Hi,

I've got a heartbeat 2.1.2 cluster with two nodes and two drbd resources
set up as described on
http://wiki.linux-ha.org/DRBD/HowTov2?highlight=%28drbd%29:

Drbd0 prefers to run as master on node1, drbd1 prefers to run as master
on node2

There's a resource group "web" that uses drbd1
<rsc_order id="drbd_web_before_web" from="web" action="start"
to="drbd_web" to_action="promote"/>

Can I just add 

<rsc_order id="drbd_cache_before_web" from="web" action="start"
to="drbd_cache" to_action="promote"/>

To make the web group require both drbd resources to be in master state?
If not: how hould I specify this requirement correnctly?

Next, I need to make sure that drbd2 always runs as master on the same
node as drbd1. would this colocation rule work?

<rsc_colocation id="cache_on_drbd_web" to="drbd_web" to_role="master"
from="drbd_cache" from_role="master" score="infinity"/>

Thanks for any hints you can give!
_______________________________________________
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