Re: XFS internal error xfs_trans_cancel at line 1163 of file fs/xfs/xfs_trans.c

Re: XFS internal error xfs_trans_cancel at line 1163 of file fs/xfs/xfs_trans.c

[Miquel van Smoorenburg]

On Thu, 2008-06-12 at 16:33 +0200, Oliver Pinter wrote:
> add CC's
> 
> On 6/12/08, lists@ku-gbr.de <lists@ku-gbr.de> wrote:
> > Hi!
> >
> > Today morning my server at home bailed out two times (reboot between):

> > Jun 12 07:23:40 zappa
> > Jun 12 07:23:40 zappa xfs_force_shutdown(sda7,0x8) called from line
> > 1164 of file fs/xfs/xfs_trans.c.  Return address = 0xffffffff802f4d8a
> > Jun 12 07:23:40 zappa Filesystem "sda7": Corruption of in-memory data
> > detected.  Shutting down filesystem: sda7
> > Jun 12 07:23:40 zappa Please umount the filesystem, and rectify the
> > problem(s)

Hmm, interesting. I'm seeing the same thing on one of my servers since I
upgraded from 2.6.ancient (14 or so) to 2.6.25, while XFS otherwise has
been very stable for me over the years:

Linux transit5.news.xs4all.nl 2.6.25.6 #1 SMP Wed Jun 11 10:59:10 CEST
2008 x86_64 GNU/Linux

Filesystem "sda4": XFS internal error xfs_trans_cancel at line 1163 of
file fs/xfs/xfs_trans.c.  Caller 0xffffffff880f1315
Pid: 3402, comm: diablo Not tainted 2.6.25.6 #1

Call Trace:
 [<ffffffff880f1315>] :xfs:xfs_create+0x1e5/0x520
 [<ffffffff880ea4a6>] :xfs:xfs_trans_cancel+0x126/0x150
 [<ffffffff880f1315>] :xfs:xfs_create+0x1e5/0x520
 [<ffffffff880fcb79>] :xfs:xfs_vn_mknod+0x1d9/0x320
 [<ffffffff8028811c>] vfs_create+0xac/0xf0
 [<ffffffff8028b6bd>] open_namei+0x61d/0x6c0
 [<ffffffff802448c0>] autoremove_wake_function+0x0/0x30
 [<ffffffff8027da0c>] do_filp_open+0x1c/0x50
 [<ffffffff8027d6e9>] get_unused_fd_flags+0x79/0x120
 [<ffffffff8027da9a>] do_sys_open+0x5a/0xf0
 [<ffffffff8020b2bb>] system_call_after_swapgs+0x7b/0x80

xfs_force_shutdown(sda4,0x8) called from line 1164 of file
fs/xfs/xfs_trans.c.  Return address = 0xffffffff880ea4bf
Filesystem "sda4": Corruption of in-memory data detected.  Shutting down
filesystem: sda4
Please umount the filesystem, and rectify the problem(s)

After a reboot, xfs_repair didn't find anything wrong with the fs. It
has happened 3 times over the last few days already.

FS is on a local SCSI raid (dpt_i2o), not SATA.

Mike.

Re: [xfs-masters] Re: XFS internal error xfs_trans_cancel at line 1163 of file fs/xfs/xfs_trans.c

[Konstantin Kletschke]

Now I got ksymoops output:

~/ > ksymoops -m /boot/System.map xfs_break1.txt
ksymoops 2.4.11 on x86_64 2.6.25-gentoo-r4.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.6.25-gentoo-r4/ (default)
     -m /boot/System.map (specified)

Error (regular_file): read_ksyms stat /proc/ksyms failed
ksymoops: No such file or directory
No modules in ksyms, skipping objects
No ksyms, skipping lsmod
Jun 12 07:23:40 zappa Pid: 2379, comm: procmail Not tainted 2.6.25-gentoo-r4 #3
Jun 12 07:23:40 zappa Call Trace:
Jun 12 07:23:40 zappa [<ffffffff802fa8f5>]
Jun 12 07:23:40 zappa [<ffffffff802f4d71>]
Jun 12 07:23:40 zappa [<ffffffff802fa8f5>]
Jun 12 07:23:40 zappa [<ffffffff803038bb>]
Jun 12 07:23:40 zappa [<ffffffff8025f1e1>]
Jun 12 07:23:40 zappa [<ffffffff80261899>]
Jun 12 07:23:40 zappa [<ffffffff8025ae00>]
Jun 12 07:23:40 zappa [<ffffffff80257010>]
Jun 12 07:23:40 zappa [<ffffffff80256d92>]
Jun 12 07:23:40 zappa [<ffffffff80257077>]
Jun 12 07:23:40 zappa [<ffffffff8020ad3b>]
Warning (Oops_read): Code line not seen, dumping what data is available


