[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Kernel paging errors



On Fri, Apr 09, 2004 at 01:26:19PM -0400, Nathaniel Eliot said
> I'm running a Debian Sarge installation, with the 2.4.22-1-386 kernel.
> Its a fairly lightweight server, running SSH and our accounting
> software.  I'm very unfamiliar with the error reporting of Debian, so
> bear with the stupid questions and conclusions.
> 
> I'm getting recurring kernel errors, that disappear for a while on
> reboot.  The error starts with "Unable to handle kernel paging request
> at virtual address".  Its initial EIP varies, but once the errors
> started the EIP remains the same until I reboot.  However, it hits many
> different processes, so I'm assuming the problem is something to do with
> a common library.
> 
> ----- error from kern.log -----
> Unable to handle kernel paging request at virtual address d0e8c3e0
>  printing eip:
> c013eb5b
> *pde = 00000000
> Oops: 0002
> CPU:    0
> EIP:    0010:[prune_dcache+135/295]    Not tainted
> EFLAGS: 00010246
> eax: ce9836f8   ebx: c0e8c368   ecx: c0e8c370   edx: d0e8c3e0
> esi: c0e8c350   edi: c0cbea00   ebp: 0000125e   esp: c12fdf60
> ds: 0018   es: 0018   ss: 0018
> Process kswapd (pid: 4, stackpage=c12fd000)
> Stack: 00000020 000001d0 00000006 00000037 c013ee53 00001919 c0128dd3
> 00000006
>        000001d0 c0233550 000001d0 00000006 c0233550 00000000 c0128e1d
> 00000020
>        c0233550 c12fc000 c02334a0 c0128efd c02334a0 00000000 c12fc245
> 0008e000
> Call Trace:    [shrink_dcache_memory+27/45] [shrink_caches+95/121]
> [try_to_free_pages_zone+48/74] [kswapd_balance_pgdat+65/136]
> [kswapd_balance+20/40] [kswapd+148/173] [rest_init+0/39]
> [arch_kernel_thread+35/45] [kswapd+0/173]
> 
> Code: 89 02 c7 41 04 00 00 00 00 c7 43 08 00 00 00 00 8b 7e 08 85
> ----- end error -----
> 
> I'm at a loss as to the source of these errors.  I've dug around with
> Google, without luck.  The memory and hard drive are both fully tested,
> and work fine.

Have you tried chkrootkit[0]?  A poorly-written rootkit could be corrupting
memory.

[0] read the BTS before posting any questions about it's output, it has
known false-positives.

-- 
Rob Weir <rweir@ertius.org> | mlspam@ertius.org  |  Do I look like I want a CC?
Words of the day:      CIDA Marxist Khaddafi assassinate radar Mafia Janet Reno

Attachment: signature.asc
Description: Digital signature


Reply to: