[ClusterProtector] How to Recover from a Failed Cluster

Follow

Summary

This Knowledgebase article describes how to recover and restore application services to the Production Site after a Cluster failure resulted in Neverfail ClusterProtector failing over and providing application services from the Disaster Recovery Site.


More Information

The following procedure restores the correct running configuration in which the application runs within the Cluster and the Neverfail Primary is active on one of the Cluster nodes. See also Knowledgebase Article #1456 Overview - The Operational Behavior of Neverfail ClusterProtector , which describes the typical operation of the Neverfail ClusterProtector solution.

NOTE

It is outside the scope of this Knowledgebase article to describe how to resolve the actual problems that caused the Cluster to fail.


Procedure

Perform the following procedure to recover from the failure.

  1. Disconnect and disable the Neverfail Channel and Principal (Public) NICs on the Cluster Node that is to be brought online.
  2. Restart the failed Cluster Node if necessary.
  3. Run the Neverfail Server Configuration wizard.
  4. Set and verify that the Cluster Node is set to Primary (passive).
  5. Click Finish to accept the changes.
  6. Start Neverfail Heartbeat on the Cluster Node.
  7. Reconnect and enable the Principal (Public) NIC.
  8. Start the cluster service on the Cluster Node.
  9. Resolve the Cluster problem.
  10. Take offline the resources that represent the application services.
  11. Reconnect and enable the Neverfail Channel NIC and bring online the Neverfail Channel resource on the Cluster Node.
  12. Ensure that the Neverfail Secondary server is active.
  13. Wait until the Secondary (active) and the Primary (passive) servers are synchronized.
  14. Perform a switchover.

    Additional Step for Exchange ONLY :

  15. If the Clustered Mailbox Server resource was allocated to the DR Cluster then the CMS resource must be moved back to the Production Cluster with the following PowerShell command; this will ensure PowerShell cmdlets return correct information

    Set-MailboxServer -Identity:<CMSServerName> -RedundantMachines:{< ProductionActiveNodeName>,<ProductionPassiveNodeName>}


Applies To

Neverfail ClusterProtector, all versions.


Related Information

None.

KBID-1457

0 out of 0 found this helpful

Comments

0 comments

Please sign in to leave a comment.