Your message dated Fri, 26 Aug 2005 10:32:07 -0700
with message-id <[EMAIL PROTECTED]>
and subject line Bug#249315: fixed in openafs 1.4rc1-1
has caused the attached Bug report to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere. Please contact me immediately.)
Debian bug tracking system administrator
(administrator, Debian Bugs database)
--------------------------------------
Received: (at submit) by bugs.debian.org; 16 May 2004 17:35:56 +0000
>From [EMAIL PROTECTED] Sun May 16 10:35:56 2004
Return-path: <[EMAIL PROTECTED]>
Received: from mail.proteosys.com [213.139.130.197]
by spohr.debian.org with esmtp (Exim 3.35 1 (Debian))
id 1BPPYK-0003q2-00; Sun, 16 May 2004 10:35:56 -0700
Received: by mail.proteosys.com (8.12.10/8.12.2) with ESMTP id i4GHZnlq007780;
Sun, 16 May 2004 19:35:50 +0200
Date: Sun, 16 May 2004 19:35:55 +0200 (CEST)
From: =?ISO-8859-1?Q?Rainer_Sch=F6pf?= <[EMAIL PROTECTED]>
X-X-Sender: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
cc: =?ISO-8859-1?Q?Rainer_Sch=F6pf?= <[EMAIL PROTECTED]>
Subject: openafs-modules-source: xfs filesystem for AFS cache partition
results in kernel Oops
Message-ID: <[EMAIL PROTECTED]>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=ISO-8859-1
X-ProteoSys-SPAM-Score: 3.271 (***)
RCVD_IN_DYNABLOCK,RCVD_IN_NJABL,RCVD_IN_NJABL_DIALUP,RCVD_IN_SORBS
X-Scanned-By: MIMEDefang 2.28 (www . roaringpenguin . com / mimedefang)
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mail.proteosys.com id
i4GHZnlq007780
Delivered-To: [EMAIL PROTECTED]
X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2004_03_25
(1.212-2003-09-23-exp) on spohr.debian.org
X-Spam-Status: No, hits=-8.0 required=4.0 tests=BAYES_00,HAS_PACKAGE
autolearn=no version=2.60-bugs.debian.org_2004_03_25
X-Spam-Level:
Package: openafs-modules-source
Version: 1.2.11-1
I have recently experimented with openafs. One particular issue is choice
of underlying filesystem for server and client.
If I put the afs client cache on an xfs partition, starting up afsd on an
initialized client cache results in a kernel Oops. I found out (see below=
)
that xfs is not supported for the cache partition. I did not find this
mentioned in the documentation.
This fact should either be clearly documented or (preferably) fixed. Even
if it's a restriction, it should not lead to a kernel NULL pointer. I'd
rather have afsd issue some warning like "cannot use cache on xfs
filesystem" and terminate gracefully when it first initializes the cache.
The gory details:
Kernel 2.4.26 on 686.
The traceback from ksymoops looks typically like this:
May 16 07:43:24 localhost kernel: Starting AFS cache scan...<1>Unable to =
handle kernel NULL pointer dereference at virtual
address 00000000
May 16 07:43:24 localhost kernel: 00000000
May 16 07:43:24 localhost kernel: Oops: 0000
May 16 07:43:24 localhost kernel: CPU: 0
May 16 07:43:24 localhost kernel: EIP: 0010:[<00000000>] Tainted: P=
F
May 16 07:43:24 localhost kernel: EFLAGS: 00010282
May 16 07:43:24 localhost kernel: eax: f8961760 ebx: f70a3d7c ecx: 00=
000001 edx: f7ca4000
May 16 07:43:24 localhost kernel: esi: f7d7dcc0 edi: f7d7dd30 ebp: f7=
0e9c3c esp: f70e9c10
May 16 07:43:24 localhost kernel: ds: 0018 es: 0018 ss: 0018
May 16 07:43:24 localhost kernel: Process afsd (pid: 825, stackpage=3Df70=
e9000)
May 16 07:43:24 localhost kernel: Stack: f8ab084a f70a3d7c 00000000 00000=
046 00000046 f70700e0 f7070150 f70a3c3c
May 16 07:43:24 localhost kernel: 00000000 f8cda000 00000000 f70e9=
c8c f8a775d2 02402d81 00000000 00000339
May 16 07:43:24 localhost kernel: 00000339 f70700e0 f7070150 00000=
001 f8961460 f70700e0 08070520 f70e9c8c
May 16 07:43:24 localhost kernel: Call Trace: [<f8ab084a>] [<f8a775d2>=
]
[af_packet:__insmod_af_packet_O/lib/modules/2.4.26-1-686/kernel/net/pa+-1=
067936/96] [<f8a79f1b>] [<f8ab4e6a>]
May 16 07:43:24 localhost kernel: [<f8ab543f>] [system_call+51/56]
May 16 07:43:24 localhost kernel: Code: Bad EIP value.
