On 9/30/07, Ivan <[EMAIL PROTECTED]> wrote:
> Hi List,
>
> I am a bit new to HA but read what was available and now I have a well
> working Xen cluster based on SLES10SP1.
>
> Background:
> My Xen domUs are running off a san, partitions are LVM2 on top of EVMS
> CSM. I cannot have OCFS2 image store for my domUs due to the nature of
> them so must use partitions instead if file images.
>
> I have resource stickiness set, and one of my nodes (2 node at this
> stage) is the preferred one so I have a score 100 location constraint
> for all domUs. (the other node is tester node and failover only)
>
> Somehow I cocked up something last week and ended up having 2 instances
> of one of my domUs. I must admit that I was playing with crm_resource
> migration when it happened.
>
> Question:
> What is the best way of specifying that only 1 instance of a certain
> resource can run at a time, it can only be started on the other if the
> running node is having issues or something.

could you attach your config pls?  without it i'm not 100% sure i know
what your asking.

although i will say that if you are cloning a group that there is a
known problem with this (which has since been fixed upstream).

> I love the fact that when I force migrate a resource to the other node
> it sets automatically up a -INFINITY to the other node doing exactly
> what I want.
>
> I tried setting up the preferred node with INFINITY and the other with
> -INFINITY but then the migration didn't work.

odd, it probably should.  can you include logs?

> Shall I just reduce the
> score to smaller value for instance 9000/-9000 or something?
> I use HA 2.0.8 with crm only.
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to