Dejan Muhamedagic <dejanmm@...> writes: > On Fri, May 27, 2011 at 08:21:08PM +0000, veghead wrote: > > 1) Is there a way to force crm to accept my configuration request > > ~before~ starting the second node? > > No before the DC is elected. There are two settings: dc-deadtime > and startup-fencing which can reduce the time for DC election. > Note that disabling startup fencing is not recommended. But I > don't know what's your use case. YMMV.
Well, I'm probably not quite the typical use case. We're using Amazon EC2 to setup and tear down testing environments. I have automated the entire process except for setting up Pacemaker. Beyond testing environments, I'd like to automate Pacemaker setup to cover the scenario where all nodes in a Pacemaker cluster crash and the entire configuration is lost. Obviously, once one node is running, setting up additional nodes becomes easy. It's just the bootstrap phase that's a challenge to automate. > > 2) Is there a way to tell Pacemaker to ignore quorum requirements > > ~before~ starting additional nodes? > > > > 2) Is there an alternate way to configure Pacemaker? > > Yes, you can modify the CIB _before_ starting pacemaker. Sth > like: > > CIB_file=/var/lib/heartbeat/crm/cib.xml crm configure ... > > But in that case you need to remove cib.xml.sig. Then you have to > make sure that pacemaker starts first on this node. Consider this > only if everything else fails. I'll give that a shot. Thanks. -S _______________________________________________ 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://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker