2.6.8.1 Oops after raid1 disk failure

2.6.8.1 Oops after raid1 disk failure

[David R]

[-- Attachment #1: Type: text/plain, Size: 167 bytes --]

I got the following Oops (attached) after a drive failed in my raid1 
pair. I'm assuming the kernel shouldn't really do this as the other 
drive was ok?

Cheers
David

[-- Attachment #2: oops.txt --]
[-- Type: text/plain, Size: 1372 bytes --]


Aug 15 21:30:35 server kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
Aug 15 21:30:35 server kernel: hdc: dma_intr: error=0x40 { Uncorrectable Error } LBAsect=100651832, sector=100651832
Aug 15 21:30:35 server kernel: end_request: I/O error, dev hdc, sector 100651832  
Aug 15 21:30:35 server kernel: raid1: Disk failure on hdc3, disabling device
Oops: 0000 [#1]
PREEMPT
Modules linked in: thermal processor fan button battery ac binfmt_misc
CPU:    0
EIP:    0060:[<c0318200>]    Not tainted 
EFLAGS: 00010002   (2.6.8.1)
EIP is at read_intr+0x60/0x190
eax: dff8f240   ebx: 00000000   ecx: 00000000   edx: c063b10c
esi: fffffff8   edi: 00000008   ebp: 00000008   esp: ddee9f38
ds: 007b   es: 007b   ss: 0068
Process klogd (pid: 1778, threadinfo=ddee8000 task=dde12dd0)
Stack: c063b10c 00000001 00000008 dde12dd0 ddee8000 c063b10c 00000286 dff8f240
       c030e59c c063b10c 00515380 c03181a0 c063b060 dfd0af60 04000001 00000000
       ddee9fc4 c010850a 0000000e dff8f240 ddee9fc4 ddee8000 0000000e c05b4100
Call Trace:
 [<c030e59c>] ide_intr+0x11c/0x1e0
 [<c03181a0>] read_intr+0x0/0x190
 [<c010850a>] handle_IRQ_event+0x3a/0x70
 [<c01088d3>] do_IRQ+0xa3/0x170
 [<c0106b28>] common_interrupt+0x18/0x20
Code: 8b 73 18 89 ef 39 ee 0f 46 fe 29 fd 8b 53 38 0f b7 42 1a 8b
 <0>Kernel panic: Fatal exception in interrupt
In interrupt handler - not syncing

Re: 2.6.8.1 Oops after raid1 disk failure

[Erik Tews]

Am So, den 15.08.2004 schrieb David R um 23:17:
> I got the following Oops (attached) after a drive failed in my raid1 
> pair. I'm assuming the kernel shouldn't really do this as the other 
> drive was ok?

Hi

I see no raid-related parts in the backtrace, so the error seems to be
somewhere in the ide-layer.