7.5 Recovering from a Failed Node in a Cluster

If one node in a cluster fails, Secure API Manager provides a process where you can redeploy a new node and have it join the system.

NOTE:The following steps apply only to non-database nodes. When you are redeploying a non-database failed node, you do not have to power down and restart the database nodes. For information about restoring a failed database node, see Restoring a Failed Database Service.

To remove and redeploy a failed node:

  1. Ensure that you have completely removed the failed node from VMware.

  2. Check the SYSTEM tab in the Deployment Manager to ensure that the remaining node or nodes in the cluster are up and communicating with the Database Service.

  3. Redeploy a new appliance with the same IP address and DNS name as the failed node had before the failure. For more information, see Deploying the Secure API Manager Appliances.

  4. Log in to the appliance management console on the new appliance as root with the new password you created when you redeployed the appliance.

    https://ip-address-or-dns-name-appliance:9443

  5. Click Deployment Manager.

  6. Select Join.

  7. Specify the DNS name for the Database Service component and specify the database user name and password you created when you deployed your system.

  8. Select Join.

  9. Approve the certificate that the Deployment Manager displays or import a trusted root certificate for this appliance.

  10. Click Go To Deployment.

  11. Access the appropriate deployment page for this component, then specify the required information for this appliance.

  12. On the last page, click Save.

  13. Select Save configuration and deploy only this appliance.

  14. On the STATUS tab, watch this node join the system. The Deployment Manager adds the configuration information for this new node to each component in the system.