A relatively simple looking process, but at times, it might make you wonder in which order I should be doing this. Lately while talking with a customer, I realized that in vRO documentation, hints of this process are given but not really in the form of step by step instructions. I then went ahead and tested following steps. I am sure these would be useful to people who might be looking at changing their vRO active-active cluster to active-standby.
1) Take a full backup of vRO Db if you are using External DB, if you are using embedded vPostgresDB as vRO DB then skip to step2
2) Take snapshot of all the vRO nodes.
3) Go into vRO control center of vRO node1, set the number of active nodes to 1. (Under vRo Control center-> Orchestrator Cluster Management -> Number of active nodes. )
4) Repeat step 3) for all the remaining vRO nodes in this cluster.
5) Restart vRO node1
6) Repeat step 5) for all the remaining vRO nodes in this cluster (one by one)
At the end of this, you will see in ‘vRO Control center -> Cluster Management’ that you have now only 1 node as Active and rest of those are stand by. Which node becomes active depends upon the order of restart of the nodes. (This can be seen from control center of any nodes in cluster)
Please note: Do not forget to remove snapshots taken in step 2) when you are done with this process.
It is a very useful blog. A very rare site to share troubleshooting on vRealize.
Thanks.