Trace; ffffffff802fa8f5 <xfs_create+44f/493>
Trace; ffffffff802f4d71 <xfs_trans_cancel+5c/f4>
Trace; ffffffff802fa8f5 <xfs_create+44f/493>
Trace; ffffffff803038bb <xfs_vn_mknod+15f/249>
Trace; ffffffff8025f1e1 <vfs_create+75/ba>
Trace; ffffffff80261899 <open_namei+19d/608>
Trace; ffffffff8025ae00 <vfs_lstat_fd+18/47>
Trace; ffffffff80257010 <do_filp_open+1c/3d>
Trace; ffffffff80256d92 <get_unused_fd_flags+6d/100>
Trace; ffffffff80257077 <do_sys_open+46/ca>
Trace; ffffffff8020ad3b <system_call_after_swapgs+7b/80>


1 warning and 1 error issued.  Results may not be reliable.





~/ > ksymoops -m /boot/System.map xfs_break2.txt
ksymoops 2.4.11 on x86_64 2.6.25-gentoo-r4.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.6.25-gentoo-r4/ (default)
     -m /boot/System.map (specified)

Error (regular_file): read_ksyms stat /proc/ksyms failed
ksymoops: No such file or directory
No modules in ksyms, skipping objects
No ksyms, skipping lsmod
Jun 12 08:15:58 zappa Pid: 2161, comm: procmail Not tainted 2.6.25-gentoo-r4 #3
Jun 12 08:15:58 zappa Call Trace:
Jun 12 08:15:58 zappa [<ffffffff802fa8f5>]
Jun 12 08:15:58 zappa [<ffffffff802f4d71>]
Jun 12 08:15:58 zappa [<ffffffff802fa8f5>]
Jun 12 08:15:58 zappa [<ffffffff803038bb>]
Jun 12 08:15:58 zappa [<ffffffff8025f1e1>]
Jun 12 08:15:58 zappa [<ffffffff80261899>]
Jun 12 08:15:58 zappa [<ffffffff8025ae00>]
Jun 12 08:15:58 zappa [<ffffffff80257010>]
Jun 12 08:15:58 zappa [<ffffffff80256d92>]
Jun 12 08:15:58 zappa [<ffffffff80257077>]
Jun 12 08:15:58 zappa [<ffffffff8020ad3b>]
Warning (Oops_read): Code line not seen, dumping what data is available


Trace; ffffffff802fa8f5 <xfs_create+44f/493>
Trace; ffffffff802f4d71 <xfs_trans_cancel+5c/f4>
Trace; ffffffff802fa8f5 <xfs_create+44f/493>
Trace; ffffffff803038bb <xfs_vn_mknod+15f/249>
Trace; ffffffff8025f1e1 <vfs_create+75/ba>
Trace; ffffffff80261899 <open_namei+19d/608>
Trace; ffffffff8025ae00 <vfs_lstat_fd+18/47>
Trace; ffffffff80257010 <do_filp_open+1c/3d>
Trace; ffffffff80256d92 <get_unused_fd_flags+6d/100>
Trace; ffffffff80257077 <do_sys_open+46/ca>
Trace; ffffffff8020ad3b <system_call_after_swapgs+7b/80>


1 warning and 1 error issued.  Results may not be reliable.


Regards, Konsti


PS.: The line xfs_check puttedt out was: "agi_freecount 1, counted 0 in ag17"
After xfs_repair'ing the message vanished, the result left no files in lost+found 
and seemed to have fixed this, the filesystem is up and running (for a
while :-) ) now.