>>EIP; 00000000 Before first symbol
>>eax; f8961760 <[xfs]linvfs_sops+0/60>
>>ebx; f70a3d7c <_end+36da031c/38509600>
>>edx; f7ca4000 <_end+379a05a0/38509600>
>>esi; f7d7dcc0 <_end+37a7a260/38509600>
>>edi; f7d7dd30 <_end+37a7a2d0/38509600>
>>ebp; f70e9c3c <_end+36de61dc/38509600>
>>esp; f70e9c10 <_end+36de61b0/38509600>
Trace; f8ab084a <[openafs]osi_UFSOpen+1ba/240>
Trace; f8a775d2 <[openafs]afs_InitCacheFile+102/340>
Trace; f8ab543f <[openafs]afs_syscall+17f/1b0>
The NULL pointer happens inside osi_UFSOpen in src/afs/LINUX/osi_file.c,
during the call to iget. The important point here is that it is not
possible to call iget on an xfs inode. That is because iget in turn calls
the filesystem's read_inode function, which doesn't exist for xfs (see
include/linux/fs.h and fs/xfs/linux/xfs_super.c in the Linux kernel tree).
One way to correct this is suggested by the IRIX specific code in
src/afs/IRIX/osi_file.c and src/afs/IRIX/osi_inode.c. But I don't have th=
e
time for implementing that. If someone is working on it I'd be happy to
try it out.
As an aside: I'm not sure that it works with reiserfs either, which has
read_inode2 instead of read_inode. But I didn't really check.
Rainer Sch=F6pf
---------------------------------------
Received: (at 249315-close) by bugs.debian.org; 26 Aug 2005 17:40:11 +0000
>From [EMAIL PROTECTED] Fri Aug 26 10:40:11 2005
Return-path: <[EMAIL PROTECTED]>
Received: from katie by spohr.debian.org with local (Exim 3.36 1 (Debian))
id 1E8i3j-00060e-00; Fri, 26 Aug 2005 10:32:07 -0700
From: Russ Allbery <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
X-Katie: $Revision: 1.56 $
Subject: Bug#249315: fixed in openafs 1.4rc1-1
Message-Id: <[EMAIL PROTECTED]>
Sender: Archive Administrator <[EMAIL PROTECTED]>
Date: Fri, 26 Aug 2005 10:32:07 -0700
Delivered-To: [EMAIL PROTECTED]
X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2005_01_02
(1.212-2003-09-23-exp) on spohr.debian.org
X-Spam-Level:
X-Spam-Status: No, hits=-6.0 required=4.0 tests=BAYES_00,HAS_BUG_NUMBER
autolearn=no version=2.60-bugs.debian.org_2005_01_02
X-CrossAssassin-Score: 3
Source: openafs
Source-Version: 1.4rc1-1
We believe that the bug you reported is fixed in the latest version of
openafs, which is due to be installed in the Debian FTP archive:
libopenafs-dev_1.4rc1-1_i386.deb
to pool/main/o/openafs/libopenafs-dev_1.4rc1-1_i386.deb
libpam-openafs-kaserver_1.4rc1-1_i386.deb
to pool/main/o/openafs/libpam-openafs-kaserver_1.4rc1-1_i386.deb
openafs-client_1.4rc1-1_i386.deb
to pool/main/o/openafs/openafs-client_1.4rc1-1_i386.deb
openafs-dbserver_1.4rc1-1_i386.deb
to pool/main/o/openafs/openafs-dbserver_1.4rc1-1_i386.deb
openafs-fileserver_1.4rc1-1_i386.deb
to pool/main/o/openafs/openafs-fileserver_1.4rc1-1_i386.deb
openafs-kpasswd_1.4rc1-1_i386.deb
to pool/main/o/openafs/openafs-kpasswd_1.4rc1-1_i386.deb
openafs-modules-source_1.4rc1-1_all.deb
to pool/main/o/openafs/openafs-modules-source_1.4rc1-1_all.deb
openafs_1.4rc1-1.diff.gz
to pool/main/o/openafs/openafs_1.4rc1-1.diff.gz
openafs_1.4rc1-1.dsc
to pool/main/o/openafs/openafs_1.4rc1-1.dsc
openafs_1.4rc1.orig.tar.gz
to pool/main/o/openafs/openafs_1.4rc1.orig.tar.gz
A summary of the changes between this version and the previous one is
attached.
Thank you for reporting the bug, which will now be closed. If you
have further comments please address them to [EMAIL PROTECTED],
and the maintainer will reopen the bug report if appropriate.
Debian distribution maintenance software
pp.
