This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


Re: [Xen-users] suspect cpu utilization level

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: Re: [Xen-users] suspect cpu utilization level
From: Mark Williamson <mark.williamson@xxxxxxxxxxxx>
Date: Fri, 19 Oct 2007 17:16:24 +0100
Cc: Marco Strullato <marco.strullato@xxxxxxxxx>
Delivery-date: Fri, 19 Oct 2007 09:17:40 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <b9f669850710080712n76341e98l1b3bb23c21114744@xxxxxxxxxxxxxx>
List-help: <mailto:xen-users-request@lists.xensource.com?subject=help>
List-id: Xen user discussion <xen-users.lists.xensource.com>
List-post: <mailto:xen-users@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
References: <b9f669850710080712n76341e98l1b3bb23c21114744@xxxxxxxxxxxxxx>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: KMail/1.9.6
> Hi all, I've a linux dom0, a dom1 winxp sp2 (WinXP) and another winxp sp2
> with 2 vcpu (WinXP_02)
> Dom2 uses all the dom0 idle: I mean this vm uses too much resources, it
> seems something wrong in configuration.
> This is strange because the other win xp uses very few resources

Are the guests completely identical?  It does seem strange that it's using 
more resources if they're the same.

> I'll attach some useful output..


> I changed the xen scheduler giving more weight to WinXP_02 but it wasn't
> useful
> [root@localhost ~]# xm sched-credit
> Name                              ID Weight Cap
> Domain-0                           0    256    0
> WinXP                              5    256    0
> WinXP_02                           4    512    0

Giving more weight to a domain will cause it to run more, not less.  So 
increasing the weight would be expected to increase its CPU utilisation (see 

> [root@localhost ~]# xm vcpu-list
> Name                              ID  VCPU   CPU State   Time(s) CPU
> Affinity
> Domain-0                           0     0     0   r--     822.8 any cpu
> Domain-0                           0     1     0   ---     431.7 any cpu
> WinXP                              5     0     0   -b-     139.1 any cpu
> WinXP                              5     1     0   -b-     140.2 any cpu
> WinXP_02                           4     0     0   ---     624.8 any cpu
> WinXP_02                           4     1     1   r--     621.8 any cpu

It says both domain 5's CPUs are blocked on IO, which suggests they're waiting 
for IO and therefore can't run.

It looks like domain 4 isn't blocked on IO and so it's continuing to run.

By the way, it looks like you have two vcpus in dom0?  That's not usually 
necessary; giving dom0 just one vcpu should still enable it to serve other 
domains well and will leave more CPU time for other uses.

Does that help you somewhat?  What are you expected / wanting to happen?


Dave: Just a question. What use is a unicyle with no seat?  And no pedals!
Mark: To answer a question with a question: What use is a skateboard?
Dave: Skateboards have wheels.
Mark: My wheel has a wheel!

Xen-users mailing list

<Prev in Thread] Current Thread [Next in Thread>