stonith-timeout doesn't work as expected in a RHEL 6 or 7 High Availability cluster with pacemaker

Issue

• I've set stonith-timeout as an attribute on my stonith device, but I'm still seeing timeouts that are shorter than the value I set
• stonith-timeout doesn't seem to work on a per-device basis like the manpages and docs say it should
• Why doesn't stonith-timeout apply correctly to my cluster's stonith devices? I have to set the cluster property stonith-timeout for all of them.
• I've set the stonith-timeout property to a higher value, but monitor actions are timing out
• Does stonith-timeout apply to monitor, list, and status actions? Or just to fencing a node (i.e., reboot).
• My cluster logs errors from my stonith device regarding a parse error from stonith-timeout:

Aug 31 08:23:39 node1 stonith-ng[2008]:  warning: log_operation: vmfence1:6167 [ Parse error: Ignoring unknown option 'stonith-timeout=120' ]