Adding Another Member to an Existing Cluster

Important - Schedule a full maintenance window to perform this procedure.

1. Install the Security Gateway you plan to add to the existing cluster as another Cluster Member. See the R80.30 Installation and Upgrade Guide.
2. On the Security Gateway computer you plan to add to the existing cluster:
   1. Configure or change the IP addresses on the applicable interfaces to match the cluster topology.
   2. Configure or change the applicable static routes to match the cluster topology.
   3. Connect to the command line.
   4. Log in to Gaia Clish or Expert mode.
   5. Run:

      cpconfig

   6. Select Enable cluster membership for this gateway > enter y to confirm.
   7. Reboot the Security Gateway.
3. In SmartConsole:
   1. Open the existing cluster object.
   2. In the Cluster Members page:
      • If the object of the Security Gateway you plan to add to the existing cluster is not defined yet, click Add > New Cluster Member
      • If the object of the Security Gateway you plan to add to the existing cluster is already defined, click Add > Add Existing Gateway

      Follow the instructions on the screen to configure this new Cluster Member.

   3. In the Network Management page:
      • Make sure all interfaces are defined correctly.
      • Make sure all IP addresses are defined correctly.
   4. Click OK.
   5. Install the Access Control Policy on the cluster object.
4. On each Cluster Member (existing and the newly added):
   1. Connect to the command line.
   2. Log in to Gaia Clish or Expert mode.
   3. Make sure all Cluster Members detect each other and agree on their cluster states. Run:

      cphaprob state

5. If Cluster Members do not detect each other, or do not agree on their cluster states:
   1. Connect to the command line on each Cluster Member (existing and the newly added).
   2. Log in to Gaia Clish or Expert mode.
   3. Restart the clustering on each Cluster Member.

      Important - This temporarily causes the Cluster Member not to be a part of the cluster. As a result, cluster failover can occur.

      Run:

      cphastop

      cphastart

   4. Make sure all Cluster Members detect each other and agree on their cluster states. Run:

      cphaprob state