Simulating a disaster recovery in a MetroCluster configuration with array LUNs

Testing for proper setup of a MetroCluster configuration with array LUNs includes simulating a disaster recovery. Simulating a site failure and recovery involves disabling and degrading mirrors.

You should never simulate disaster site failure and recovery in production environments without prior planning and downtime.
  1. Disable the HA interconnect between the Data ONTAP systems.
  2. Power down one of the systems. The system you power down simulates the site of the disaster.
  3. On the surviving system, complete the following steps:
    1. Enter the following command to activate a forced, manual takeover: cf forcetakeover -d
    2. Enter the following command to validate that the forced takeover has occurred: cf status
    3. Enter the following commands to validate the status of the aggregates and of the volumes that they include: aggr statusvol status The display should show an online status for aggregates and volumes, but mirrors, which are disabled, should be displayed as “degraded.”
  4. Reconnect the HA interconnect.
  5. On the surviving system, enter the following commands:
    1. Enter the following command to rejoin the aggregates: aggr mirror surviving_aggregate -v victim_aggregate
      Note: Data ONTAP uses parentheses to indicate the degraded aggregate. For example, aggr0_b indicates the aggregate that is not degraded, and aggr0_b(1) is degraded.
    2. Enter the following command to validate that forced takeover has occurred: cf status
    3. Enter the following command to validate aggregate status: aggr status The display should show an online and mirrored status for aggregates.
    4. Enter the following command to validate volume status: vol status The display should show an online and mirrored status for volumes.
  6. Power up the system that you powered down to simulate the disaster.
  7. On the surviving system, enter the following command to reactivate the site that simulated the disaster: cf giveback
  8. To validate that the HA pair, aggregates, and mirrors are online and operational, enter the following commands at both sites: cf status aggr status vol status The display should indicate that the HA pair is enabled, and that aggregates and volumes are online and mirrored.