public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
* XFS corruption with rsync
@ 2012-09-19 13:57 Emmanuel Florac
  2012-09-19 17:39 ` Eric Sandeen
  0 siblings, 1 reply; 3+ messages in thread
From: Emmanuel Florac @ 2012-09-19 13:57 UTC (permalink / raw)
  To: xfs


I'"ve seen this error showing up on the list previously: while
rsync'ing a large volume (66 TB copied to a 114 TB filesystem), I've got
this error showing up:


Pid: 2069, comm: rsync Not tainted 3.2.28-storiq64-opteron #1
Call Trace:
 [<ffffffff811e037f>] ? xfs_alloc_fixup_trees+0x22f/0x370
 [<ffffffff811f6ee3>] ? xfs_btree_get_rec+0x53/0x90
 [<ffffffff811e32ee>] ? xfs_alloc_ag_vextent_near+0x83e/0xbe0
 [<ffffffff811e3b4b>] ? xfs_alloc_fix_freelist+0x3cb/0x460
 [<ffffffff811e3755>] ? xfs_alloc_ag_vextent+0xc5/0xf0
 [<ffffffff811e40ba>] ? xfs_alloc_vextent+0x3aa/0x5f0
 [<ffffffff81222d4a>] ? xfs_trans_read_buf+0x22a/0x420
 [<ffffffff8120a857>] ? xfs_ialloc_ag_alloc+0x3c7/0x4d0
 [<ffffffff8120aa2b>] ? xfs_dialloc+0xcb/0x870
 [<ffffffff81077ffd>] ? up+0xd/0x40
 [<ffffffff8120d705>] ? xfs_ialloc+0x75/0x760
 [<ffffffff811df986>] ? kmem_zone_alloc+0x66/0xe0
 [<ffffffff811dbc23>] ? xfs_dir_ialloc+0x93/0x330
 [<ffffffff812198bd>] ? xfs_trans_reserve+0x9d/0x210
 [<ffffffff811df705>] ? xfs_create+0x455/0x670
 [<ffffffff811d4286>] ? xfs_vn_mknod+0xa6/0x1b0
 [<ffffffff8110f129>] ? vfs_create+0x89/0xc0
 [<ffffffff811107a0>] ? do_last+0x550/0x750
 [<ffffffff811112bb>] ? path_openat+0xdb/0x3d0
 [<ffffffff81103617>] ? do_sync_write+0xc7/0x100
 [<ffffffff810e43e6>] ? handle_mm_fault+0x226/0x270
 [<ffffffff811116dd>] ? do_filp_open+0x4d/0xc0
 [<ffffffff8111d923>] ? alloc_fd+0x43/0x130
 [<ffffffff81102fb5>] ? do_sys_open+0x105/0x1e0
 [<ffffffff81424652>] ? system_call_fastpath+0x16/0x1b


Rsync logs says that "structure needs cleaning". After running
xfs_repair, there's a few files making reference to a free block. 

This machine has had serious filesystem failure previously, but the
RAID array and filesystem have been rebuild from the ground up, so it
should be OK; however I'd like to be really sure there's no hidden
hardware glitch waiting to bite us again...

-- 
------------------------------------------------------------------------
Emmanuel Florac     |   Direction technique
                    |   Intellique
                    |	<eflorac@intellique.com>
                    |   +33 1 78 94 84 02
------------------------------------------------------------------------

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 3+ messages in thread
* Re: XFS corruption with rsync
  2012-09-19 13:57 XFS corruption with rsync Emmanuel Florac
@ 2012-09-19 17:39 ` Eric Sandeen
  2012-09-19 19:58   ` Emmanuel Florac
  0 siblings, 1 reply; 3+ messages in thread
From: Eric Sandeen @ 2012-09-19 17:39 UTC (permalink / raw)
  To: Emmanuel Florac; +Cc: xfs

On 9/19/12 8:57 AM, Emmanuel Florac wrote:
> 
> I'"ve seen this error showing up on the list previously: while
> rsync'ing a large volume (66 TB copied to a 114 TB filesystem), I've got
> this error showing up:

I think you trimmed some relevant lines before these, what actual
error did it hit, there's probably a file & line number?

