Re: xfs_iunlink_remove: xfs_inotobp() returned error 22 -- debugging

[符永涛 <yongtaofu@xxxxxxxxx>]

Hi Brian and Eric,

Here's the  meta_dump file of one server xfs repair log. And again this happens exactly when one of the glusterfs finished rebalance.
https://docs.google.com/file/d/0B7n2C4T5tfNCdDFwN24zdkVmdHM/edit?usp=sharing

https://docs.google.com/file/d/0B7n2C4T5tfNCOGdpOGhIaVFRV28/edit?usp=sharing

The server which finishes rebalance and trigger xfs shutdown has the following log on glusterfs:

[2013-04-18 22:42:52.063196] E [dht-rebalance.c:1194:gf_defrag_migrate_data] 0-test2-dht: migrate-data failed for /fytest/8/58037
[2013-04-18 22:42:52.067867] I [dht-rebalance.c:639:dht_migrate_file] 0-test2-dht: /fytest/8/58040: attempting to move from test2-replicate-1 to test2-replicate-3
[2013-04-18 22:42:52.070530] W [dht-rebalance.c:353:__dht_check_free_space] 0-test2-dht: data movement attempted from node (test2-replicate-1) with higher disk space to a node (test2-replicate-3) with lesser disk space (/fytest/8/58040)
[2013-04-18 22:42:52.070613] E [dht-rebalance.c:1194:gf_defrag_migrate_data] 0-test2-dht: migrate-data failed for /fytest/8/58040
[2013-04-18 22:43:11.679797] I [dht-common.c:2337:dht_setxattr] 0-test2-dht: fixing the layout of /fytest/2
[2013-04-18 22:43:11.682162] I [dht-rebalance.c:1055:gf_defrag_migrate_data] 0-test2-dht: migrate data called on /fytest/2
[2013-04-18 22:43:14.246209] I [dht-rebalance.c:1611:gf_defrag_status_get] 0-glusterfs: Rebalance is completed
[2013-04-18 22:43:14.246278] I [dht-rebalance.c:1614:gf_defrag_status_get] 0-glusterfs: Files migrated: 7203, size: 745897761616, lookups: 79881, failures: 9002
[2013-04-18 22:43:14.247021] W [glusterfsd.c:831:cleanup_and_exit] (-->/lib64/libc.so.6(clone+0x6d) [0x3ec22e767d] (-->/lib64/libpthread.so.0() [0x3ec2607851] (-->/usr/sbin/glusterfs(glusterfs_sigwaiter+0xdd) [0x405c9d]))) 0-: received signum (15), shutting down
[2013-04-18 22:43:14.247202] E [rpcsvc.c:1155:rpcsvc_program_unregister_portmap] 0-rpc-service: Could not unregister with portmap

The shutdown happens at 2013-04-18 22:43:14.

Thank you.

 

2013/4/18 符永涛 <yongtaofu@xxxxxxxxx>

Hi Brian and Eric,

The shutdown is not easy to produce but finally right now 2 of our servers in our test cluster xfs was shutdown.

the trace output as following
https://docs.google.com/file/d/0B7n2C4T5tfNCLXRYUWJ0b19JcWc/edit?usp=sharing

Sorry but the systemtap is interrupt and I didn't noticed that so I didn't get systemtap logs.

/var/log/message is same as before
Apr 18 22:43:14 10 kernel: XFS (sdb): xfs_iunlink_remove: xfs_inotobp() returned error 22.
Apr 18 22:43:14 10 kernel: XFS (sdb): xfs_inactive: xfs_ifree returned error 22
Apr 18 22:43:14 10 kernel: XFS (sdb): xfs_do_force_shutdown(0x1) called from line 1184 of file fs/xfs/xfs_vnodeops.c.  Return address = 0xffffffffa02d44aa
Apr 18 22:43:14 10 kernel: XFS (sdb): I/O Error Detected. Shutting down filesystem
Apr 18 22:43:14 10 kernel: XFS (sdb): Please umount the filesystem and rectify the problem(s)
Apr 18 22:43:20 10 kernel: XFS (sdb): xfs_log_force: error 5 returned.

