Bug 5856 - Oops when fs nearly full
Summary: Oops when fs nearly full
Status: REJECTED INSUFFICIENT_DATA
Alias: None
Product: File System
Classification: Unclassified
Component: XFS (show other bugs)
Hardware: i386 Linux
: P2 high
Assignee: XFS Guru
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-01-09 04:06 UTC by Marc Dequ
Modified: 2007-02-17 11:58 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.6.14.3-grsec
Subsystem:
Regression: ---
Bisected commit-id:


Attachments

Description Marc Dequ 2006-01-09 04:06:28 UTC
Distribution: Debian GNU/Linux (testing)


Problem Description:

Filling /home with ISOs got an Oops when reaching the last 4-5% free space
available (last report from stats before they die). Then all processes using any
XFS partition hung, and softirqs brought CPU usage to 100%. Other (few)
softwares were still working nicely. After a necessary reboot, no corruption on
the fs was found.

Trace:
Jan  6 22:41:35 Orfeo kernel: Unable to handle kernel paging request at virtual
address befa9e3b
Jan  6 22:41:35 Orfeo kernel:  printing eip:
Jan  6 22:41:35 Orfeo kernel: c028f927
Jan  6 22:41:36 Orfeo kernel: *pgd =    0
Jan  6 22:41:36 Orfeo kernel: *pmd =    0
Jan  6 22:41:36 Orfeo kernel: Oops: 0000 [#1]
Jan  6 22:41:36 Orfeo kernel: PREEMPT 
Jan  6 22:41:36 Orfeo kernel: CPU:    0
Jan  6 22:41:36 Orfeo kernel: EIP:    0060:[xfs_ail_insert+39/144]    Not
tainted VLI
Jan  6 22:41:36 Orfeo kernel: EFLAGS: 00010212   (2.6.14.3-grsec-ipot) 
Jan  6 22:41:36 Orfeo kernel: eax: f79c1c14   ebx: ece45348   ecx: bb5343d2  
edx: befa9e2f
Jan  6 22:41:37 Orfeo kernel: esi: ece45348   edi: 000012ba   ebp: 0000363d  
esp: c1fdfe30
Jan  6 22:41:37 Orfeo kernel: ds: 007b   es: 007b   ss: 0068
Jan  6 22:41:37 Orfeo kernel: Process xfslogd/0 (pid: 188, threadinfo=c1fde000
task=c1f3f030)
Jan  6 22:41:37 Orfeo kernel: Stack: ece45348 f79c1c14 0000363d 000012ba
c028f5ef f79c1c14 ece45348 f79c1c00 
Jan  6 22:41:37 Orfeo kernel:        00000000 f55faf50 ece45348 0000363d
000012ba eea85bbc c028f1b4 f79c1c00 
Jan  6 22:41:37 Orfeo kernel:        ece45348 0000363d 000012ba 00000000
f79c1c00 00000000 c1fde000 eea84194 
Jan  6 22:41:37 Orfeo kernel: Call Trace:
Jan  6 22:41:37 Orfeo kernel:  [xfs_trans_update_ail+95/320]
Jan  6 22:41:37 Orfeo kernel:  [xfs_trans_chunk_committed+84/320]
Jan  6 22:41:37 Orfeo kernel:  [xfs_trans_committed+82/272]
Jan  6 22:41:37 Orfeo kernel:  [xlog_state_do_callback+380/768]
Jan  6 22:41:37 Orfeo kernel:  [xlog_state_done_syncing+115/176]
Jan  6 22:41:37 Orfeo kernel:  [xlog_iodone+76/208]
Jan  6 22:41:37 Orfeo kernel:  [worker_thread+477/704]
Jan  6 22:41:37 Orfeo kernel:  [pagebuf_iodone_work+0/80]
Jan  6 22:41:37 Orfeo kernel:  [default_wake_function+0/32]
Jan  6 22:41:37 Orfeo kernel:  [worker_thread+0/704]
Jan  6 22:41:37 Orfeo kernel:  [kthread+182/192]
Jan  6 22:41:37 Orfeo kernel:  [kthread+0/192]
Jan  6 22:41:38 Orfeo kernel:  [kernel_thread_helper+5/12]
Jan  6 22:41:38 Orfeo kernel: Code: 18 c3 89 f6 83 ec 10 8b 44 24 14 89 74 24 04
89 1c 24 89 7c 24 08 89 6c 24 0c 8b 74 24 18 8b 50 04 39 c2 74 45 8b 7e 0c 8b 6e
08 <39>
 7a 0c 8b 4a 08 74 31 72 07 8b 52 04 39 d0 75 ef 8b 02 89 56 
Jan  6 22:41:38 Orfeo kernel:  <6>note: xfslogd/0[188] exited with preempt_count 1

Kernel config for XFS:
CONFIG_XFS_FS=y
CONFIG_XFS_QUOTA=y
# CONFIG_XFS_SECURITY is not set
CONFIG_XFS_POSIX_ACL=y
# CONFIG_XFS_RT is not set


Steps to reproduce:

Was not yet able to reproduce, sorry.
Comment 1 Adrian Bunk 2006-12-07 07:50:33 UTC
Is this issue still present in kernel 2.6.19?
Comment 2 Adrian Bunk 2007-02-17 11:58:28 UTC
Please reopen this bug if it's still present with kernel 2.6.20.

Note You need to log in before you can comment on or make changes to this bug.