WARNING - OLD ARCHIVES

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/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

Re: [Xen-devel] Remus and Kemari...

To: gilberto nunes <gilberto.nunes@xxxxxxxxxxxxxxx>
Subject: Re: [Xen-devel] Remus and Kemari...
From: Brendan Cully <brendan@xxxxxxxxx>
Date: Wed, 20 Jan 2010 16:20:17 -0800
Cc: "xen-devel@xxxxxxxxxxxxxxxxxxx" <xen-devel@xxxxxxxxxxxxxxxxxxx>
Delivery-date: Wed, 20 Jan 2010 16:20:38 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <4B4F0581.30005@xxxxxxxxxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Mail-followup-to: gilberto.nunes@xxxxxxxxxxxxxxx, xen-devel@xxxxxxxxxxxxxxxxxxx
References: <4B4F0581.30005@xxxxxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mutt/1.5.20 (2009-12-10)
On Thursday, 14 January 2010 at 09:52, gilberto nunes wrote:
> I installed the Remus but had some problems.
> He leaves the VM very slow. I'm using a partition with DRBD / LVM.
> May even be that you say that has many layers (DRBD / LVM) that can
> influence on performance.
> But the interesting thing is that when Remus is not running the VM is light
> even with 512M of memory. And look what I'm talking about a Windows 2003
> Standard Edition.
> The Remus to stop working inexplicably (at least for me)
> and let the VM's started in a state of the two nodes ...

It's taken me a while to get Xen unstable running again, but now that
I have I can't reproduce any of these problems with an XP guest. It
remains responsive while Remus is running, and doesn't fail over
unless I kill it.

Disk access isn't likely to make a huge performance difference
(although I should say, simply parking a Remus VM on top of DRBD is
not safe, since there is no way of rolling back changes that have been
written since the most recent checkpoint). It sounds like your network
link between the primary and backup is either low capacity or flaky in
some way, or you have a loaded dom0.

I'd recommend using single-processor dom0 and guest to start with, and
pinning the VCPUs to their own separate physical cores. I'd also do a
bit of network measurement along the link you're using to the backup,
to check capacity, latency, and jitter.

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

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