> Pid: 2069, comm: rsync Not tainted 3.2.28-storiq64-opteron #1
> Call Trace:
>  [<ffffffff811e037f>] ? xfs_alloc_fixup_trees+0x22f/0x370
>  [<ffffffff811f6ee3>] ? xfs_btree_get_rec+0x53/0x90
>  [<ffffffff811e32ee>] ? xfs_alloc_ag_vextent_near+0x83e/0xbe0
>  [<ffffffff811e3b4b>] ? xfs_alloc_fix_freelist+0x3cb/0x460
>  [<ffffffff811e3755>] ? xfs_alloc_ag_vextent+0xc5/0xf0
>  [<ffffffff811e40ba>] ? xfs_alloc_vextent+0x3aa/0x5f0
>  [<ffffffff81222d4a>] ? xfs_trans_read_buf+0x22a/0x420
>  [<ffffffff8120a857>] ? xfs_ialloc_ag_alloc+0x3c7/0x4d0
>  [<ffffffff8120aa2b>] ? xfs_dialloc+0xcb/0x870
>  [<ffffffff81077ffd>] ? up+0xd/0x40
>  [<ffffffff8120d705>] ? xfs_ialloc+0x75/0x760
>  [<ffffffff811df986>] ? kmem_zone_alloc+0x66/0xe0
>  [<ffffffff811dbc23>] ? xfs_dir_ialloc+0x93/0x330
>  [<ffffffff812198bd>] ? xfs_trans_reserve+0x9d/0x210
>  [<ffffffff811df705>] ? xfs_create+0x455/0x670
>  [<ffffffff811d4286>] ? xfs_vn_mknod+0xa6/0x1b0
>  [<ffffffff8110f129>] ? vfs_create+0x89/0xc0
>  [<ffffffff811107a0>] ? do_last+0x550/0x750
>  [<ffffffff811112bb>] ? path_openat+0xdb/0x3d0
>  [<ffffffff81103617>] ? do_sync_write+0xc7/0x100
>  [<ffffffff810e43e6>] ? handle_mm_fault+0x226/0x270
>  [<ffffffff811116dd>] ? do_filp_open+0x4d/0xc0
>  [<ffffffff8111d923>] ? alloc_fd+0x43/0x130
>  [<ffffffff81102fb5>] ? do_sys_open+0x105/0x1e0
>  [<ffffffff81424652>] ? system_call_fastpath+0x16/0x1b
> 
> 
> Rsync logs says that "structure needs cleaning". After running
> xfs_repair, there's a few files making reference to a free block. 

Including the xfs_repair output would be helpful too.

> This machine has had serious filesystem failure previously, but the
> RAID array and filesystem have been rebuild from the ground up, so it
> should be OK; however I'd like to be really sure there's no hidden
> hardware glitch waiting to bite us again...

-Eric


_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 3+ messages in thread
* Re: XFS corruption with rsync
  2012-09-19 17:39 ` Eric Sandeen
@ 2012-09-19 19:58   ` Emmanuel Florac
  0 siblings, 0 replies; 3+ messages in thread
From: Emmanuel Florac @ 2012-09-19 19:58 UTC (permalink / raw)
  To: Eric Sandeen; +Cc: xfs

Le Wed, 19 Sep 2012 12:39:53 -0500 vous écriviez:

> On 9/19/12 8:57 AM, Emmanuel Florac wrote:
> > 
> > I'"ve seen this error showing up on the list previously: while
> > rsync'ing a large volume (66 TB copied to a 114 TB filesystem),
> > I've got this error showing up:
> 
> I think you trimmed some relevant lines before these, what actual
> error did it hit, there's probably a file & line number?

Nothing else in dmesg but this error repeated many times. I'll check
through the /var/log/messages if there is more information.

> > Rsync logs says that "structure needs cleaning". After running
> > xfs_repair, there's a few files making reference to a free block. 
> 
> Including the xfs_repair output would be helpful too.

Unfortunately I forgot that all of xfs_repair output is to STDERR, so I
lost it... It said a lot of lines like "data fork in inode XXX claims
free block YYY". However no data was dropped to lost+found, and
apparently all is OK.

-- 
------------------------------------------------------------------------
Emmanuel Florac     |   Direction technique
                    |   Intellique
                    |	<eflorac@intellique.com>
                    |   +33 1 78 94 84 02
------------------------------------------------------------------------

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 3+ messages in thread
end of thread, other threads:[~2012-09-19 19:57 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-09-19 13:57 XFS corruption with rsync Emmanuel Florac
2012-09-19 17:39 ` Eric Sandeen
2012-09-19 19:58   ` Emmanuel Florac

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox