Re: 2.1.129: oops during boot from raid

Mark Lord (mlord@pobox.com)
Sat, 21 Nov 1998 16:59:42 +0000

Me Too! Me Too! Same Ooops, same place,
but with Fast! IDE drives instead of SCSI.

I thought for a moment that it might be a hard-to-find
IDE driver bug, but I see that Andreas managed to get it
to happen on a SCSI-only RAID0.

I first noticed this problem with .128/.129/.129+flufix
but that doesn't mean it didn't happen earlier.

In fact, the worrisome issue is that I get silent corruption
of data when WRITING to a filesystem on /dev/md0, and the
Ooops only shows up rarely.

We're talking *SEVERE* data loss here, folks. What broke it???

--
mlord@pobox.com




Andreas Schuldei wrote:
...
> Nov 20 22:07:07 sigrid kernel: md: Setting up md0 as a striped device.
> Nov 20 22:07:07 sigrid kernel: REGISTER_DEV sda5 to md0 done
> Nov 20 22:07:07 sigrid kernel: REGISTER_DEV sdb5 to md0 done
> Nov 20 22:07:07 sigrid kernel: raid0 : Allocating 8 bytes for hash.
> Nov 20 22:07:07 sigrid kernel: md: Loading md0.
...
> Unable to handle kernel NULL Pointer dereference at virtual address 00000008
> current->tss.cr3 = 00101000, %cr3 = 00101000
> *pde = 00000000
> Oops: 0000
> CPU:    1
> EIP:    0010:[<c016e0de>]
> EFLAGS: 00010212
> eax: 00005580   ebx: c4808028   ecx: 00000080   edx: 0000001e
> esi: 00000000   edi: 0015600e   ebp: 00000040   esp: c0095de0
> ds: 0018   es: 0018   ss: 0018
> Prosess swapper (pid: 1, process nr: 2, stackpage=c0095000)
> Stack: c3fdd84e 00000009 00000000 c4806000 002ac01c c4804000 00023620 00000000
>        00000004 c016c626 c01fadd4 c3fdd84e c3fdd850 00000002 00000000 c0095e5c
>        00000001 c016aeb1 00000000 c3fdd84e c3fdd850 00000002 0015600e c001b000
...
> >>EIP: c016e0de <raid0_map+a6/13c>
> 
> Code: c016e0de <raid0_map+a6/13c>
> Code: c016e0de <raid0_map+a6/13c>  8b 46 08             movl   0x8(%esi),%eax
> Code: c016e0e1 <raid0_map+a9/13c>  03 06                addl   (%esi),%eax
> Code: c016e0e3 <raid0_map+ab/13c>  39 c7                cmpl   %eax,%edi
> Code: c016e0e5 <raid0_map+ad/13c>  7c 27                jl     c016e10e <raid0_map+d6/13c>
> Code: c016e0e7 <raid0_map+af/13c>  8b 5b 04             movl   0x4(%ebx),%ebx
> Code: c016e0f0 <raid0_map+b8/13c>  85 db                testl  %ebx,%ebx
> Code: c016e0f2 <raid0_map+ba/13c>  75 1e                jne    c016e10c <raid0_map+d4/13c>
> Code: c016e0f4 <raid0_map+bc/13c>  57                   pushl  %edi
> Code: c016e0f5 <raid0_map+bd/13c>  68 6a dc 00 90       pushl  $0x9000dc6a
> Code: c016e100 <raid0_map+c8/13c>  90                   nop
> Code: c016e101 <raid0_map+c9/13c>  90                   nop



-- 
mlord@pobox.com


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