Re: swapper: page allocation failure. - random reboot problem

["Haar János" <djani22@netcenter.hu>]

----- Original Message ----- 
From: "Nick Piggin" <nickpiggin@yahoo.com.au>
To: "Haar János" <djani22@netcenter.hu>
Cc: <kernel@kolivas.org>; <cw@f00f.org>; <linux-kernel@vger.kernel.org>
Sent: Tuesday, May 23, 2006 12:45 PM
Subject: Re: swapper: page allocation failure. - random reboot problem


> Haar János wrote:
> > ----- Original Message ----- 
>
> >>But is the power supply rated enough to support all the drives?
> >>I have seen random reboots where the power supply wasn't good
> >>enough.
> >
> >
> > This is the 3rd modell, currently 550W.
> > The system is P4 3G.
> >
> > The another 2 stable node uses only 460W, and all hardware is equal.
> > But i tried to swap ps between the stable and unstabe nodes, but nothing
is
> > changed....
>
> Not sure then, sorry.

OK, what do you recommend? :-)


>
> If it is a software problem, then if you can narrow it down further
> (eg. kernel 2.6.15 worked, 2.6.16 did not), or find a reproducable
> test case for it, then some more progress might be made.

Yes, you have right!
I try it allready, but i cannot step back enough, because my sata card works
only on 2.6.16+

Before i use the promise sata cards, and the sata 300G hdds, i use 2.6.15
kernel, and all my 4 nodes was really stable.
But this is not enough to exactly find the problem. :-(

Anyway, Hetbert Xu helps me a lot to track down the problem, but we only can
close out some thing.
If i have right, the problem is about libata, promise driver,
sata-error-handling or similar, but not so sure.

If there is no way to software debug, i will swap all the 12 hdd between 2
disk nodes, and this can show i have right, or not.

Cheers,
Janos

>
> -- 
> SUSE Labs, Novell Inc.
> Send instant messages to your online friends http://au.messenger.yahoo.com