com > > tel. 734-761-4689 > fax. 734-769-8938 > cel. 734-216-5309 > --0016e64aff62f20d2404abd32d38 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable We= 're using 1.4.14 on the file servers (Ubuntu Natty) and 1.4.12 on the c= lients (Ubuntu Lucid). We don't know many details about the network top= ology as our servers are sitting on EC2, though as I stated previously, our= scp benchmarks are significantly faster.

The fileserver configuration is the default from the reposit= ory: /usr/lib/openafs/fileserver -p 23 -busyat 600 -rxpck 400 -s 1200 -l 12= 00 -cb 65535 -b 240 -vc 1200

On Wed, Aug 31, 2011 at 6:01 AM, Matt W. Benjamin <matt@linuxbox.com> wrote:
Hi,

I think some more detail on the network/network path would be helpful.

Regards,

Matt

----- "Ken Elkabany" <Ken@Elkabany.com> wrote:

> Hi,
>
>
> On a network capable of scp-ing files between machines at 60MB/sec, we=
> are only able to achieve 2-3MB/sec of throughput when using AFS. We= 9;ve
> been conducting tests on 300MB files, by copying them from the /afs/*<= br> > mount to the local filesystem with memcache enabled. In production, we=
> will not be using memcache, but we wanted to eliminate the overhead of=
> the on-disk cache manager. We've tried tweaking our options as
> follows: "-verbose -nosettime -memcache -chunksize 20 -stat 2800<= br> > -daemons 5 -volumes 128". The most important option is the chunks= ize,
> which did not have a measurable effect. Any other tips?
>
>
> Thanks!
> Ken

--

Matt Benjamin

The Linux Box
206 South Fifth Ave. Suite 150
Ann Arbor, MI =A048104

http://linuxbox.com
tel. 734-761-4689 fax. 734-769-8938 cel. 734-216-5309

--0016e64aff62f20d2404abd32d38-- From matt@linuxbox.com Wed Aug 31 21:51:20 2011 From: matt@linuxbox.com (Matt W. Benjamin) Date: Wed, 31 Aug 2011 16:51:20 -0400 (EDT) Subject: [OpenAFS] Tips for increasing throughput In-Reply-To: <244592002.66.1314823522769.JavaMail.root@thunderbeast.private.linuxbox.com> Message-ID: <1412193856.70.1314823880066.JavaMail.root@thunderbeast.private.linuxbox.com> Hi, Well, others will have more to say about versions, but I imagine the recommendation will be to have more modern servers and clients. Also, as you suggest, the EC2 environment might very well put afs at a disadvantage, for example, there might be NAT (though I have no real idea). It's not really surprising that AFS doesn't compete well with SCP in this environment, though again, people have been working on performance of the existing UDP transport, as well as on a TCP transport which I might expect to do better. Matt ----- "Ken Elkabany" wrote: > We're using 1.4.14 on the file servers (Ubuntu Natty) and 1.4.12 on > the clients (Ubuntu Lucid). We don't know many details about the > network topology as our servers are sitting on EC2, though as I stated > previously, our scp benchmarks are significantly faster. > > > The fileserver configuration is the default from the repository: > /usr/lib/openafs/fileserver -p 23 -busyat 600 -rxpck 400 -s 1200 -l > 1200 -cb 65535 -b 240 -vc 1200 > > > > On Wed, Aug 31, 2011 at 6:01 AM, Matt W. Benjamin < matt@linuxbox.com > > wrote: > > > Hi, > > I think some more detail on the network/network path would be helpful. > > Regards, > > Matt > > > > > ----- "Ken Elkabany" wrote: > > > Hi, > > > > > > On a network capable of scp-ing files between machines at 60MB/sec, > we > > are only able to achieve 2-3MB/sec of throughput when using AFS. > We've > > been conducting tests on 300MB files, by copying them from the > /afs/* > > mount to the local filesystem with memcache enabled. In production, > we > > will not be using memcache, but we wanted to eliminate the overhead > of > > the on-disk cache manager. We've tried tweaking our options as > > follows: "-verbose -nosettime -memcache -chunksize 20 -stat 2800 > > -daemons 5 -volumes 128". The most important option is the > chunksize, > > which did not have a measurable effect. Any other tips? > > > > > > Thanks! > > Ken > > -- > > Matt Benjamin > > The Linux Box > 206 South Fifth Ave. Suite 150 > Ann Arbor, MI 48104 > > http://linuxbox.com > > tel. 734-761-4689 > fax. 734-769-8938 > cel. 734-216-5309 -- Matt Benjamin The Linux Box 206 South Fifth Ave. Suite 150 Ann Arbor, MI 48104 http://linuxbox.com tel. 734-761-4689 fax. 734-769-8938 cel. 734-216-5309