[ILUG] Ubuntu Hoary box getting kernel OOPS

[Niall O Broin]

On 1 Sep 2005, at 04:11, Paul Jakma wrote:

> On Thu, 1 Sep 2005, Niall O Broin wrote:
>
>> I have a box getting kernel oops which it is surviving, somewhat i.e. 
>> I can still read the log but e.g. ps -ef just hangs, as does uname 
>> -a, ls / (but simply ls in /root works fine).  Looking in the log for 
>> "Call Trace" I see:
>>
>> Aug 31 00:17:08 localhost kernel: Call Trace:
>> Aug 31 00:17:08 localhost kernel:  [do_lookup+23/114] 
>> do_lookup+0x17/0x72
>
> Is that all? Only one line of a backtrace? I'm guessing there's more 
> but you've snipped all but the first line of each trace.

That's right - I was mindful of the length of the post. Each oops 
logged starts with an address (I presume - an 8 digit hex string) and 
then a line saying
kernel: PREEMPT,    and then gives a module list, flags, identifies the 
oopsing process, gives a few bytes of stack and then the call trace.

The hardware is now up and running on knoppix (earlier on 3.4 - now 
3.5) and I've had no oopses. It did hang once, but I'm putting that 
down to my dropping the keyboard and mouse.

I've been pushing it in the same way as I was the other box (running 
convert on a bunch of images).

So, is it an Ubuntu kernel issue maybe, related to this hardware? 
[Background - the hardware is my former deskunder, repurposed as a 
spare server. While under my desk it was solid as a rock running SuSE 
of various versions. The installed Ubuntu is on a disk which was in 
another box, and it never oopsed there, though it wasn't used much]


Niall