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-users

Re: [Xen-users] Re: [Xen-devel] Re: XEN and failover

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: Re: [Xen-users] Re: [Xen-devel] Re: XEN and failover
From: Ernst Bachmann <e.bachmann@xxxxxxxx>
Date: Thu, 9 Feb 2006 11:49:58 +0100
Delivery-date: Thu, 09 Feb 2006 11:00:19 +0000
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <43EB13DC.2050308@xxxxxxxxx>
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: <43EA8A1B.6090901@xxxxxxxxx> <43EB13DC.2050308@xxxxxxxxx>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: KMail/1.9.1
On Thursday 09 February 2006 11:05, Thorolf Godawa wrote:
> Hi all,
>
> where do I find more information about the following:
>
> xc_save xcfd iofd domid maxit maxf flags
> xc_restore xcfd iofd domid nr_pfns store_evtchn console_evtchn
>
> What are the commands doing exactly and how are they doing it, what do
> the parameters mean exactly?
>
> Is it possible to realise a failover with this, can xc_save make
> snap-shots of a currently running VM without disturbing it much so if
> this host/guest crashes can I restore it quickly with the most actual
> status?

Well, to restore a snapshot, the on-disk and in-memory state has to match, 
otherwise you'd most likely badly corrupt your filesystem.

so you'd have to xm save your domain (stopping it), create a snapshot of all 
mounted filesystems and swap partitions, and restart the domain. I don't 
think that could be done in less than a second, iow, quite disturbing to your 
primary domU.

Next problem would be: if your master domU fails, your memory dump most likely 
already contains the reason for the failure (runaway process, memory leak, 
corrupted pointers...) so you don't gain much with that sort of failover, 
your secondary domU will most likely fail too in short order.

If you use an minimal kernel and tuned startup scripts starting only the 
absolutely needed services, a fresh reboot of a domU shouldn't take much more 
time than a restore from dump would, so for real-world applications that 
seems to be the better solution.

or, run a second domU with the same services in hot-standby, and just switch 
over the IP and MAC adress in case of failure.

/Ernst

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

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