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


[Xen-users] Best practices iSCSI / redundancy

To: <xen-users@xxxxxxxxxxxxxxxxxxx>
Subject: [Xen-users] Best practices iSCSI / redundancy
From: "netz-haut - stephan seitz" <s.seitz@xxxxxxxxxxxx>
Date: Fri, 4 Sep 2009 12:20:44 +0200
Delivery-date: Fri, 04 Sep 2009 03:23:18 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
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/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
Thread-index: AcotSVzEmwdSRpk8Ed7GhwBQBPU9LA==
Thread-topic: Best practices iSCSI / redundancy
Hi there,

I'ld like to ask if someone has some best practices at hand for following 

We're currently running some Xen Hosts each equipped with hardware raid with 8 
to 12 disks and 4 nics dedicated for iSCSI.
Currently, there's no real installation consistence. Some domUs (all PV) are 
running with phy-backend over local LVM. Some others are using iSCSI where each 
domU has its own LUN.

We thought about adding some redundancy by building soft raid-1 over iSCSI LUN 
and local LVM. Well, this will affect fast and easy migration, but that's not a 
primary topic.

We ended up with two different possible configurations:
a) Having an iSCSI LUN for each physical Server with the size of one big 
physical partition. Adding LUN and physical partition into a raid-1 and putting 
LVM on top of this raidset. The virtual Server would use a LV on top of this 
big raid-1 PV/VG.
b) Setting a big VG on top of the physical partition, and having an iSCSI LUN 
and an LV for each virtual Server. Adding each LUN and each LV into an 
independent raid-1. The virtual Server would use a raid-1 MD build from one 
iSCSI and one local part.

Maybe some other scenarios are also possible, any thoughts are very welcome.



Mit freundlichen Gruessen

Stephan Seitz
Senior System Administrator

*netz-haut* e.K.
multimediale kommunikation

zweierweg 22
97074 würzburg

fon: +49 931 2876247
fax: +49 931 2876248

web: http://www.netz-haut.de/

registriergericht: amtsgericht würzburg, hra 5054

Xen-users mailing list

<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-users] Best practices iSCSI / redundancy, netz-haut - stephan seitz <=