I'd highly recommend getting pacemaker 1.0 which (finally!) sorted out the scoring mess that failure_stickiness created. http://clusterlabs.org
oh, and the new version also allows: ptest -L -s which will show you the current scores. On Fri, Mar 20, 2009 at 22:01, adam <tragic_ins...@yahoo.com> wrote: > Hi list- > > Heartbeat newbie here, combed the docs and google and can't figure out what > I'm not doing. If someone could open my eyes to whatever it is I am totally > overlooking I"d really appreciate it. > > Basically can't get resources to respect score calculations, if they are even > being calculated at all. Is there a way to examine what the scores are? > Through crm_verify? Or buried in cib.xml? Dummy_A comes up on it's prefered > node, but after killing heartbeat and bringing it back up twice, the resource > still comes up on the now degraded node. Here is my simple config using the > Dummy OCF. Any insight would be appreciated. > > Thanks in advance. > > <resources> > <primitive class="ocf" id="Dummy_A" provider="heartbeat" type="Dummy"> > <meta_attributes id="dummy-ma"> > <attributes> > <nvpair name="target_role" id="dummystarted" value="started"/> > <nvpair name="resource_stickiness" id="ma-1" value="100"/> > <nvpair name="resource_failure_stickiness" id="ma-2" value="-100"/> > </attributes> > </meta_attributes> > </primitive> > </resources> > <constraints> > <rsc_location id="dummy_on_kilo" rsc="Dummy_A"> > <rule id="kilo_native" score="200"> > <expression id="be_on_kilo" attribute="#uname" operation="eq" > value="kilo"/> > </rule> > <rule id="juliet_backup" score="150"> > <expression id="fail_to_juliet" attribute="#uname" > operation="eq" value="juliet"/> > </rule> > </rsc_location> > </constraints> > > > > > > _______________________________________________ > Linux-HA mailing list > Linux-HA@lists.linux-ha.org > http://lists.linux-ha.org/mailman/listinfo/linux-ha > See also: http://linux-ha.org/ReportingProblems > _______________________________________________ Linux-HA mailing list Linux-HA@lists.linux-ha.org http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
