Promoting a node to master

If a master node fails, you might want to promote a different node to master while you resolve the failure.

About this task

When you are promoting a node to master, ensure that you adhere to the cluster architecture rules. For example, you must specify the supplementary masters in order. You cannot specify tertiary and quaternary masters if there is no secondary master. For a complete list of the cluster configuration rules, see Cluster configuration rules.

Promoting a node to a master falls into two main categories:
  • Promoting a node to a supplementary master - secondary master, tertiary master, or quaternary master.
  • Promoting a node to primary master.

Promoting a node to a supplementary master

Procedure

You can use the local management interface of the primary master to update the cluster configuration and select the supplementary masters. To promote a node to secondary, tertiary, or quaternary master, complete these steps:
  1. Open the Cluster Configuration page from the primary master local management interface.
  2. Go to the General tab.
  3. Change the values in the master fields. That is, Secondary master, Tertiary master, Quaternary master.
  4. Save and deploy the updates.

Promoting a node to primary master when the original primary master is unavailable

About this task

  • Nodes are automatically updated with information for the new primary master. If a node is not reachable by the primary master at the time of promotion, there is a delay of up to 15 minutes from the time that connectivity is restored before the node is notified of the new primary master.
  • If the original primary master is reconnected to the primary master, it is automatically demoted to the role of a normal node.
  • If the network is segregated and two different nodes are promoted to primary master in the different networks, automatic recovery is not possible when the network connectivity is re-established. In this situation, a manual merge of the segregated cluster is required. This step is achieved by removing all nodes from one of the clusters and joining these nodes back into the other cluster. This situation occurs only when both of the following conditions are met:
    • Connectivity in the cluster is lost.
    • The administrator promotes two different nodes to the primary master role while network connectivity is lost.

Procedure

Use the local management interface of the appliance that you are promoting to primary master to update the configuration. You can promote a non-master node or one of the supplementary masters if available. To promote the selected node to primary master, complete these steps:
  1. Access the local management interface of the node that you want to promote to primary master.
  2. Select System > Network Settings > Cluster Configuration.
  3. Select the General tab.
  4. Select Set this appliance as a Primary Master.
  5. Use the available menu to set the Primary master IP address. Select the first management interface of the appliance.
  6. Save and deploy the changes.

Promoting a node to primary master when the original primary master is available

About this task

  • You must deploy any undeployed changes on all master nodes before promotion.
  • Both primary and secondary cluster statuses must be accessible and synchronized.
  • If you have already attempted to promote the secondary node and it is not synchronized, you must break the cluster and rebuild it before following these instructions.

Procedure

To promote the secondary master node, follow these steps:
  1. Access the Local Management Interface (LMI) of the secondary node that you want to promote to primary master.
  2. Select System > Network Settings > Cluster Configuration.
  3. Select the General tab.
  4. Select Set this appliance as a Primary Master. This action automatically populates the 'Primary Master' field with the hostname or IP address of this node.
  5. Clear the following fields from bottom to top:
    • Quaternary Master
    • Tertiary Master
    • Master External Reference Entity
    • Secondary Master
  6. Save and deploy the changes.
    • Wait for the cluster to synchronize.
    • If you are using the internal HVDB, the cluster will have only one copy of the data at this point.
    • The original Primary node (or a different node with AAC/Fed activated) must be made as the Secondary node, so that it serves as the standby for HVDB.
  7. Update the configurations for the Secondary, ERE, and other master nodes as required.