Current (default of 15min) value for the cluster-recheck-interval is in-optimal.
OpenStack HA guide [0] mentions that:
"Pacemaker uses an event-driven approach to cluster state processing. However, certain Pacemaker actions occur at a configurable interval, cluster-recheck-interval, which defaults to 15 minutes. It is usually prudent to reduce this to a shorter interval, such as 5 or 3 minutes."
Clusterlabs pacemaker guide [1] mentions that:
"If you rely on time based rules, it is essential that you set the cluster-recheck-interval option."
There is also interesting blogpost [2] about the cluster-recheck-interval and the failure-timeout correlation.
[0] http://docs.openstack.org/high-availability-guide/content/_setting_basic_cluster_properties.html
[1] http://clusterlabs.org/doc/en-US/Pacemaker/1.1/html/Pacemaker_Explained/s-failure-migration.html#ftn.id718985
[2] http://blog.kennyrasschaert.be/blog/2013/12/18/pacemaker-high-failability/
Given that, and bearing in mind that Fuel configures the failure-timeout values for resources in the [30; 180] sec range,
the cluster-recheck-interval should be changed to the value which would be near to the 180 sec, like 190 or so.
That should make the pacemaker to process some events for defined resources much faster, up to x5 times basically.
Fix proposed to branch: master
Review: https:/