The metadump file is large I'll share it to you soon.

2013/4/18 Brian Foster <bfoster@xxxxxxxxxx>

On 04/18/2013 04:25 AM, 符永涛 wrote:
> Hi Brian and Eric,
> Can I change as following to bypass it?

This is probably not a wise thing to do. The problem we're seeing here
is indicative of a potentially larger problem than this particular error
path. An inode is being unlinked and inactivated, but we aren't finding
on the list where we expect it to be. Killing the error return doesn't
eliminate the larger problem.

So while changes could end up being made in this area as part of a fix,
I would not suggest making any changes beyond those designed to help
debug until we have a better idea of root cause.

Brian

> --- a/xfs_inode.c
> +++ b/xfs_inode.c
> @@ -1764,7 +1764,7 @@ xfs_iunlink_remove(
>                  */
>                 next_agino = be32_to_cpu(agi->agi_unlinked[bucket_index]);
>                 last_ibp = NULL;
> -               while (next_agino != agino) {
> +               while (next_agino != agino && next_agino != NULLAGINO) {
>                         /*
>                          * If the last inode wasn't the one pointing to
>                          * us, then release its buffer since we're not
> @@ -1786,6 +1786,14 @@ xfs_iunlink_remove(
>                         ASSERT(next_agino != NULLAGINO);
>                         ASSERT(next_agino != 0);
>                 }
> +               if (next_agino == NULLAGINO) {
> +                       /*
> +                        *After search the list for the inode being free
> +                        *we still can't find it.
> +                        */
> +                       xfs_err(mp, "%s ino %lld not found in unlinked
> list.",
> +                                    __func__, (unsigned long
> long)ip->i_ino);
> +               }
>                 /*
>                  * Now last_ibp points to the buffer previous to us on
>                  * the unlinked list.  Pull us from the list.
> @@ -1810,16 +1818,20 @@ xfs_iunlink_remove(
>                 } else {
>                         xfs_trans_brelse(tp, ibp);
>                 }
> -               /*
> -                * Point the previous inode on the list to the next inode.
> -                */
> -               last_dip->di_next_unlinked = cpu_to_be32(next_agino);
> -               ASSERT(next_agino != 0);
> -               offset = last_offset + offsetof(xfs_dinode_t,
> di_next_unlinked);
> -               xfs_trans_inode_buf(tp, last_ibp);
> -               xfs_trans_log_buf(tp, last_ibp, offset,
> -                                 (offset + sizeof(xfs_agino_t) - 1));
> -               xfs_inobp_check(mp, last_ibp);
> +               if (next_agino != NULLAGINO) {
> +                       /*
> +                       * If only find the inode being free then we modify
> +                       * the unlinked list.
> +                       * Point the previous inode on the list to the
> next inode.
> +                       */
> +                       last_dip->di_next_unlinked =
> cpu_to_be32(next_agino);
> +                       ASSERT(next_agino != 0);
> +                       offset = last_offset + offsetof(xfs_dinode_t,
> di_next_unlinked);
> +                       xfs_trans_inode_buf(tp, last_ibp);
> +                       xfs_trans_log_buf(tp, last_ibp, offset,
> +                                         (offset + sizeof(xfs_agino_t)
> - 1));
> +                       xfs_inobp_check(mp, last_ibp);
> +               }
>         }
>         return 0;
>  }
>
> Thank you.
>
>

> 2013/4/18 符永涛 <yongtaofu@xxxxxxxxx <mailto:yongtaofu@xxxxxxxxx>>

>
>     Hi Brain and Eric,
>     If the problem is the agno can't be found in the unlinked list. Can
>     we just bypass it instead of passing ino=0xffffffff to xfs_inotobp?
>     Thank you.
>
>

