Rgmanger now checks to see if it has been killed by the OOM killer and if
so, reboots the node.
Resolves: rhbz#488072
[2.0.48-1]
clulog now accepts ‘-’ as the first character in messages.
If expire_time is 0 max_restarts is no longer ignored.
SAP scripts have been updated.
Empty PID files no longer cause resource start failures.
Recovery policy of type restart now works properly when using a resource
based on ra-skelet.sh
startup_wait option has been added to the mysql resource agent.
samba.sh now kills the pid listed in the proper pid file.
Handling of ‘-F’ has been improved to fix issues with rgmanager crashing if
no members of a restricted failover domain are online and rgmanager failing
to correctly restart service is they fail on the first node.
Enabled ability to prioritize services.
It is now possible to cap the number of simultaneious status checks to
prevent load spikes.
Enabling a frozen service no longer fails and leaves the service in a failed
state.
Forking and cloning during status checks has been optimized to reduce load
spikes.
rg_test no longers hangs when running against a cluster due to the removal
of an 8MB memory cap.