Fuel for OpenStack

crm doesn't show actual state after resource refresh

Bug #1269449 reported by Aleksandr Shaposhnikov
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Fuel for OpenStack
Invalid
Undecided
Aleksandr Shaposhnikov
Fuel for OpenStack 4.1

Bug Description

Confirmed behavior of corosync that described here:
http://bugs.clusterlabs.org/show_bug.cgi?id=5150

I've seen this behavior with some HA services that managed by corosync and pacemaker on fuel installed cloud.

Revision history for this message
Vladimir Kuklin (vkuklin) wrote :

Alex, please provide steps you are doing. What is expected behaviour and what is the actual behaviour.

Changed in fuel:
status: New → Incomplete
milestone: none → 4.1
Revision history for this message
Dmitry Borodaenko (angdraug) wrote :

Aleksandr, please reopen and provide more info if you still have this problem.

Changed in fuel:
assignee: nobody → Aleksandr Shaposhnikov (alashai8)
status: Incomplete → Invalid
Revision history for this message
Andrew Beekhof (1nbr3w) wrote :

'crm_resource --reprobe' is used to tell the cluster to redetect the state of the named resource.

It is supposed to wait for replies from each of the active nodes before exiting, however there was a bug that prevented those replies from being received - so the command would hang before eventually timing out.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.
Subscribing...

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.