I've got a new dual processor Opteron system I'm suspecting is flaky
when doing heavy disk I/O as I've had lots of kernel panics during
installs. Latest crash is the following while running bonnie++:
May 20 10:11:46 coop01 kernel: general protection fault: 0000 [1] SMP
May 20 10:11:46 coop01 kernel: CPU 1
May 20 10:11:46 coop01 kernel: Pid: 2456, comm: bonnie++ Not tainted
2.6.5-1.358smp
May 20 10:11:46 coop01 kernel: RIP: 0010:[<ffffffff80130cdf>]
<ffffffff80130cdf>{__wake_up
_common+55}
May 20 10:11:46 coop01 kernel: RSP: 0018:0000010002dadbd8 EFLAGS: 00010006
May 20 10:11:46 coop01 kernel: RAX: 0000010081b3ddf0 RBX:
0000000000000000 RCX: 0000010085
b3ddf0
May 20 10:11:46 coop01 kernel: RDX: 0000000000000000 RSI:
0000000000000003 RDI: 0000010085
b3ddd8
May 20 10:11:46 coop01 kernel: RBP: 0000010002dadc08 R08:
000001007f5fd580 R09: 000000037f
b0d4e0
May 20 10:11:46 coop01 kernel: R10: 0000000000000000 R11:
0000000000000000 R12: 0000010081
b3dde8
May 20 10:11:46 coop01 kernel: R13: 000000005b5a5958 R14:
7776757473727170 R15: 0000000000
000000
May 20 10:11:46 coop01 kernel: FS: 0000002a958624c0(0000)
GS:ffffffff80481f00(005b) knlGS
:00000000557752a0
May 20 10:11:46 coop01 kernel: CS: 0010 DS: 002b ES: 002b CR0:
000000008005003b
May 20 10:11:46 coop01 kernel: CR2: 0000002a9595ecc8 CR3:
00000000fbfa2000 CR4: 0000000000
0006e0
May 20 10:11:46 coop01 kernel: Process bonnie++ (pid: 2456,
stackpage=100f8fd2180)
May 20 10:11:47 coop01 kernel: Stack: 0000000381b3dde8 0000010081b3dde8
0000000000000000 0
000000000001000
May 20 10:11:47 coop01 kernel: 0000000000001000 0000000055570000
0000010002dadc28 f
fffffff80130d2b
May 20 10:11:47 coop01 kernel: 0000000000000246 0000000000001000
May 20 10:11:47 coop01 kernel: Call
Trace:<ffffffff80130d2b>{__wake_up+30} <ffffffff80150b
c4>{generic_file_aio_write_nolock+1952}
May 20 10:11:47 coop01 kernel:
<ffffffff80150da6>{generic_file_aio_write+93} <fffff
fffa0012faa>{:ext3:ext3_file_write+22}
May 20 10:11:47 coop01 kernel:
<ffffffff8016a980>{do_sync_write+102} <ffffffff801b8
0b4>{inode_has_perm+91}
May 20 10:11:47 coop01 kernel:
<ffffffff801b9c43>{selinux_file_permission+215} <fff
fffff801b9c7b>{selinux_file_permission+271}
May 20 10:11:47 coop01 kernel: <ffffffff8016aa7c>{vfs_write+208}
<ffffffff8016ab40>
{sys_write+57}
May 20 10:11:47 coop01 kernel: <ffffffff80122bd3>{cstar_do_call+27}
May 20 10:11:47 coop01 kernel:
May 20 10:11:47 coop01 kernel: Code: ff 51 f8 85 c0 74 0a 41 80 e5 01 74
04 ff cb 74 10 4c
89 f1
May 20 10:11:47 coop01 kernel: RIP
<ffffffff80130cdf>{__wake_up_common+55} RSP <0000010002
dadbd8>
memtest86+ checks out fine. Not sure what else to do to try to narrow
down the problem. Any help would be appreciated.