Russ Allbery <[EMAIL PROTECTED]> (supplier of updated openafs package)
(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing [EMAIL PROTECTED])
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Format: 1.7
Date: Fri, 26 Aug 2005 01:30:21 -0700
Source: openafs
Binary: openafs-client libopenafs-dev openafs-dbserver openafs-modules-source
openafs-fileserver openafs-kpasswd libpam-openafs-kaserver
Architecture: source i386 all
Version: 1.4rc1-1
Distribution: unstable
Urgency: low
Maintainer: Sam Hartman <[EMAIL PROTECTED]>
Changed-By: Russ Allbery <[EMAIL PROTECTED]>
Description:
libopenafs-dev - The AFS distributed filesystem- development libraries
libpam-openafs-kaserver - The AFS distributed filesystem- kaserver PAM module
openafs-client - The AFS distributed filesystem- client support
openafs-dbserver - The AFS distributed filesystem- database server
openafs-fileserver - The AFS distributed filesystem- file server
openafs-kpasswd - The AFS distributed filesystem- old password changing
openafs-modules-source - The AFS distributed filesystem- Module Sources
Closes: 120097 249315 322638 323582 324694
Changes:
openafs (1.4rc1-1) unstable; urgency=low
.
* New upstream version.
* Install pam_afs.krb.so on hppa. I believe the new PAM module build
process fixes the issue that was being worked around before.
* Improve the documentation and scripts for setting up new servers or a
new cell. afs-rootvol can now be run from a client configured to use
dynroot. afs-newcell generates the server CellServDB directly to work
around oddities with bos addhost. There are many other, more minor
improvements as well. Thanks to Faheem Mitha and Sergio Gelato for
analysis and patches. (Closes: #322638)
* Update the configuration transcript to reflect the current packages,
messages, and instructions. Thanks, Faheem Mitha.
* Fill in the server CellServDB with information about the local cell,
if available in the client CellServDB, rather than always initializing
it to contain just the name of the cell.
* Detect AFS caches on non-ext2/ext3 file systems and abort AFS client
initialization. (Closes: #249315)
* Provide a way of setting sysname in afs.conf. (Closes: #324694)
* Don't create the unused /etc/openafs/AFSLog file. (Closes: #120097)
* Redo how library object files are found for the PAM module build to
avoid assuming C locale character set behavior. (Closes: #323582)
* Remove the openafs-client warning against dynroot for the first system
in a cell now that afs-rootvol can cope. Provide some basic
documentation of the CellAlias syntax in that message until we have a
real man page.
* Suggest openafs-doc.
* Update standards version to 3.6.2 for the kernel module packages (no
changes required).
* Translation updates.
- Czech, thanks Martin Sin.
- Vietnamese, thanks Clytie Siddall.
- French, thanks Christian Perrier.
Files:
6f73aebf36a64ab4cccf90267de52cf4 1108 net optional openafs_1.4rc1-1.dsc
8cac45d3fc3c0d63d68092fd9638ee69 4355767 net optional
openafs_1.4rc1.orig.tar.gz
1bc1c0820a608446f009053d5ac86189 115168 net optional openafs_1.4rc1-1.diff.gz
077ea2b10ad48b29261eeca1ce1ef45a 4469272 net extra
openafs-modules-source_1.4rc1-1_all.deb
948acee67bd91462ef4b2cbf13b06b08 1491022 net optional
openafs-client_1.4rc1-1_i386.deb
8eb36ac3b740ea0025369ec91eec6263 193674 net extra
openafs-kpasswd_1.4rc1-1_i386.deb
f3c14db41c5f0d81110fd4b9d29130ac 774976 net optional
openafs-fileserver_1.4rc1-1_i386.deb
4c457b6371e9813752f004bec3b01108 449610 net optional
openafs-dbserver_1.4rc1-1_i386.deb
d6a13c5639e29fc5e3fec8b9db3aa5bc 1128886 libdevel extra
libopenafs-dev_1.4rc1-1_i386.deb
d2488d4711afe347a71c13b666746901 396030 net extra
libpam-openafs-kaserver_1.4rc1-1_i386.deb
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
iQEVAwUBQw9Qb7f2jhx5fmQdAQKBZQgAt1z5p6so2JuXwZnAwScUqnZXfvENSJB2
e8E4LTKAIUNOwlKiyaSLhQPR17jxBV6fvVBI0RZoQxe8R4OG1QoVNAJ6rLE9pWfD
LnA2fSSEFJNh4tea0444fLx7wG9Z0b37JawZp1FCZoy4llzbAJqfToHvwWHot/0c
DWlpGaww0NIJXuU0LgUP+BCBCyaAE7Zd7oJ0hb141xvwg/GCsMAOZHRo5rCRoKkL
OTdcw0EcG0Efe1z+IP9MgytWuGUyn6zDHhdkLSubp+MXp2oa2AsaJUFWIXhQRmXB
otnDFFWjgJ5IUI102ubWyuJWdDQTO7F4SeYyY4j8sKdPdjG1mqThrg==
=igG2
-----END PGP SIGNATURE-----
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
