Red Hat Bugzilla – Bug 114279
reiserfs oops 74min into copying 50G
Last modified: 2007-11-30 17:10:35 EST
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040120 Description of problem: I was copying a tree from a reiserfs volume on a GDTH hardware RAID0 set to a software RAID0 ATA pair. Boom: kernel BUG at buffer.c:604! invalid operand: 0000 ide-cd cdrom via82cxxx_audio ac97_codec uart401 sound soundcore nfs lockd sunrpc parport_pc lp parport autofs e100 floppy sg raid0 keybdev mousedev hid input CPU: 0 EIP: 0060:[<c014529e>] Not tainted EFLAGS: 00010286 EIP is at __insert_into_lru_list [kernel] 0x1e (2.4.22-1.2149.nptl) eax: 8bbe968e ebx: 00000002 ecx: f0220080 edx: c03d604c esi: f0220080 edi: f0220080 ebp: 00001000 esp: e6edbe64 ds: 0068 es: 0068 ss: 0068 Process cp (pid: 16878, stackpage=e6edb000) Stack: 00000002 c0145c8b f0220080 00000002 f0220080 00001000 c0146882 f0220080 00000000 00014000 00000000 c1f22e20 efd51980 c014707f efd51980 c1f22e20 00000000 00001000 e6edbee8 00014000 00000000 efd51980 f886538d e7696200 Call Trace: [<c0145c8b>] __refile_buffer [kernel] 0x5b (0xe6edbe68) [<c0146882>] __block_commit_write [kernel] 0xb2 (0xe6edbe7c) [<c014707f>] generic_commit_write [kernel] 0x3f (0xe6edbe98) [<f886538d>] reiserfs_commit_write [reiserfs] 0x13d (0xe6edbebc) [<f8887090>] .LC30 [reiserfs] 0x2c (0xe6edbee8) [<c01366f3>] do_generic_file_write [kernel] 0x283 (0xe6edbf18) [<c0136c86>] generic_file_write [kernel] 0x136 (0xe6edbf68) [<c0143a13>] sys_write [kernel] 0xa3 (0xe6edbf94) [<c014b9a9>] sys_fstat64 [kernel] 0x49 (0xe6edbfa8) [<c0109747>] system_call [kernel] 0x33 (0xe6edbfc0) Code: 0f 0b 5c 02 17 0a 28 c0 8b 02 85 c0 75 07 89 0a 89 49 24 8b Version-Release number of selected component (if applicable): kernel-2.4.22-1.2149.nptl How reproducible: Didn't try Steps to Reproduce: 1. cp -a /vol1/* /vol2/ Actual Results: 34G copied Expected Results: 50G copied Additional info: This machine was a dual PIII till it caught fire. We replaced mobo, RAM, CPU (to an Athlon but I wasn't allowed to install the Athlon kernel..), video card, RAID card, PSU and one hard drive (which is where the fire was).. I believe the RAM is OK (we regularly run memtest86) but will find time to run it again..
Preliminary memtest (default tests) ran fine for 2 passes; tried the cp again and it worked fine; now running memtest (all tests..)--will follow up if it fails! Should we put this down to gamma rays/the dodgy music coming out of the machine next door?
memtest (all tests) was fine (for one pass). It's now been running 2.6.1-1.57 happily for 24h.
the vm updates circa 2163 should have also fixed this.