Sometimes can be interesting to know which is the master node in a cluster, in my case, this is needed becouse the machines of cluster have different hardware, being one more powerfull than others because of this I prefer like the master node the more powerfull machine (more CPUs and memory).
The first rule (and logical) is
The default master is always the first node that is started in the cluster.
The problem is when we have had some reboots and the timeline is not very clearly. In this case you have two ways;
By locating automatic OCR backup
It returned some rows, the more recent backup perform by one of nodes is the master.
By searching the string “MASTER” crsd.l* on all nodes:
grep “MASTER” $ORA_CRS_HOME/log/$HOST/crsd/crsd.l*
HTH – Antonio NAVARRO