Hi,

I'm trying to setup a simple drbd cluster using Pacemaker and the ocf RA
provided by the drbd team.

When the master-slave object is created, both drbd resources (on both
nodes) start as slaves. This is the expected behavior: when using plain
drbd (with no CRM involved), both nodes start as slave and syncing only
starts when one node is manually promoted to master.

However, since I use only the CRM/RA for drbd management (except for
editing drbd.conf and creating device metadata), I don't know of any
simple way to promote one resource instance to master.

All the docs/examples that I have seen rely on resource dependency and
implicit promotion of one drbd resource instance once the dependant
resource is started/migrated to a specific node.

Is there any way to manually promote a (drbd) resource instance to
master?

TIA,

Radu Rendec


_______________________________________________
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