on the 2950 we only have a few VM's..  sipX with some windows 2003 servers.
 Our plan is to add additional ACD servers in the sipX cluster (once we get
that working).  We do have a large Dell R900 where I have about 15 VM's
running with multiple sipX servers, Windows servers, linux servers etc..  I
have not experienced that pause issue with either VM host..  that is
interesting.   We have internal storage on the 2950 and a Dell MD1000 SAS
array for the R900.
One thought..  on the rogue VM..  how many vCPU's do you have assigned?  If
you have the max assigned I could see this affecting other VM's possibly.
 Another thought is try limiting which cores the VM is allowed to use.  On
the VM properties  -->resources tab --> Advanced CPU setting.  Here you can
select which processor (cores) the VM can use.  we have 2 proc quad core in
our 2950, so all 8 cores show up there.  You can simply check off a handful
of cores, so the VM doesn't have access to all of the host CPU.  I'd be
curious if you still have the pause issue after configuring this.

James


On Thu, Jun 18, 2009 at 11:11 AM, Simon Stockdale <
[email protected]> wrote:

>  Are you running much else on the VM box? I have a similar (if not
> identical server) running a number of non-SipX VMs and I do often notice
> that if one VM goes rogue and spins on CPU it can disrupt other VMs. This
> often manifests itself as a 2-3 second pause on the other VMs. We have
> limited the resource of the sometimes rouge VMs but it still causes the
> strange pauses.
>
>
>
> FYI – we are running ESXi 3.5 with about 20ish VMs running on that box with
> a mix of local storage and iSCSI storage.
>
>
>
> Simon
>
>
>
_______________________________________________
sipx-users mailing list [email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-users
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-users
sipXecs IP PBX -- http://www.sipfoundry.org/

Reply via email to