When Master Node, Data Node and Witness Node have been deployed, the Continuous Availability feature can be easily enabled in vRealize Operations Manager.
To obtain best performance and a reliable vROps infrastructure, additional Data Nodes should be deployed in your environment.
Blog Series
vRealize Operations Manager 8.3 enable Continuous Availability - pt.1
vRealize Operations Manager 8.3 enable Continuous Availability - pt.2
vRealize Operations Manager 8.3 enable Continuous Availability - pt.3
Deploy additional Data Nodes
For production environments is recommended to deploy two pairs of Data Nodes. Two additional nodes are deployed to fulfill the best practices by following the same procedure covered in part 2.
One Data Node is deployed in the primary site where the Master Node is located.
The second Data Node is deployed in the secondary site in order to have an even number of Data Nodes to create the pair.
Click Finishing Adding New Node(s) button to add the two new nodes to the vROps cluster.
After some minutes all new nodes are Online and Running.
Enable the Continuous Availability
Click Enable CA button to proceed with Continuous Availability configuration.
By default the Master Node is placed in Fault Domain 1. Drag and drop other Data Nodes to Fault Domains 1 and 2 in order to create some pairs. Data Nodes placed in Fault Domain 1 must belong to same physical location as well as Data Nodes for Fault Domain 2.
When done, click OK.
Continuous Availability feature is being enabled.
During the CA enabling procedure, the vROps cluster will go offline.
After some minutes, the cluster goes back online and Continuous Availability feature is enabled.
The vROps cluster is now taking benefit of the CA feature ensuring high availability of the cluster even in the event of a complete failure of one Fault Domain.
Shutdown and startup the vROps cluster
Log into the Administration GUI at the address https://<IP-vrops-master>/admin then click on Take The Cluster Offline button. This will start the graceful shutdown of services running in the cluster.
Enter a Reason and click OK to confirm.
The vROps cluster is going offline.
Wait until the cluster is reported as offline before proceeding with the appliances shutdown operation.
Access the vCenter and shutdown the nodes following this order:
- Remote Collector nodes if installed
- Data Nodes
- Master Replica Node
- Master Node
Startup the vROps cluster
To startup the cluster just follow the inverse order:
- Primary node
- Primary replica node
- Data nodes
- Remote Collector nodes
Following the correct shutdown and startup order is very important to avoid negative effects on the vRealize Operations Manager cluster.
