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] xen sparce files on ocfs2/gfs

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: Re: [Xen-users] xen sparce files on ocfs2/gfs
From: "Christopher G. Stach II" <cgs@xxxxxxxxx>
Date: Thu, 04 Jan 2007 17:59:35 -0200
Delivery-date: Thu, 04 Jan 2007 11:59:50 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <2F8578C9-D80E-42E4-A9A5-FEDAF6AC4E06@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: <2F8578C9-D80E-42E4-A9A5-FEDAF6AC4E06@xxxxxxxxx>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Thunderbird 1.5.0.9 (X11/20061219)
Brock Palen wrote:
> Well my two node xen cluster is coming along.  The nodes will mirror two
> partitions (using drbd) one for holding the xen images the other for
> user data  (global /home).    I figure i will export a ocfs2 or gfs
> filesystem to both node dom0's via iscsi.  They will then hold sparse
> files for the xen images (regular ext3).
> 
> Is this doable?   Is there a better way?  A dedicated shared storage box
> is out of the question for the foreseeable future.  I want to use
> heartbeat to provide HA ability and DRBD will keep the data mirrored.  
> I worry about having the iSCSI target having a floating ip with drbd. 
> And split brain,  both nodes try to attached to the sparefile for one domU.
> 
> Has anyone done anything similar?  And what about using GFS vs OCFS2 ? 
> I  would like to avoid sparce files, and export real devices,  something
> like a cluster aware LVM but finding information on such a thing on-top
> of DRBD has not been fruitful, and all the GFS options for providing
> such a feature has confused me with lack of decent documentation.  Help
> please.
> 
> I guess simplest is there a HOW-TO for a similar set up?  Two nodes no
> (physical) shared disk HA ability.

Use DRBD to replicate your block device.

Use CLVM on the block device to manage your volumes.  (Go figure.)

Use GFS or OCFS2 on the volumes that are shared (mounted on multiple
machines at the same time).

Use ext3 on the volumes that aren't.

Don't use files for VM images.

What more do you need to know?

-- 
Christopher G. Stach II


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

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