A two-VM Oracle RAC cluster was recently configured with Linux Red Hat using a Nutanix Volume Group. At the time Of performing the failover tests Of simulating a failure in each Of the three Nutanix cluster nodes, the cluster administrator noticed that tile virtual machines Of the RAC cluster do not come up and the failover fails.
The administrator must to follow the best practices from Nutanix to construct the RAC cluster using AHV.
What should the administrator Veri'/ in the Steps performed to configure the RAC cluster to prevent this situation from occurring again?
https://www.nutanix.com/architecture
Nutanix volume groupsare collections of vDisks that can be exposed to virtual machines or physical hosts using iSCSI protocol. Nutanix volume groups have a shared attribute that allows multiple VMs to access them concurrently.
In the step for attach the volume group to the RAC VMs, he must create it with an attribute of shared = true. This is because if the volume group is not created with shared = true, then only one VM can access it at a time, which would prevent Oracle RAC from working properly.
Currently there are no comments in this discussion, be the first to comment!