Testing MetroCluster configuration during a switch failure

To ensure that an FC switch is configured correctly and the failure of the switch does not cause a node failover, you must manually test the FC switch used in a MetroCluster configuration.

You must have collected all the required information for the components used in a MetroCluster configuration.

A switch failure must not initiate a disaster recovery and therefore during a switch failure, you must be able to access the data in both the nodes in a MetroCluster configuration.

  1. Fail one FC switch by powering it off.
  2. Verify that the data on both nodes is still accessible.
  3. Verify that takeover capability is enabled by entering the following command:cf status
    Note: After powering off the switch, the takeover capability might be disabled briefly. It is subsequently reenabled automatically.
  4. Power on the switch.
  5. Verify that the system returns to healthy state and all the disks are available.