On 13/04/17 05:07 PM, neeraj ch wrote: > Hi, > > I have three node cluster set up. My corosync config is as follows. > > compatibility: whitetank > totem { > version: 2 > secauth: on > threads: 0 > interface { > member{ > memberaddr: ip > } > member{ > memberaddr:ip > } > member{ > memberaddr:ip > } > bindnetaddr:172.31.31.12 > ringnumber: 0 > mcastport: 5405 > ttl: 1 > } > transport: udpu > } > logging { > fileline: off > to_stderr:yes > to_logfile: yes > to_syslog:yes > logfile:/var/log/cluster/corosync.log > debug:off > timestamp: on > logger_subsys { > subsys: AMF > debug:off > } > } > service { > name: pacemaker > ver: 1 > } > amf { > mode: disabled > } > > > > I went on to partition the master, the rest of the cluster considered > the master down and elected a new master. > > But the corosync on the partitoned node didnt realize loss of quoram and > consequently pacemaker didnt shut down the resources. Giving me a > multi-master scenario. > > I have tried adding , quorum tool as votequoram and it seems to give the > same result. > > Why isn't corosync not detecting this scenario ? any ideas. > > I used IP tables to block all UDP traffic in and out of the node . I > didn't block traffic on lo. > > any help would be appreciated.
On EL6, cman is the quorum provider. You need to have a skeleton cluster.conf setup with it's fencing passed over to pacemaker via fence_pcmk. It you use pcs/pcsd, it will configure everything for you. Do NOT configure corosync directly; You need to configure it inside cman itself. Reset corosync.conf back to defaults. Reference; http://clusterlabs.org/doc/en-US/Pacemaker/1.1-plugin/html-single/Clusters_from_Scratch/index.html -- Digimer Papers and Projects: https://alteeve.com/w/ "I am, somehow, less interested in the weight and convolutions of Einstein’s brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshops." - Stephen Jay Gould _______________________________________________ Users mailing list: Users@clusterlabs.org http://lists.clusterlabs.org/mailman/listinfo/users Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://bugs.clusterlabs.org