Here is the environment specs we have tested this failure of backing up an Oracle cluster:
Oracle RAC 11
ESXi 5.1
storage provided via NFS
Backup method: Netbackup 7.5 using hot-add method( basically using vADP)
There seems to be a major un-documented bug in ESXi 5.1, as anytime we tried to backup the vmdk files from an Oracle RAC cluster running on Vmware, that particular node is being ejected from the cluster.
It seems that invoking the quiescence feature of the Hot-add method in Netbackup, the Oracle RAC is ejecting that particular node from the cluster.
We were able to replicate this problem every time we try to backup the files stored in a vmdk volume that is provisioned via an NFS datastore.
Has anyone run into this problem before? How did you fix it?
