2012/12/11 Arnold Krille <[email protected]> > > > Better discard the anti-colocation and use two positive location > constraints. And when that value is higher then your stickiness, > services will fall back to their original node when all is well. When > the value is below the stickiness, your services will stay where they > are even as the second node comes back online. >
Sure, but in practise what are the rules you advise to write ? can you give an example ? I thought of the method of the MS resource because it's easy to force a resource to stick to a master node and to another resource to stick to the slave node. But we could also imagine a fake MS (not necessarily a drbd one), I would call it the MS-Guide. In fact the rules : colocation res1-ON-slave inf: nfs-client1 ms-guide:Slave colocation res2-ON-master inf: nfs-client1 ms-guide:Master really work. and typing : crm resource move res1 forces the ms-guide to be promoted the node res1 resides on, then res1 starts up on the other node, and in the meantime res2 replaces it. This way res1 and res2 never run on the same node. Of course, we are not on the scenario of a master/spare pair of servers, but another paradygm. _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
