I have setup an 8 node cluster. The cluster has 15 resources. I setup the system such that all 15 resources are distributed on the 7 primary nodes when the cluster starts up. I would like it such that when a node fails, the resources are migrated to the backup node first and then subsequently balancing the resource distribution among the remaining nodes. I'm running HB 2.1.4. It is my understanding that if the score values are equal then the node with the fewest resources will be selected as the target node. During my testing I don't find this to be the case. It seems like the node selected is the first node listed in the crm_mon output:
Node: fs24 (1ff11fab-a613-4335-b302-33dd812bf99b): online Node: fs23 (85a8a44b-9e1e-4aaa-bedf-d023398b5553): online Node: fs22 (bf727f13-62d9-450b-b389-86353994ffe1): online Node: fs21 (4d36cf33-7540-45e5-b6ae-d3d34661de75): online Node: fs20 (11d49119-471f-4ce1-bbed-2603edee32fc): online Node: fs19 (f10a906e-5ad2-4e49-af91-f9ebbfa2b52d): online Node: fs18 (d538b892-ddee-4dac-aaf1-e29dcf85817b): online Node: fs17 (9073d06c-e040-46f9-9a69-9c3e7719333c): online Can anyone confirm the behavior of Heartbeat 2.1.4 for node selection in the case where scores are equal? thanks, kevin _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