>     2013/4/18 符永涛 <yongtaofu@xxxxxxxxx <mailto:yongtaofu@xxxxxxxxx>>

>
>         Hi Eric,
>         The shutdown issue is still not reproduced yet. But I get the
>         following error today during test.
>
>         Apr 18 07:42:51 10 kernel: Call Trace:
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02d91ef>] ?
>         xfs_buf_cond_lock+0x2f/0xc0 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffff814fe6a5>]
>         schedule_timeout+0x215/0x2e0
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02d5f07>] ?
>         kmem_zone_alloc+0x77/0xf0 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffff814ff5c2>] __down+0x72/0xb0
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02da652>] ?
>         _xfs_buf_find+0x102/0x280 [xfs]
>         Apr 18 07:42:51 10 kernel: "echo 0 >
>         /proc/sys/kernel/hung_task_timeout_secs" disables this message.
>         Apr 18 07:42:51 10 kernel: glusterfsd    D ffffffff8160b3c0
>         0 14522      1 0x00000083
>         Apr 18 07:42:51 10 kernel: ffff882015a63a28 0000000000000082
>         0000000000000000 0000000000000000
>         Apr 18 07:42:51 10 kernel: ffff882015a639b8 ffffffffa02d91ef
>         ffff882015a639d8 0000000000000246
>         Apr 18 07:42:51 10 kernel: ffff880e70491af8 ffff882015a63fd8
>         000000000000fb88 ffff880e70491af8
>         Apr 18 07:42:51 10 kernel: Call Trace:
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02d91ef>] ?
>         xfs_buf_cond_lock+0x2f/0xc0 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffff814fe6a5>]
>         schedule_timeout+0x215/0x2e0
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02d5f07>] ?
>         kmem_zone_alloc+0x77/0xf0 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffff814ff5c2>] __down+0x72/0xb0
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02da652>] ?
>         _xfs_buf_find+0x102/0x280 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffff81097ef1>] down+0x41/0x50
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02da493>]
>         xfs_buf_lock+0x53/0x110 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02da652>]
>         _xfs_buf_find+0x102/0x280 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02da83b>]
>         xfs_buf_get+0x6b/0x1a0 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02daeac>]
>         xfs_buf_read+0x2c/0x100 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02d0af8>]
>         xfs_trans_read_buf+0x1f8/0x400 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02b3444>]
>         xfs_read_agi+0x74/0x100 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02b967b>]
>         xfs_iunlink+0x5b/0x180 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffff810724c7>] ?
>         current_fs_time+0x27/0x30
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02d12a7>] ?
>         xfs_trans_ichgtime+0x27/0xa0 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02d15fb>]
>         xfs_droplink+0x5b/0x70 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02d2f9e>]
>         xfs_remove+0x27e/0x3a0 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffff81186fd3>] ?
>         generic_permission+0x23/0xb0
>         Apr 18 07:42:51 10 kernel: [<ffffffffa02e0968>]
>         xfs_vn_unlink+0x48/0x90 [xfs]
>         Apr 18 07:42:51 10 kernel: [<ffffffff81188c0f>] vfs_unlink+0x9f/0xe0
>         Apr 18 07:42:51 10 kernel: [<ffffffff8118795a>] ?
>         lookup_hash+0x3a/0x50
>         Apr 18 07:42:51 10 kernel: [<ffffffff8118b143>]
>         do_unlinkat+0x183/0x1c0
>         Apr 18 07:42:51 10 kernel: [<ffffffff81017938>] ?
>         syscall_trace_enter+0x1d8/0x1e0
>         Apr 18 07:42:51 10 kernel: [<ffffffff8118b196>] sys_unlink+0x16/0x20
>         Apr 18 07:42:51 10 kernel: [<ffffffff8100b308>] tracesys+0xd9/0xde
>
>         Thank you.
>
>
>         2013/4/17 Eric Sandeen <sandeen@xxxxxxxxxxx

