Cluster: Update a Cluster

If a cluster is to be updated, it is recommended that you always start with the last node.

• If it is a 2-node cluster:
  start the update of Node2 and then repeat the same update process for Node1.

• If it is a 3-node cluster:
  start the update of Node3, then repeat the same update process for Node2 and then for Node1.

Prerequisite:

• It is important that all nodes in the cluster are updated to the same version!

• When a node is updated, the update on the next node should not begin until the cluster is operational again.

2-node cluster:

Start the update of Node2 and then repeat the same update process for Node1.
In a cluster with 2-nodes, Node2 must first be fully operational again after the update before Node1 can be updated.

Updating a 2-node cluster running a version prior to 5.2.2

When updating a 2-node cluster running a version prior to 5.2.2, all nodes may transition to a non-primary state.
In this case, Force into primary must be executed to restore the cluster to its properly functioning state. Refer to Cluster Troubleshooting.

1. Log in to your Next Generation Hardware Appliance.

2. Open the Overview page.

3. All nodes are listed on the right-hand side in the Cluster Overview section.

4. Check for any error messages e.g. a node has not yet been restored yet.

5. Ensure that all nodes are active and connected again.

6. Next switch to the Cluster page.
   To do this, use the tab Cluster in the top menu bar or click on Configure Cluster in the Cluster Overview section.

7. Information about the cluster is also visible on the Cluster page in the Cluster Status section.

   • The warning sign indicates that a node is not functional.

   • Further more a red banner indicates that Cluster is not operational!
     This happens, for example, if a node is still in initialization mode.
     The warning sign disappears as soon as the node is connected.

8. Open the Settings page.

9. Start to update the appliance to the required version.
   See Appliance Update for more information.

10. An EJBCA license is required after a reboot.
    See License Management for more information.

3-node cluster:

The process is the same as for a 2-node cluster.

Start updating Node3, then repeat the same update process for Node2 and then for Node1.

Between each update of a cluster node, it must be monitored that all nodes are operational until the update process is successfully completed.

Verification of Successful Update Completion

The following describes how to verify if an update for a cluster was successful.

After updating all cluster nodes as described above, make sure the cluster connection is established again.

1. Log in to your Next Generation Hardware Appliance and open the Overview page.

2. All nodes are listed on the right-hand side in the Cluster Overview section.

3. Check for any error messages e.g. a node has not yet been restored yet.

4. Ensure that all nodes are active and connected again.

5. Next switch to the Cluster page. To do this, use the tab Cluster in the top menu bar or click on Configure Cluster in the Cluster Overview section.

6. Information about the cluster is also visible on the Cluster page in the Cluster Status section.

   The warning sign indicates that a node is not functional.
   Further more a red banner indicates that Cluster is not operational!
   This happens, for example, if a node is still in initialization mode.
   The warning sign disappears as soon as the node is connected.

7. Use the drop down function in the title bar to switch between nodes.

8. Start with any node and check for the Version of the current Next Generation Hardware Appliance.
   To do this, you can scroll down to the bottom of any page in Webconf.
   The current version number is always visible here.

   All nodes must of course have been updated to the same version.