My favorites | Sign in
Project Home Downloads Wiki Issues Source
READ-ONLY: This project has been archived. For more information see this post.
Search
for
  Advanced search   Search tips   Subscriptions
Issue 98: Selective relocation Crashing
  Back to list
Status:  WontFix
Owner:  checkout...@gmail.com
Closed:  Aug 2009


 
Project Member Reported by checkout...@gmail.com, Jul 18, 2009
Select relocation crashing.
Getting a oops message.

Looking into it.
Jul 18, 2009
Project Member #1 sandeepksinha
Can you provide the trace for the same????
Summary: Selective relocation Crashing
Jul 18, 2009
Project Member #2 checkout...@gmail.com
This is the trace.

WARNING: at fs/fs-writeback.c:396 __writeback_single_inode+0x39/0x30c()
[11128.311513] Hardware name:         
[11128.311516] Modules linked in: ohsm ext2 dm_mirror dm_region_hash dm_log
dm_multipath dm_mod [last unloaded: ohsm]
[11128.311535] Pid: 7349, comm: ohsm Tainted: G        W  2.6.30.1 #3
[11128.311540] Call Trace:
[11128.311547]  [<c1026fb8>] warn_slowpath_common+0x60/0x90
[11128.311555]  [<c1026ff5>] warn_slowpath_null+0xd/0x10
[11128.311562]  [<c10a574f>] __writeback_single_inode+0x39/0x30c
[11128.311570]  [<c10a5a3e>] sync_inode+0x1c/0x26
[11128.311580]  [<f88d5a0e>] ohsm_sync_inode+0x29/0x30 [ohsm]
[11128.311590]  [<f88d5cf0>] ohsm_realloc_mcp+0x2db/0x31a [ohsm]
[11128.311601]  [<f88d6245>] ohsm_get_inodes_for_relocation+0x288/0x315 [ohsm]
[11128.311612]  [<f88d65b0>] device_ioctl+0x2de/0x55e [ohsm]
[11128.311618]  [<c109a44b>] vfs_ioctl+0x50/0x69
[11128.311625]  [<c109a8aa>] do_vfs_ioctl+0x446/0x47f
[11128.311632]  [<c102b2d8>] ? __do_softirq+0x121/0x129
[11128.311639]  [<c109a923>] sys_ioctl+0x40/0x5a
[11128.311646]  [<c1002a04>] sysenter_do_call+0x12/0x22
[11128.311651] ---[ end trace 4dba69f08dbca8b9 ]---

Jul 18, 2009
Project Member #3 sandeepksinha
Vineet any updates on this?
Jul 19, 2009
Project Member #4 checkout...@gmail.com
I tried with BUG_ON but no clues as to why this oops message was generated.

But when tested on a fresh file system with correct inputs. The whole thing is
working fine.
Aug 1, 2009
Project Member #5 sandeepksinha
I think this was because of the testing being formed on a corrupt filesystem.
The file system got corrupted due to crash in some of the previous tests.
This issue has no more been seen in the recent tests.

Hence closing the issue.

Status: WontFix

Powered by Google Project Hosting