kernel BUG at buffer.c

From: Andreas Schmidt
Date: Fri Apr 23 2004 - 11:52:01 EST


Hi,

lately, I've repeatedly found kernel-bug warnings in my logs. The context varies, but time and again the message "kernel BUG at buffer. c:575!" pops up.
The box that happens on is running Debian (Woody) with kernel 2.4.25. It has an K7TA266 Mainboard, AMD 700MHz processor, 512MB RAM and 3 IDE- harddisks (Samsung SV1203N),

I suspect that the latest error was a result of over-burdening the poor computer: it was to compress a 25GB diskimage with bzip2 and had been running for about 12 hrs before segfaulting. Here the log:

Apr 22 22:41:45 stralsunder-10 kernel: kernel BUG at buffer.c:575!
Apr 22 22:41:45 stralsunder-10 kernel: invalid operand: 0000
Apr 22 22:41:45 stralsunder-10 kernel: CPU: 0
Apr 22 22:41:45 stralsunder-10 kernel: EIP:
0010:[__insert_into_lru_list+28/112] Tainted: P
Apr 22 22:41:45 stralsunder-10 kernel: EFLAGS: 00010206
Apr 22 22:41:45 stralsunder-10 kernel: eax: 00000000 ebx: 00000002 ecx: d2fa00c0 edx: c02ab754
Apr 22 22:41:45 stralsunder-10 kernel: esi: d2fa00c0 edi: d2fa00c0 ebp: 00000001 esp: dc071ebc
Apr 22 22:41:45 stralsunder-10 kernel: ds: 0018 es: 0018 ss: 0018
Apr 22 22:41:45 stralsunder-10 kernel: Process bzip2 (pid: 2263,
stackpage=dc071000)
Apr 22 22:41:45 stralsunder-10 kernel: Stack: 00000002 c0135976 d2fa00c000000002 d2fa00c0 00001000 c013598a d2fa00c0
Apr 22 22:41:45 stralsunder-10 kernel: c0136383 d2fa00c0 00001000 df029180 31dd3000 00000000 00001000 00000000
Apr 22 22:41:45 stralsunder-10 kernel: c01369f4 df029180 c134caa4 00000000 00001000 c134caa4 40015000 d3327000
Apr 22 22:41:45 stralsunder-10 kernel: Call Trace: [__refile_buffer +86/96] [refile_buffer+10/16] [__block_commit_write+131/208]
[generic_commit_write+52/96] [do_generic_file_write+654/976]
Apr 22 22:41:45 stralsunder-10 kernel: [generic_file_write+259/288]
[sys_write+150/240] [system_call+51/56]
Apr 22 22:41:45 stralsunder-10 kernel:
Apr 22 22:41:45 stralsunder-10 kernel: Code: 0f 0b 3f 02 65 4c 21 c0 83 3a 00 75 05 89 0a 89 49 24 8b 02


In another case, the error occured more randomly. I was upping some files per ftp when, once again, a bug occured at buffer.c; however, there also seems to have been something wrong with the ext3-journal, as can be seen in this entry:

