On 29.05.2012 10:15, Anton Altaparmakov wrote: > On 28 May 2012, at 23:46, Steven Silk wrote: >> I am trying to setup a two node system making NFS highly available.... >> We have run this in the past with heartbeat and drbd. Now we would like to >> use pacemaker and corosync. I have been told not to use lvm, and this must >> be in a unicast environment. >> Is there anything obviously wrong with this plan? I have stumbled around >> and gotten the preliminaries set up and working, but when I start >> configuring the primitives in crm I get tons of errors. > Nothing wrong with this plan at all. We have three systems like that running > here in production without any problems. Though ours are more complicated as > each of our clusters is made up of three nodes so that quorum works. > Just as an example here is our crm configuration which works without any > errors (cerberus and minotaur are the two storage containing nodes and qs2 is > the quorum node which cannot run any of the resources - you can obviously > ignore a lot of our configuration because you only have two nodes): > > node cerberus \ > attributes standby="off" > node minotaur \ > attributes standby="off" > node qs2 \ > attributes standby="off"
You can set qs2 to standby permanently. As long as its online, the quorum will work and the one storage-node that sees the qs2 will be the active part of the cluster. Have fun, Arnold -- Dieses Email wurde elektronisch erstellt und ist ohne handschriftliche Unterschrift gültig.
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Pacemaker mailing list: [email protected] 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