>         <mailto:sandeen@xxxxxxxxxxx>>

>
>             On Apr 16, 2013, at 8:48 PM, 符永涛 <yongtaofu@xxxxxxxxx

>             <mailto:yongtaofu@xxxxxxxxx>> wrote:
>
>>             Hi Brain,
>>             Can I change as following?
>
>             ASSERTS are no-ops in a non-debug kernel, so this won't
>             change any behavior.  I hope we'll know more if we get new
>             traces from your patched kernel....
>
>             Eric
>
>>             --- a/xfs_inode.c
>>             +++ b/xfs_inode.c
>>             @@ -1773,6 +1773,8 @@ xfs_iunlink_remove(
>>                                     if (last_ibp != NULL) {
>>                                             xfs_trans_brelse(tp,
>>             last_ibp);
>>                                     }
>>             +                        ASSERT(next_agino != NULLAGINO);
>>             +                        ASSERT(next_agino != 0);
>>                                     next_ino = XFS_AGINO_TO_INO(mp,
>>             agno, next_agino);
>>                                     error = xfs_inotobp(mp, tp,
>>             next_ino, &last_dip,
>>                                                         &last_ibp,
>>             &last_offset, 0);
>>             @@ -1783,8 +1785,6 @@ xfs_iunlink_remove(
>>                                             return error;
>>                                     }
>>                                     next_agino =
>>             be32_to_cpu(last_dip->di_next_unlinked);
>>             -                       ASSERT(next_agino != NULLAGINO);
>>             -                       ASSERT(next_agino != 0);
>>                             }
>>                             /*
>>                              * Now last_ibp points to the buffer
>>             previous to us on
>>
>>             Thank you.
>>
>>
>>             2013/4/17 符永涛 <yongtaofu@xxxxxxxxx

>>             <mailto:yongtaofu@xxxxxxxxx>>

>>
>>                 Hi Brain,
>>                 If it is because NULLAGINO is passed in  to xfs_inotobp().
>>                 Can I move the following two lines before xfs_inotobp?
>>
>>                 For example:
>>
>>                 1767                 while (next_agino != agino) {
>>                 1768                         /*
>>                 1769                          * If the last inode
>>                 wasn't the one pointing to
>>                 1770                          * us, then release its
>>                 buffer since we're not
>>                 1771                          * going to do anything
>>                 with it.
>>                 1772                          */
>>                 1773                         if (last_ibp != NULL) {
>>                 1774
>>                 xfs_trans_brelse(tp, last_ibp);
>>                 1775                         }
>>                 1776                         next_ino =
>>                 XFS_AGINO_TO_INO(mp, agno, next_agino);
>>                 +                               ASSERT(next_agino !=
>>                 NULLAGINO);
>>                 +                               ASSERT(next_agino != 0);
>>                 1777                         error = xfs_inotobp(mp,
>>                 tp, next_ino, &last_dip,
>>                 1778
>>                 &last_ibp, &last_offset, 0);
>>                 1779                         if (error) {
>>                 1780                                 xfs_warn(mp,
>>                 1781                                         "%s:
>>                 xfs_inotobp() returned error %d.",
>>                 1782                                         __func__,
>>                 error);
>>                 1783                                 return error;
>>                 1784                         }
>>                 1785                         next_agino =
>>                 be32_to_cpu(last_dip->di_next_unlinked);
>>                 -                               //ASSERT(next_agino !=
>>                 NULLAGINO);
>>                 -                               //ASSERT(next_agino != 0);
>>                 1788                 }
>>                 I don't understand xfs well and correct me if I'm
>>                 totally wrong.
>>                 Thank you very much.
>>
>>
>>                 2013/4/17 符永涛 <yongtaofu@xxxxxxxxx

