Re: BUG/CRASH help - 2.6.20 (fc5) - spinlock bad magic on CPU#0, kswapd

Re: BUG/CRASH help - 2.6.20 (fc5) - spinlock bad magic on CPU#0, kswapd

[Robert Hancock]

Ian Kirk wrote:
> Hi,
> 
> I just wondered if the below type of crash was a known thing, and if
> there are any obvious things I can do to prevent/fix it ?
> 
> BUG: spinlock bad magic on CPU#0, kswapd0/242 (Not tainted)
> lock: c06c9380, .magic: c06c9380, .owner:
> [<c06eba2a>] _raw_spin_lock+0x1a/0xd9
> [<c061e72f>] _spin_lock_irqsave+0x9/0xd
> [<c041febf>] __wake_up+0x18/0x43
> [<c04379ce>] __wake_up_bit+0x2e/0x33
> [<c045a496>] shrink_inactive_list+0x483/0x6c9

RAM issues are a possibility for this sort of problem, tried running 
Memtest86+?

-- 
Robert Hancock      Saskatoon, SK, Canada
To email, remove "nospam" from hancockr@nospamshaw.ca
Home Page: http://www.roberthancock.com/

BUG/CRASH help - 2.6.20 (fc5) - spinlock bad magic on CPU#0, kswapd

[Ian Kirk]

Hi,

I just wondered if the below type of crash was a known thing, and if
there are any obvious things I can do to prevent/fix it ?

BUG: spinlock bad magic on CPU#0, kswapd0/242 (Not tainted)
lock: c06c9380, .magic: c06c9380, .owner:
[<c06eba2a>] _raw_spin_lock+0x1a/0xd9
[<c061e72f>] _spin_lock_irqsave+0x9/0xd
[<c041febf>] __wake_up+0x18/0x43
[<c04379ce>] __wake_up_bit+0x2e/0x33
[<c045a496>] shrink_inactive_list+0x483/0x6c9
[<c0459eae>] shrink_active_list+0x334/0x33c
[<c045a799>] shrink_zone+0xbd/0xe2
[<c045ac0e>] kswapd+0x296/0x3bb
[<c04379e9>] autoremove_wake_function+0x0/0x35
[<c045a978>] kswapd+0x0/0x3bb
[<c043792f>] kthread+0xb0/0xd9
[<c043787f>] kthread+0x0/0xd9
[<c0404b47>] kernel_thread_helper+0x7/0x10
=======================

The .owner: field contains a long list of unprintable characters.

The server is running kernel-smp-2.6.20-1.2307.fc5.i686.rpm.
Hardware of note is:

1x AMD AM2 DUALCORE 3800
2x Corsair DDR2 XMS 1Gb PC6400
2x 400G SEAGATE SATA2 ST3400633AS (traditional s/w RAID1 w/ ext3)
1x ASUS M2V-MX AM2 MOBO (onboard NIC unused)
1x e100 driver using NIC

I've a few things to try (disabling serial/par ports, unused NIC, USB).
It also crashed with a UP kernel.

Apologies if I'm duplicating any questions asked, I Googled for the
error messages but found quite different stack traces and contextx
(e.g. resiferfs vs ext3 - i'm only running ext3)

Regards,
Ian

(normally posting from blob@blob.co.uk but can't get to that as the
server is crashed!)