Apr 20 06:22:59 stralsunder-10 kernel: kernel BUG at buffer.c:575!
Apr 20 06:22:59 stralsunder-10 kernel: invalid operand: 0000
Apr 20 06:22:59 stralsunder-10 kernel: CPU: 0
Apr 20 06:22:59 stralsunder-10 kernel: EIP:
0010:[__insert_into_lru_list+28/112] Tainted: P
Apr 20 06:22:59 stralsunder-10 kernel: EFLAGS: 00010206
Apr 20 06:22:59 stralsunder-10 kernel: eax: 00000000 ebx: 00000002 ecx: d06140c0 edx: c02ab754
Apr 20 06:22:59 stralsunder-10 kernel: esi: d06140c0 edi: d06140c0 ebp: 00000001 esp: d227be6c
Apr 20 06:22:59 stralsunder-10 kernel: ds: 0018 es: 0018 ss: 0018
Apr 20 06:22:59 stralsunder-10 kernel: Process proftpd (pid: 32653,
stackpage=d227b000)
Apr 20 06:22:59 stralsunder-10 kernel: Stack: 00000002 c0135976 d06140c0 00000002 d06140c0 00001000 c013598a d06140c0
Apr 20 06:22:59 stralsunder-10 kernel: c0136383 d06140c0 00000220 c2bb3500 0792f220 00000000 00001000 00000000
Apr 20 06:22:59 stralsunder-10 kernel: c01369f4 c2bb3500 c1425168 00000000 00000220 c1425168 080abb24 c2bb3500
Apr 20 06:22:59 stralsunder-10 kernel: Call Trace: [__refile_buffer +86/96] [refile_buffer+10/16] [__block_commit_write+131/208]
[generic_commit_write+52/96] [ext3_commit_write+305/448]
Apr 20 06:22:59 stralsunder-10 kernel: [do_generic_file_write +654/976] [generic_file_write+259/288] [ext3_file_write+35/192] [sys_write+150/240] [system_call+51/56]
Apr 20 06:22:59 stralsunder-10 kernel:
Apr 20 06:22:59 stralsunder-10 kernel: Code: 0f 0b 3f 02 65 4c 21 c0 83 3a 00 75 05 89 0a 89 49 24 8b 02
Apr 20 06:22:59 stralsunder-10 kernel: <0>Assertion failure in journal_start() at transaction.c:251: "handle->h_transaction->t_journal == journal"
Apr 20 06:22:59 stralsunder-10 kernel: kernel BUG at transaction.c:251!
Apr 20 06:22:59 stralsunder-10 kernel: invalid operand: 0000
Apr 20 06:22:59 stralsunder-10 kernel: CPU: 0
Apr 20 06:22:59 stralsunder-10 kernel: EIP: 0010:[journal_start +74/192] Tainted: P
Apr 20 06:22:59 stralsunder-10 kernel: EFLAGS: 00010286
Apr 20 06:22:59 stralsunder-10 kernel: eax: 0000006c ebx: d09e7440 ecx: df886000 edx: 00000001
Apr 20 06:22:59 stralsunder-10 kernel: esi: dfe4a600 edi: d227a000 ebp: 00000040 esp: d227bc10
Apr 20 06:22:59 stralsunder-10 kernel: ds: 0018 es: 0018 ss: 0018
Apr 20 06:22:59 stralsunder-10 kernel: Process proftpd (pid: 32653,
stackpage=d227b000)
Apr 20 06:22:59 stralsunder-10 kernel: Stack: c021a740 c021a96c c021a720 000000fb c021a940 d09e7440 c1590400 c5ec1700
Apr 20 06:22:59 stralsunder-10 kernel: c015e958 dfe4a600 00000002 c5ec1700 c1590400 00000001 c014635e c5ec1700
Apr 20 06:22:59 stralsunder-10 kernel: c5ec1700 c5ec176c dded1140 c0128c83 c5ec1700 00000001 c5ec1700 c5ec176c
Apr 20 06:22:59 stralsunder-10 kernel: Call Trace: [ext3_dirty_inode
+88/256] [__mark_inode_dirty+46/144] [do_generic_file_write+211/976]
[check_free_space+290/320] [generic_file_write+259/288]
Apr 20 06:22:59 stralsunder-10 kernel: [ext3_file_write+35/192]
[do_acct_process+571/592] [acct_process+25/39] [do_exit+105/592]
[do_invalid_op+0/160] [die+86/96]
Apr 20 06:22:59 stralsunder-10 kernel: [do_invalid_op+140/160]
[__insert_into_lru_list+28/112] [ext3_get_block_handle+426/640]
[ext3_get_block_handle+242/640] [error_code+52/60] [__insert_into_lru_list+28/112]
Apr 20 06:22:59 stralsunder-10 kernel: [__refile_buffer+86/96]
[refile_buffer+10/16] [__block_commit_write+131/208]
[generic_commit_write+52/96] [ext3_commit_write+305/448]
[do_generic_file_write+654/976]
Apr 20 06:22:59 stralsunder-10 kernel: [generic_file_write+259/288]
[ext3_file_write+35/192] [sys_write+150/240] [system_call+51/56]
Apr 20 06:22:59 stralsunder-10 kernel:
Apr 20 06:22:59 stralsunder-10 kernel: Code: 0f 0b fb 00 20 a7 21 c0 83 c4 14 ff 43 08 89 d8 eb 59 8d 74

I was planning on upgrading the kernel to 2.4.26 anyway; would that be sufficient to resolve these issues, or is there something completely different wrong?

Best regards

Andreas


PS: Please CC me as I'm not on this list.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/