On Mon, 2018-11-12 at 10:47 +0100, John Gogu wrote: > Hello guys, > I need here an advice in regards of the following situation: > > Online: [Node1 Node2] > Full list of resources: > Resource1 (lsb: Resource1): Started Node1 > Resource2 (lsb: Resource2): Started Node2 > > > In case of Resource1@ Node1 is failing and is moved to Node2 / or the > Node1 is failing itself, I want to stop the Resource2@ Node2 , only > XX seconds after successfully start of Resource1@ Node2 ( the > scenario is needed to speed up application recovery: Resource1 needs > to pick up some data from Resource2) > > Online: [Node1 Node2] > Full list of resources: > Resource1 (lsb: Resource1): Started Node2 > Resource2 (lsb: Resource2): Stopped >> stop > is triggered only XX seconds after successfully start of Resource1@ > Node2 > > > Many thanks in advance for your ideas, > BR, > JohnG
There's no out-of-the-box solution for this, but you might be able to get there via node attribute rules. Rather than use a fixed 10 seconds, which sounds like it might be subject to race conditions, is there a way to detect when resource1 has finished picking up the data? If so, you could set a node attribute at that point, with a rule saying that resource2 can no longer run when that attribute is set. -- Ken Gaillot <kgail...@redhat.com> _______________________________________________ Users mailing list: Users@clusterlabs.org https://lists.clusterlabs.org/mailman/listinfo/users Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://bugs.clusterlabs.org