Consistent disk errors and oopsen

From: Michael T. Babcock (mbabcock@fibrespeed.net)
Date: Wed Sep 25 2002 - 11:24:10 EST


I have a client running 2.4.18 with FreeS/WAN 1.98b patched in and using
software RAID-1 on two IDE disks with Ext-3 on top of that. They are
getting consistent errors where the drives become un-sync'd but running
a raidhotadd seems to repair them just fine. They are also getting
these errors in the syslog:

Sep 25 10:46:51 gw kernel: journal_bmap_Rc2009e7d: journal block not
found at offset 7185 on md
Sep 25 10:46:51 gw kernel: Aborting journal on device md(9,3).
Sep 25 10:46:51 gw kernel: attempt to access beyond end of device
Sep 25 10:46:51 gw kernel: 09:03: rw=1, want=1625616132, limit=4192192
[... repeat the above two lines a few times ...]
Sep 25 10:46:51 gw kernel: Assertion failure in
__journal_remove_journal_head() at journal.c:1730: "buffer_jbd(bh)"

[... and then an oops: ]

>>EIP; d08181ce <[jbd]__journal_remove_journal_head+7e/e0> <=====
Trace; d0819e60 <[jbd].rodata.end+1a91/4cb9>
Trace; d0818a98 <[jbd].rodata.end+6c9/4cb9>
Trace; d08188bf <[jbd].rodata.end+4f0/4cb9>
Trace; d0818ab6 <[jbd].rodata.end+6e7/4cb9>
Trace; d0818279 <[jbd]journal_unlock_journal_head+49/60>
Trace; d0814804 <[jbd]journal_commit_transaction+854/e6a>
Trace; c010844d <do_IRQ+6d/b0>
Trace; c010846c <do_IRQ+8c/b0>
Trace; c010a498 <call_do_IRQ+5/d>
Trace; c0105a2e <__switch_to+3e/d0>
Trace; c01126b6 <schedule+2c6/2f0>
Trace; d08169d6 <[jbd]kjournald+106/1a0>
Trace; d08168b0 <[jbd]commit_timeout+0/10>
Trace; c0105726 <kernel_thread+26/30>
Trace; d08168d0 <[jbd]kjournald+0/1a0>
Code; d08181ce <[jbd]__journal_remove_journal_head+7e/e0>
00000000 <_EIP>:
Code; d08181ce <[jbd]__journal_remove_journal_head+7e/e0> <=====
   0: 0f 0b ud2a <=====
Code; d08181d0 <[jbd]__journal_remove_journal_head+80/e0>
   2: 83 c4 14 add $0x14,%esp
Code; d08181d3 <[jbd]__journal_remove_journal_head+83/e0>
   5: 39 1e cmp %ebx,(%esi)
Code; d08181d5 <[jbd]__journal_remove_journal_head+85/e0>
   7: 74 23 je 2c <_EIP+0x2c> d08181fa
<[jbd]__journal_remove_journal_head+aa/e0>
Code; d08181d7 <[jbd]__journal_remove_journal_head+87/e0>
   9: 68 c5 8a 81 d0 push $0xd0818ac5
Code; d08181dc <[jbd]__journal_remove_journal_head+8c/e0>
   e: 68 c3 06 00 00 push $0x6c3
Code; d08181e1 <[jbd]__journal_remove_journal_head+91/e0>
  13: 68 00 00 00 00 push $0x0

-- 
Michael T. Babcock
C.T.O., FibreSpeed Ltd.
http://www.fibrespeed.net/~mbabcock

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Mon Sep 30 2002 - 22:00:22 EST