-- 
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E  A080 1E69 3FDA EF62 FCEF

Re: XFS internal error xfs_trans_cancel at line 1163 of file fs/xfs/xfs_trans.c

[Miquel van Smoorenburg]

On Fri, 2008-06-13 at 13:28 +0200, Miquel van Smoorenburg wrote:
> On Fri, 2008-06-13 at 13:08 +1000, Dave Chinner wrote: 
> > On Fri, Jun 13, 2008 at 12:27:09AM +0200, Miquel van Smoorenburg wrote:
>
> > > Linux transit5.news.xs4all.nl 2.6.25.6 #1 SMP Wed Jun 11 10:59:10 CEST
> > > 2008 x86_64 GNU/Linux
> > > 
> > > Filesystem "sda4": XFS internal error xfs_trans_cancel at line 1163 of
> > > file fs/xfs/xfs_trans.c.  Caller 0xffffffff880f1315
> > > Pid: 3402, comm: diablo Not tainted 2.6.25.6 #1
> > 
> > This commit in 2.6.26 will probably fix it.
> > 
> > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=75de2a91c98a6f486f261c1367fe59f5583e15a3
> 
> "At ENOSPC, we can get a filesystem shutdown due to a cancelling a dirty
> transaction in xfs_mkdir or xfs_create."
> 
> But The filesystem is only used for 35%. It might have hit 100% somewhere in
> the recent past though (a few reboots ago).
> I've applied the patch to 2.6.25.6 just in case, I'll let it run over
> the weekend to see what happens.

Well, the box has been up for 3 days now. When this problem first
appeared it only stayed up for a day max, so I'm reasonably positive
it's fixed.

The patch applies cleanly to 2.6.25.6 - perhaps it should go into
-stable.

Thanks,

Mike.

Re: XFS internal error xfs_trans_cancel at line 1163 of file fs/xfs/xfs_trans.c

[Konstantin Kletschke]

Again. Before strace looks weird, this looks like the old one:

ksymoops 2.4.11 on x86_64 2.6.26-rc6.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.6.26-rc6/ (default)
     -m /usr/src/linux-2.6.26-rc6/System.map (specified)

Error (regular_file): read_ksyms stat /proc/ksyms failed
ksymoops: No such file or directory
No modules in ksyms, skipping objects
No ksyms, skipping lsmod
Pid: 2137, comm: fetchnews Not tainted 2.6.26-rc6 #1
Call Trace:
 [<ffffffff802ffa42>]
 [<ffffffff802fa35a>]
 [<ffffffff802ffa42>]
 [<ffffffff80308a29>]
 [<ffffffff80262799>]
 [<ffffffff8026500a>]
 [<ffffffff80223506>]
 [<ffffffff8023116c>]
 [<ffffffff8025a0fa>]
 [<ffffffff8020afab>]
Warning (Oops_read): Code line not seen, dumping what data is available


Trace; ffffffff802ffa42 <xfs_create+41d/462>
Trace; ffffffff802fa35a <xfs_trans_cancel+56/ee>
Trace; ffffffff802ffa42 <xfs_create+41d/462>
Trace; ffffffff80308a29 <xfs_vn_mknod+148/229>
Trace; ffffffff80262799 <vfs_create+75/ba>
Trace; ffffffff8026500a <do_filp_open+1dc/7c5>
Trace; ffffffff80223506 <ns_to_timeval+9/27>
Trace; ffffffff8023116c <__remove_hrtimer+6b/78>
Trace; ffffffff8025a0fa <do_sys_open+4a/f1>
Trace; ffffffff8020afab <system_call_after_swapgs+7b/80>


1 warning and 1 error issued.  Results may not be reliable.


-- 
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E  A080 1E69 3FDA EF62 FCEF

Re: XFS internal error xfs_trans_cancel at line 1163 of file fs/xfs/xfs_trans.c

[Dave Chinner]

On Fri, Jun 20, 2008 at 07:27:17AM +0200, Konstantin Kletschke wrote:
> Am 2008-06-13 13:08 +1000 schrieb Dave Chinner:
> 
> > This commit in 2.6.26 will probably fix it.
> > 
> > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=75de2a91c98a6f486f261c1367fe59f5583e15a3
> 
> If I am correct, this fix is included in 2.6.26-rc6:

Yes, so if you're seeing it again then there's a different problem.

Please provide a pointer to a xfs_metadump image of the filesystem
and the steps to reproduce the error from the image so we can
get to the bottom of it...

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com