Dustin J. Mitchell wrote:
> On Mon, Aug 31, 2009 at 11:51 PM, Tom Robinson<tom.robinson AT motec.com DOT 
> au> wrote:
>   
>> While the disk is reaching saturation (and recovering quickly) I'm
>> thinking that the all the retransmissions would be slowing things down more.
>>
>> I don't see any errors on the client interface but there are four on the
>> server interface over the last four days.
>>     
>
> Hmm, the causation may be going the other way -- if the disk is
> generating too many IRQs for the CPU to handle, then network packets
> might get dropped.  Alternately, perhaps the PCI bus is maxed out?
> Anyway, this sounds like a problem local to the client.  Is there a
> way to slow down the disk IO so that it doesn't wedge the machine?
>   
Thanks Dustin,

I've found that our very old (RH7.1 seawolf), running a very old kernel
(2.4.20) has a bug in the ide driver. I can't say categorically that
this is the root cause of the dump issue I saw but, finally, I've got
permission to move forward with a planned upgrade that I've been pushing
for some time now.

For those that are interested, I suspect this is the problem:
https://bugzilla.redhat.com/show_bug.cgi?id=134579

Thanks for all the help

Regards,

Tom

-- 

Tom Robinson
System Administrator

MoTeC

121 Merrindale Drive
Croydon South
3136 Victoria
Australia

T: +61 3 9761 5050
F: +61 3 9761 5051   
M: +61 4 3268 7026
E: tom.robinson AT motec.com DOT au