On 06/10/12 22:21, Arnold Krille wrote: > On 10.06.2012 21:48, Florian Haas wrote: >> However, why do you want automatic failback? If your cluster nodes are >> interchangeable in terms of performance, you shouldn't need to care >> which node is the master. In other words the concept of having a >> "preferred master" is normally moot in well-designed clusters. > > One of the things on our road-map is providing some of our customers > with a two-node cluster that in normal operations does load-balancing > (simply by having two vm each for half of the samba-resources based on > two (or more) drbd-resources). Then normally the vms would run each on > one node and only in case of an error both run on one node. Then you > want the resources to fall back to distributed location when all is well > again. > I tried this once with an anti-colocation but I don't know if that is > easy to set up also with drbd-masters. That probably requires a well > adjusted value for the anti-colocation to get it between the master- and > slave-scores of drbd-master...
Yes. I would call that pretty tricky. :) Cheers, Florian -- Need help with High Availability? http://www.hastexo.com/now _______________________________________________ Pacemaker mailing list: Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://bugs.clusterlabs.org
