lkml.org 
[lkml]   [2005]   [Aug]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: rc5 seemed to kill a disk that rc4-mm1 likes. Also some X trouble.
Danny ter Haar wrote:

>Andrew Morton <akpm@osdl.org> wrote:
>
>
>>Helge Hafting <helgehaf@aitel.hist.no> wrote:
>>
>>
>>>2.6.13-rc5 seemed to kill a scsi disk (sdb) for me, where 2.6.13-rc4-mm1
>>>have no problems with the same disk.
>>>
>>>
>
>Sort of same with me:
>2.6.12-mm1 runs for _weeks_ where others keep crashing:
>
>
>
>>The latest -git kernel (or 2.6.13-rc6 if it's there) with APCI enabled is
>>the one to test, please.
>>
>>
>
>no rc6 yet, i did however experience the following:
>
>reboot system boot 2.6.12-mm1 Sun Aug 7 18:20 (00:36)
>dth pts/1 zaphod.dth.net Sun Aug 7 15:41 - crash (02:38)
>reboot system boot 2.6.13-rc5-git5 Sun Aug 7 14:04 (04:52)
>reboot system boot 2.6.13-rc5-git4 Sun Aug 7 10:05 (03:43)
>reboot system boot 2.6.13-rc5-git3 Fri Aug 5 16:55 (1+17:07)
>
>git3 lasted near 2 days
>git4 ran for nearly 5 hours than i upgraded to
>git5 didn't last longer than 2.5 hours
>
>Fortunatly some info was found in the log files.
>What i dont "get" is that ethernet also goes down when the scsi
>controller goes bezerk.
>I'm pretty sure it's not a hardware problem since 2.6.12-mm1 survives
>and brings this usenet host in the worldwide top 1000.
>
>
Interesting.
I have no idea what the core problem is, but one problem will often lead
to others. My scsi problem froze some apps that couldn't be paged in
from the "failing" disk, for example.

Something going wrong in the kernel can delay other devices for too long,
maybe your network driver was hit by nasty latency in the middle of
something
as the scsi controller reset itself. It may also be memory scribbling.

I sometimes gets x lockups with rc5. Sometimes they just lock one display,
sometimes the whole machine locks solid necessitating a reset. sysrq+B
did not work, on either keyboard.

rc5 is no good for amd64, and it doesn't need power management to go wrong.

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

\
 
 \ /
  Last update: 2005-08-08 13:22    [W:0.194 / U:0.720 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site