ocfs2: do not panic if ocfs2_start_trans fails
diff mbox

Message ID 2015071014353683945641@h3c.com
State New
Headers show

Commit Message

Zhangguanghui July 10, 2015, 6:35 a.m. UTC
Hi?
When storage network is unstable, it may trigger the panic in ocfs2_start_trans.
 ocfs2_start_trans -> ocfs2_abort ->panic?whether there is a better way to solve this problem?
Just mark as a mistake instead of ocfs2_abort.
Finally, any feedback about this process (positive or negative) would be greatly appreciated.

Callback as follows?
Jul 8 15:34:16 cvk99 kernel: [72056.516720] (o2hb-5CED29F793,5052,1):o2hb_bio_end_io:381 ERROR: IO Error -5
Jul 8 15:34:16 cvk99 kernel: [72056.516723] (o2hb-5CED29F793,5052,1):o2hb_do_disk_heartbeat:993 ERROR: status = -5
Jul 8 15:34:16 cvk99 kernel: [72056.516730] (pool,7938,6):ocfs2_commit_trans:393 ERROR: status = -5
Jul 8 15:34:16 cvk99 kernel: [72056.516733] (pool,7938,6):ocfs2_write_begin_inline:1515 ERROR: status = -30
Jul 8 15:34:16 cvk99 kernel: [72056.516735] (pool,7938,6):ocfs2_try_to_write_inline_data:1594 ERROR: status = -30
Jul 8 15:34:16 cvk99 kernel: [72056.516738] (pool,7938,6):ocfs2_write_begin_nolock:1721 ERROR: status = -30
Jul 8 15:34:16 cvk99 kernel: [72056.516741] (pool,7938,6):ocfs2_write_begin:1937 ERROR: status = -30
Jul 8 15:34:16 cvk99 kernel: [72056.516772] (pool,8052,0):ocfs2_start_trans:368 ERROR: status = -30
Jul 8 15:34:16 cvk99 kernel: [72056.516776] OCFS2: abort (device sdb): ocfs2_start_trans: Detected aborted journal

________________________________
zhangguanghui

???? zhangguanghui 10102<mailto:zhang.guanghui@h3c.com>
????? 2015-07-06 10:29
???? Joseph Qi<mailto:joseph.qi@huawei.com>; Andrew Morton<mailto:akpm@linux-foundation.org>
??? Mark Fasheh<mailto:mfasheh@suse.com>; Joel Becker<mailto:jlbec@evilplan.org>; ocfs2-devel@oss.oracle.com<mailto:ocfs2-devel@oss.oracle.com>
??? Re: [PATCH] ocfs2: do not BUG if buffer not uptodate in __ocfs2_journal_access

This patch is currently verified and working well for two weeks.
So it is reasonable for the case.

Finally, any feedback about this process (positive or negative) would be greatly appreciated.
________________________________
zhangguanghui

From: Joseph Qi<mailto:joseph.qi@huawei.com>

Date: 2015-07-06 10:05
To: Andrew Morton<mailto:akpm@linux-foundation.org>
CC: Mark Fasheh<mailto:mfasheh@suse.com>; Joel Becker<mailto:jlbec@evilplan.org>; zhangguanghui 10102 (CCPL)<mailto:zhang.guanghui@h3c.com>; ocfs2-devel@oss.oracle.com<mailto:ocfs2-devel@oss.oracle.com>
Subject: [PATCH] ocfs2: do not BUG if buffer not uptodate in __ocfs2_journal_access

When storage network is unstable, it may trigger the BUG in
__ocfs2_journal_access because of buffer not uptodate. We can retry the
write in this case or return error instead of BUG.

Reported-by: Zhangguanghui <zhang.guanghui@h3c.com>
Signed-off-by: Joseph Qi <joseph.qi@huawei.com>

---
 fs/ocfs2/journal.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

--
1.8.4.3



-------------------------------------------------------------------------------------------------------------------------------------
????????????????????????????????????????
????????????????????????????????????????
????????????????????????????????????????
???
This e-mail and its attachments contain confidential information from H3C, which is
intended only for the person or entity whose address is listed above. Any use of the
information contained herein in any way (including, but not limited to, total or partial
disclosure, reproduction, or dissemination) by persons other than the intended
recipient(s) is prohibited. If you receive this e-mail in error, please notify the sender
by phone or email immediately and delete it!

Patch
diff mbox

diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index f8206d1..e731374 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -670,7 +670,23 @@  static int __ocfs2_journal_access(handle_t *handle,
                mlog(ML_ERROR, "giving me a buffer that's not uptodate!\n");
                mlog(ML_ERROR, "b_blocknr=%llu\n",
                     (unsigned long long)bh->b_blocknr);
-               BUG();
+
+               lock_buffer(bh);
+               /*
+                * A previous attempt to write this buffer head failed.
+                * Nothing we can do but to retry the write and hope for
+                * the best.
+                */
+               if (buffer_write_io_error(bh) && !buffer_uptodate(bh)) {
+                       clear_buffer_write_io_error(bh);
+                       set_buffer_uptodate(bh);
+               }
+
+               if (!buffer_uptodate(bh)) {
+                       unlock_buffer(bh);
+                       return -EIO;
+               }
+               unlock_buffer(bh);
        }

        /* Set the current transaction information on the ci so