>>                 <mailto:yongtaofu@xxxxxxxxx>>

>>
>>                     Hi Brain,
>>                     I want to ask a question, according to the
>>                     shutdown trace. The ino in  xfs_iunlink_remove
>>                     is 0x113, why xfs_imap got ino=0xffffffff ?
>>
>>                     --- xfs_imap --
>>                     module("xfs").function("xfs_imap@fs/xfs/xfs_ialloc.c:1257").return
>>                     -- return=0x16
>>                     vars: mp=0xffff882017a50800 tp=0xffff881c81797c70
>>                     ino=0xffffffff
>>
>>                     --- xfs_iunlink_remove --
>>                     module("xfs").function("xfs_iunlink_remove@fs/xfs/xfs_inode.c:1680").return
>>                     -- return=0x16
>>                     vars: tp=0xffff881c81797c70 ip=0xffff881003c13c00
>>                     next_ino=? mp=? agi=? dip=?
>>                     agibp=0xffff880109b47e20 ibp=? agno=? agino=?
>>                     next_agino=? last_ibp=?
>>                     last_dip=0xffff882000000000 bucket_index=?
>>                     offset=? last_offset=0xffffffffffff8810 error=?
>>                     __func__=[...]
>>                     ip: i_ino = 0x113, i_flags = 0x0
>>
>>                     Thank you.
>>
>>
>>
>>                     2013/4/17 符永涛 <yongtaofu@xxxxxxxxx

>>                     <mailto:yongtaofu@xxxxxxxxx>>

>>
>>                         Hi Brain,
>>                         Thank you for your update, and I have applied
>>                         your last kernel patch. However it is not easy
>>                         to reproduce especially in out test
>>                         environment. Till now is not happens again.
>>                         I'll update the kernel patch now. BTW is there
>>                         any findings in the logs of previous thread?
>>                         http://oss.sgi.com/archives/xfs/2013-04/msg00327.html
>>                         I guess it tend to happen during glusterfs
>>                         rebalance because glusterfs moves a lot of
>>                         file from one server to another and then
>>                         unlink it.
>>
>>                         Thank you.
>>
>>
>>                         2013/4/17 Brian Foster <bfoster@xxxxxxxxxx

>>                         <mailto:bfoster@xxxxxxxxxx>>

>>
>>                             On 04/16/2013 12:24 PM, Dave Chinner wrote:
>>                             > On Mon, Apr 15, 2013 at 07:14:39PM
>>                             -0400, Brian Foster wrote:
>>                             >> Hi,
>>                             >>
>>                             >> Thanks for the data in the previous thread:
>>                             >>
>>                             >>
>>                             http://oss.sgi.com/archives/xfs/2013-04/msg00327.html
>>                             >>
>>                             ...
>>                             >>
>>                             >>      echo 1 >
>>                             /sys/kernel/debug/tracing/events/xfs/xfs_iunlink/enable
>>                             >>      echo 1 >
>>                             /sys/kernel/debug/tracing/events/xfs/xfs_iunlink_remove/enable
>>                             >>      ... reproduce ...
>>                             >>      cat
>>                             /sys/kernel/debug/tracing/trace > trace.output
>>                             >
>>                             > It's better to use trace-cmd for this.
>>                             it will result in less
>>                             > dropped events. i.e.:
>>                             >
>>                             >       $ trace-cmd record -e xfs_iunlink\*
>>                             >       ... reproduce ...
>>                             >       ^C
>>                             >       $ trace-cmd report > trace.output
>>                             >
>>                             >> --- a/fs/xfs/linux-2.6/xfs_trace.h
>>                             >> +++ b/fs/xfs/linux-2.6/xfs_trace.h
>>                             >> @@ -581,6 +581,8 @@
>>                             DEFINE_INODE_EVENT(xfs_file_fsync);
>>                             ...
>>                             >
>>                             > I would suggest that the the tracing
>>                             shoul dbe at entry of the
>>                             > function, otherwise we won't get a
>>                             tracepoint for the operation that
>>                             > triggers the shutdown. (That's the
>>                             reason most tracepoints in XFS
>>                             > are at function entry...)
>>                             >
>>
>>                             Good points, thanks Dave. A v2 that pulls
>>                             up the tracepoints towards
>>                             function entry is appended.
>>
>>                             Brian
>>
>>                             From
>>                             280943e78ebe0b97a774cba51e7815c42f044b55
>>                             Mon Sep 17 00:00:00 2001
>>                             From: Brian Foster <bfoster@xxxxxxxxxx

