[Bug 414763] New: XFS file system corruption
https://bugzilla.novell.com/show_bug.cgi?id=414763 User tkutergin@mail.ru added comment https://bugzilla.novell.com/show_bug.cgi?id=414763#c1 Summary: XFS file system corruption Product: openSUSE 11.1 Version: Factory Platform: i686 OS/Version: openSUSE 11.0 Status: NEW Severity: Major Priority: P5 - None Component: Kernel AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: tkutergin@mail.ru QAContact: qa@suse.de Found By: Customer /home mounted on XFS file system was unexpectedly corrupted making it unavailable. User session was unusable after this since /home was not accessible. Immediate activity before crash was internet browsing (before crash there were several popups complaining about inability to write metadata and suggestions to disable beagle) System restart made this available again. Here is mount output: /dev/mapper/system-root on / type xfs (rw) proc on /proc type proc (rw) sysfs on /sys type sysfs (rw) debugfs on /sys/kernel/debug type debugfs (rw) udev on /dev type tmpfs (rw) devpts on /dev/pts type devpts (rw,mode=0620,gid=5) /dev/sda1 on /boot type ext3 (rw,acl,user_xattr) /dev/mapper/system-home on /home type xfs (rw) securityfs on /sys/kernel/security type securityfs (rw) fusectl on /sys/fs/fuse/connections type fusectl (rw) gvfs-fuse-daemon on /var/lib/gdm/.gvfs type fuse.gvfs-fuse-daemon (rw,nosuid,nodev,user=gdm) none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw) none on /proc/fs/vmblock/mountPoint type vmblock (rw) gvfs-fuse-daemon on /home/tkutergi/.gvfs type fuse.gvfs-fuse-daemon (rw,nosuid,nodev,user=tkutergi) Here is df output Filesystem 1K-blocks Used Available Use% Mounted on /dev/mapper/system-root 7469056 4689268 2779788 63% / udev 399628 108 399520 1% /dev /dev/sda1 69972 18265 48094 28% /boot /dev/mapper/system-home 11216896 571596 10645300 6% /home df: `/var/lib/gdm/.gvfs': Permission denied gvfs-fuse-daemon 7469056 4689268 2779788 63% /home/tkutergi/.gvfs Messages in /var/log/messages: Aug 4 20:26:12 linux-rn6q kernel: Filesystem "dm-0": XFS internal error xfs_trans_cancel at line 1163 of file fs/xfs/xfs_trans.c. Caller 0xf29b5bca Aug 4 20:26:12 linux-rn6q kernel: Pid: 3226, comm: firefox Tainted: G N 2.6.25.5-1.1-default #1 Aug 4 20:26:12 linux-rn6q kernel: [<c01071d9>] dump_trace+0x63/0x227 Aug 4 20:26:12 linux-rn6q kernel: [<c0107c8a>] show_trace+0x15/0x29 Aug 4 20:26:12 linux-rn6q kernel: [<c02e6538>] _etext+0x5b/0x65 Aug 4 20:26:12 linux-rn6q kernel: [<f299655b>] xfs_error_report+0x2c/0x2e [xfs] Aug 4 20:26:12 linux-rn6q kernel: [<f29b047c>] xfs_trans_cancel+0x48/0xd3 [xfs] Aug 4 20:26:12 linux-rn6q kernel: [<f29b5bca>] xfs_create+0x38b/0x3be [xfs] Aug 4 20:26:12 linux-rn6q kernel: [<f29beaac>] xfs_vn_mknod+0x12f/0x215 [xfs] Aug 4 20:26:12 linux-rn6q kernel: [<f29bebab>] xfs_vn_create+0xa/0xc [xfs] Aug 4 20:26:12 linux-rn6q kernel: [<c017e681>] vfs_create+0x12e/0x19d Aug 4 20:26:12 linux-rn6q kernel: [<c018054f>] open_namei+0x159/0x596 Aug 4 20:26:12 linux-rn6q kernel: [<c0175543>] do_filp_open+0x20/0x36 Aug 4 20:26:12 linux-rn6q kernel: [<c0175599>] do_sys_open+0x40/0xbb Aug 4 20:26:12 linux-rn6q kernel: [<c0175656>] sys_open+0x1e/0x26 Aug 4 20:26:12 linux-rn6q kernel: [<c0105a62>] syscall_call+0x7/0xb Aug 4 20:26:12 linux-rn6q kernel: [<ffffe422>] 0xffffe422 Aug 4 20:26:12 linux-rn6q kernel: ======================= Aug 4 20:26:12 linux-rn6q kernel: xfs_force_shutdown(dm-0,0x8) called from line 1164 of file fs/xfs/xfs_trans.c. Return address = 0xf29b0492 Aug 4 20:26:12 linux-rn6q kernel: Filesystem "dm-0": Corruption of in-memory data detected. Shutting down filesystem: dm-0 Aug 4 20:26:12 linux-rn6q kernel: Please umount the filesystem, and rectify the problem(s) Aug 4 20:26:56 linux-rn6q gconfd (tkutergi-2872): Failed to open saved state file: Failed: Failed to open gconfd logfile; won't be able to restore listeners after gconfd shutdown (Input/output error) Aug 4 20:26:56 linux-rn6q gconfd (tkutergi-2872): Could not open saved state file '/home/tkutergi/.gconfd/saved_state.tmp' for writing: Input/output error Aug 4 20:27:26 linux-rn6q gconfd (tkutergi-2872): Failed to open saved state file: Failed: Failed to open gconfd logfile; won't be able to restore listeners after gconfd shutdown (Input/output error) Aug 4 20:27:26 linux-rn6q gconfd (tkutergi-2872): Could not open saved state file '/home/tkutergi/.gconfd/saved_state.tmp' for writing: Input/output error -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=414763
Timofey Kutergin
https://bugzilla.novell.com/show_bug.cgi?id=414763
Cyril Hrubis
https://bugzilla.novell.com/show_bug.cgi?id=414763
User jeffm@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=414763#c1
Jeff Mahoney
https://bugzilla.novell.com/show_bug.cgi?id=414763
User markgw@sgi.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=414763#c2
Mark Goodwin
SGI guys, there's been a flurry of commits upstream for XFS in the last two weeks.
yes there has - the patches addressing assorted corruption issues have been released as a PTF, see BZ 374973.
Have you seen this problem before and do any of the commits address it?
This particular bug does not seem to fit any of the traces seen in BZ 374973. In this case, it looks like you're getting EIO on a dm volume. Are you sure there is no h/w error involved here, or a piece of the volume is not on-line? Thanks -- Mark Goodwin markgw@sgi.com Engineering Manager for XFS and PCP Phone: +61-3-99631937 SGI Australian Software Group Cell: +61-4-18969583 ------------------------------------------------------------- -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=414763
User susexfs@sgi.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=414763#c3
--- Comment #3 from SGI XFS Developers
https://bugzilla.novell.com/show_bug.cgi?id=414763
User markgw@sgi.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=414763#c4
--- Comment #4 from Mark Goodwin
Where did you get EIO from?
Saw the I/O error for gconfd, but reading that again, it's probably not related.
The filesystem was shutdown because we encountered an error while we had a dirty transaction in progress. This could have been any error.
We have a known issue with alignment during inode allocation. See this email thread:
yes - and that's not resolved yet I don't think. Thanks -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=414763
User jeffm@novell.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=414763#c5
Jeff Mahoney
https://bugzilla.novell.com/show_bug.cgi?id=414763
User markgw@sgi.com added comment
https://bugzilla.novell.com/show_bug.cgi?id=414763#c6
Mark Goodwin
Any news here?
Dave Chinner recently posted a patch to fix a corruption problem with very similar stack trace: http://oss.sgi.com/archives/xfs/2008-09/msg00484.html This patch fixes a known test case and is under review but we haven't yet taken it in. -- Mark -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=414763
Greg Kroah-Hartman
participants (1)
-
bugzilla_noreply@novell.com