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

Re: FW: [Xen-users] Problem with Dell Poweredge 1850


  • To: xen-users@xxxxxxxxxxxxxxxxxxx
  • From: David Harms <davidh.davidh@xxxxxxxxx>
  • Date: Mon, 9 May 2005 19:51:59 -0700
  • Delivery-date: Tue, 10 May 2005 02:51:34 +0000
  • Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=YPUEjSpFQsNNv4IVdBhsr+q/XdzMLo1tDdktDNCCBM11WaSu75suZaYKYn3hHUEWgTx0oJm/m74gndg1EWKEtk8hpbhgs0LEPXp4zpGYF3YXF8D+N6uHIageplSdpF/CJZUnoj4Pgx0k8Uaq/S8D27cn/Napd7Bqwe4abVLD6ec=
  • List-id: Xen user discussion <xen-users.lists.xensource.com>

I think I may be seeing the same problem on a set of supermicro server
based on the same Intel E7520 "Lindenhurst" chipset as the 1850.  I
can run xen-2.0.3 from the Jan 24th tarball using kernel 2.6.10 on
these systems without any problem but if I use the latest 2.0.5,
testing or unstable tarballs with 2.6.11 the server hangs under IO
load.  It often happens when I tar/untar, cp, or mkfs.  

booting xen with "ignorebiostables" seems to solve the problem.

I am running Redhat EL 3 & 4.
 
I don't know if it's relevant but it looks like there was a patch
added in 2.6.10/2.6.11 specifically for this chipset. "Disable SW
irqbalance/irqaffinity for E7520/E7320/E7525"
 
 David
 
> > -----Original Message-----
> > From: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
> > [mailto:xen-users-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of michal
> > urbanski
> > Sent: Monday, May 09, 2005 2:50 PM
> > To: xen-users@xxxxxxxxxxxxxxxxxxx
> > Subject: Re: [Xen-users] Problem with Dell Poweredge 1850
> >
> > On Sun, May 08, 2005 at 06:43:59PM +0200, Arie Kraai wrote:
> > > On Sun, May 08, 2005 at 10:15:45AM -0400, Rob See wrote:
> > > >
> > > > Any time the system is placed under high IO ( for example
> > untar/bz2ing a
> > > > tarball to create the fs for a xenU partiton) the machine hangs.
> > >
> > > Is this problem really xen-related; can you do the same actions booted
> > > from a "native" kernel without a crash?
> > > It looks like a memory or heating problem...
> >
> > I can state for certain that it is a xen problem. I've shelved the xen
> > stuff on our 2850 for the past few months, and the machine has been
> > running non-stop since then on a non-xen kernel.
> >
> > Just a data-point for you, fyi :)
> >
> > -michal
> >
> > _______________________________________________
> > Xen-users mailing list
> > Xen-users@xxxxxxxxxxxxxxxxxxx
> > http://lists.xensource.com/xen-users
> >
>

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.