9.5 Recovering from a Disaster

Use snapshots of the nodes to recover from a disaster. It is important to take snapshots of each node in the cluster regularly so you do not lose information.

To recover from a disaster:

  1. On a regular basis, take snapshots of the nodes in the cluster.

    1. Power off the working node, then take a snapshot.

      or

      Take a snapshot of the running node including the virtual machine’s memory.

    2. Repeat Step 1.a for each node in the cluster, within a short time.

  2. When a failure happens, restore the master node snapshot first.

  3. Restore the other nodes in the cluster.

Use these steps only for disaster recovery. Never restore one snapshot. Access Gateway for Cloud contains a database that is time-sensitive. Restoring one node only and not the others causes corruption in the appliance.