[Pacemaker] Always Run Clone Resource

Andrew Martin amartin at xes-inc.com
Wed Mar 21 13:51:53 EDT 2012


I have a pacemaker/heartbeat cluster that uses several DRBD primitives. The cluster resources are all colocated and ordered to start after the DRBD primitives . I have configured an ocf:heartbeat:MailTo primitive and clone for notifying me of any changes in the cluster state: 

primitive p_notify ocf:heartbeat:MailTo \ 
params email="myaddr at example.com" \ 
params subject="Alert Pacemaker Change" \ 
op start interval="0" timeout="10" \ 
op stop interval="0" timeout="10" \ 
op monitor interval="10" timeout="10" 
clone cl_notify p_notify 

This MailTo primitive and clone are NOT part of the colocation or order rules for the rest of the resources. Last night all of the resources in the cluster stopped because of a syntax error in the DRBD config. This also stopped the MailTo resources, and thus I was not notified of the error state in the cluster. How can I configure the MailTo resources to be running always, regardless of the state of the rest of the resources in the cluster? 


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.clusterlabs.org/pipermail/pacemaker/attachments/20120321/17446150/attachment-0002.html>

More information about the Pacemaker mailing list