XFS Oopses with 2.4.5 and 2.4.14?
A message that is not in the archive (sent off-list or not yet received): [email protected]
Hi Pascal -
Did you compile these kernels yourself, and if so, what compiler did you use?
Can you reproduce this reliably?
I'd be happy to help with debugging this if you'd like, you might also take this over to [email protected].
-Eric
In article <[email protected]>, "Pascal Haakmat"
Quoted 56 lines · <[email protected]> wrote:
Given the following Oopses, is it wise to continue running the XFS filesystem, or might there be some other underlying problem that is causing these Oopses?
Kernel 2.4.5 + XFS 1.0.1:
Nov 25 07:28:23 awacs kernel: Unable to handle kernel paging request at virtual address 0cb73058 Nov 25 07:28:23 awacs kernel: printing eip: Nov 25 07:28:23 awacs kernel: c01d6493 Nov 25 07:28:23 awacs kernel:*pde = 00000000 Nov 25 07:28:23 awacs kernel: Oops: 0000 Nov 25 07:28:23
awacs kernel: CPU: 0 Nov 25 07:28:23 awacs kernel: EIP:
0010:[xfs_inactive_free_eofblocks+615/720] Nov 25 07:28:23 awacs kernel:
EFLAGS: 00010206 Nov 25 07:28:23 awacs kernel: eax:0cb73000ebx:dfecf400ecx:dfecf400edx:ce19b830Nov 25 07:28:23 awacs kernel:
esi:c0042bd8edi:da742f20ebp: 00000000 esp:dffe7f2cNov 25
07:28:23 awacs kernel: ds: 0018 es: 0018 ss: 0018 Nov 25 07:28:23
awacs kernel: Process kupdated (pid: 6, stackpage=dffe7000) Nov 25 07:28:23 awacs kernel: Stack:dfec6c0000000000dfec6c44dffe665000000001dfecf51400000010 00000001 Nov 25 07:28:23 awacs kernel:dfecf51400000000 00000010 00000001 00000000 00000008 00000008 00000040Nov 25 07:28:23 awacs kernel: 00000000 00000000 00000000 de64d500
df196f2000000026dfaeaa0000000000 Nov 25 07:28:23 awacs kernel: Call
Trace: [xfs_inactive_free_eofblocks+17/720] [load_msg+176/240] [block_write+1347/1376] [do_remount+122/172] [do_mount+537/740] [kernel_thread+35/48] Nov 25 07:28:23 awacs kernel:Nov 25 07:28:23 awacs kernel: Code: f7 40 58 ff 01 00 00 75 14 66 83 be52 01 00 00 00 75 0a 8b
Kernel 2.4.14 + XFS 1.0.2:
Nov 26 04:16:23 awacs kernel: Unable to handle kernel paging request at virtual address ff689108 Nov 26 04:16:23 awacs kernel: printing eip: Nov 26 04:16:23 awacs kernel: c01c5955 Nov 26 04:16:23 awacs kernel:*pde = 00000000 Nov 26 04:16:23 awacs kernel: Oops: 0000 Nov 26 04:16:23
awacs kernel: CPU: 1 Nov 26 04:16:23 awacs kernel: EIP:
0010:[xfs_syncsub+2309/3056] Not tainted Nov 26 04:16:23 awacs
kernel: EFLAGS: 00010246 Nov 26 04:16:23 awacs kernel: eax: 00000000
ebx: 00000000 ecx:dffbe914edx:c490bbd8Nov 26 04:16:23 awacs
kernel: esi:ff689100edi:c84ab740ebp:c03c4760esp:c1955f24Nov
26 04:16:23 awacs kernel: ds: 0018 es: 0018 ss: 0018 Nov 26 04:16:23
awacs kernel: Process kupdated (pid: 7, stackpage=c1955000) Nov 26 04:16:23 awacs kernel: Stack:dffbe91400000001 00000001 00000000 00000001 00000008 00000008 00000040 Nov 26 04:16:23 awacs kernel:
00000000 00000000 00000000dd2bda00c79a91a0c197bfacc0114539c1955f9cNov 26 04:16:23 awacs kernel: de302f40 dbde0000 0000001a 0000004600000286 00000001 00000286 00000003 Nov 26 04:16:23 awacs kernel: Call
Trace: [schedule+969/1504] [xfs_sync+21/32] [linvfs_write_super+42/48] [sync_supers+199/256] [sync_old_buffers+44/128] Nov 26 04:16:23 awacs
kernel: [kupdate+317/336] [_stext+0/64] [kernel_thread+38/48]
[kupdate+0/336] Nov 26 04:16:23 awacs kernel:Nov 26 04:16:23 awacs kernel: Code: 39 56 08 0f 85 22 f8 ff ff 8b 44 2478 05 14 01 00 00 50 e8 -
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Quoted 4 lines · Eric Sandeen
Hi Pascal -
Did you compile these kernels yourself, and if so, what compiler did you use?
Yes:
Linux version 2.4.5-xfs-1.0.1 ([email protected]) (gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)) #10 SMP Fri Sep 21 18:34:40 CEST 2001
Linux version 2.4.14-xfs-1.0.2 ([email protected]) (gcc version 2.96 20000731 (Red Hat Linux 7.1 2.96-98)) #2 SMP Sun Nov 25 08:15:50 CET 2001
Quoted 1 line
Can you reproduce this reliably?
No, that is to say, I haven't tried.
Somebody else has suggested FS corruption as the cause of these Oopses. That might very well be the case. On multiple occassions (perhaps coinciding with the Oopses, sorry I can't be more specific), my system has hung when trying to write to a file, and I would be left with files looking somewhat like this (from memory):
$ ls -lhsa spook.wav 0 -rw-r--r-- 1 p p 165k Nov 25 09:04 spook.wav
Up until now I just rm'd these files and continued, but I suppose I can try an xfs_repair from the boot CD.
Quoted 2 lines
I'd be happy to help with debugging this if you'd like, you might also take this over to [email protected].
Thanks.