>>                             <mailto:bfoster@xxxxxxxxxx>>

>>                             Date: Mon, 15 Apr 2013 18:16:24 -0400
>>                             Subject: [PATCH v2] xfs: add tracepoints
>>                             for xfs_iunlink and
>>                             xfs_iunlink_remove
>>
>>                             ---
>>                              fs/xfs/linux-2.6/xfs_trace.h |    2 ++
>>                              fs/xfs/xfs_inode.c           |    4 ++++
>>                              2 files changed, 6 insertions(+), 0
>>                             deletions(-)
>>
>>                             diff --git a/fs/xfs/linux-2.6/xfs_trace.h
>>                             b/fs/xfs/linux-2.6/xfs_trace.h
>>                             index adc6ec4..338a0f9 100644
>>                             --- a/fs/xfs/linux-2.6/xfs_trace.h
>>                             +++ b/fs/xfs/linux-2.6/xfs_trace.h
>>                             @@ -583,6 +583,8 @@
>>                             DEFINE_INODE_EVENT(xfs_file_fsync);
>>                              DEFINE_INODE_EVENT(xfs_destroy_inode);
>>                              DEFINE_INODE_EVENT(xfs_dirty_inode);
>>                              DEFINE_INODE_EVENT(xfs_clear_inode);
>>                             +DEFINE_INODE_EVENT(xfs_iunlink);
>>                             +DEFINE_INODE_EVENT(xfs_iunlink_remove);
>>
>>                              DEFINE_INODE_EVENT(xfs_dquot_dqalloc);
>>                              DEFINE_INODE_EVENT(xfs_dquot_dqdetach);
>>                             diff --git a/fs/xfs/xfs_inode.c
>>                             b/fs/xfs/xfs_inode.c
>>                             index 19900f0..d705c77 100644
>>                             --- a/fs/xfs/xfs_inode.c
>>                             +++ b/fs/xfs/xfs_inode.c
>>                             @@ -1615,6 +1615,8 @@ xfs_iunlink(
>>
>>                                     mp = tp->t_mountp;
>>
>>                             +       trace_xfs_iunlink(ip);
>>                             +
>>                                     /*
>>                                      * Get the agi buffer first.  It
>>                             ensures lock ordering
>>                                      * on the list.
>>                             @@ -1694,6 +1696,8 @@ xfs_iunlink_remove(
>>                                     mp = tp->t_mountp;
>>                                     agno = XFS_INO_TO_AGNO(mp, ip->i_ino);
>>
>>                             +       trace_xfs_iunlink_remove(ip);
>>                             +
>>                                     /*
>>                                      * Get the agi buffer first.  It
>>                             ensures lock ordering
>>                                      * on the list.
>>                             --
>>                             1.7.7.6
>>
>>
>>
>>
>>                         --
>>                         符永涛
>>
>>
>>
>>
>>                     --
>>                     符永涛
>>
>>
>>
>>
>>                 --
>>                 符永涛
>>
>>
>>
>>
>>             --
>>             符永涛
>>             _______________________________________________
>>             xfs mailing list

>>             xfs@xxxxxxxxxxx <mailto:xfs@xxxxxxxxxxx>

>>             http://oss.sgi.com/mailman/listinfo/xfs
>
>
>
>
>         --
>         符永涛
>
>
>
>
>     --
>     符永涛
>
>
>
>
> --
> 符永涛

--
符永涛

--
符永涛