Bug 247 - xfs_force_shutdown
: xfs_force_shutdown
Status: RESOLVED DUPLICATE of bug 755
Product: XFS
Classification: Unclassified
Component: XFS kernel code
: unspecified
: All Linux
: P1 normal
: ---
Assigned To: XFS power people
:
:
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2003-05-29 19:45 CDT by Masanobu Shimura
Modified: 2008-12-27 03:16 CST (History)
0 users

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Masanobu Shimura 2003-05-29 19:45:47 CDT
I am using Turbo linux sever 8 with xfs.

I experienced xfs_force_shutdown almost every day while deleting back up 
directory with rm -rf xxxx/ command.

Log Messages show as follows.

May 30 05:21:34 mpzserver2 kernel: xfs_force_shutdown(ide1(22,2),0x8) called 
from line 4065 of file xfs_bmap.c.  Return address = 0xd32a15e1
May 30 05:21:34 mpzserver2 kernel: Corruption of in-memory data detected.  
Shutting down filesystem: ide1(22,2)
May 30 05:21:34 mpzserver2 kernel: Please umount the filesystem, and rectify 
the problem(s)

After unmount device and perform xfs_repair usually show following messages.

Phase 1 - find and verify superblock...
Phase 2 - using internal log
        - zero log...
ERROR: The filesystem has valuable metadata changes in a log which needs to
be replayed.  Mount the filesystem to replay the log, and unmount it before
re-running xfs_repair.  If you are unable to mount the filesystem, then use
the -L option to destroy the log and attempt a repair.
Note that destroying the log may cause corruption -- please attempt a mount
of the filesystem before doing this.

Then, I remounted device and unmont again and perform xfs_repair to fix it.

Is this considered real memory problem or some bug in xfs file system?
I am using regular DIMM ram without ECC.

Please help me.

Mike Shimura
Comment 1 Max N. Boyarov 2006-01-24 09:44:55 CST
XFS corruption after transfering large ammount of data (~360GB) to RAID5. System
has been working during 4 days. Data transfers via CIFS/SMB with breaks and with
a few connections at the same time.

[root@HOST23002006 root]# xfs_repair /dev/evms/Volume1
Phase 1 - find and verify superblock...
Phase 2 - using internal log
- zero log...
ERROR: The filesystem has valuable metadata changes in a log which needs to
be replayed. Mount the filesystem to replay the log, and unmount it before
re-running xfs_repair. If you are unable to mount the filesystem, then use
the -L option to destroy the log and attempt a repair.
Note that destroying the log may cause corruption -- please attempt a mount
of the filesystem before doing this.
[root@HOST23002006 root]# mount -o ro /dev/evms/Volume1 /drives/Volume1/
mount: Unknown error 990

uname  -rmis
Linux 2.6.10-sanmina-aw22 armv5tel unknown
Comment 2 Christoph Hellwig 2008-12-27 03:16:43 CST
The second issue reported is the same as bug #755, as these are the symptoms for
the cache aliasing problems on virtual indexed caches.  As for the first one
without actually seeing the filesystems we can't really say much about it.

*** This bug has been marked as a duplicate of 755 ***