The IBM FlashSystem supports the capability to protect from disaster scenarios through the disaster recovery capability provided by the policy-based asynchronous replication feature. In order to verify the viability of the disaster recovery configuration, the FlashSystem provides storage administrators the capability to simulate a disaster recovery failover through the disaster recovery test feature. A disaster recovery test can be initiated on the recovery volume group, which provides read and write access to the most recent checkpoint version of the volumes located in the recovery volume group. Host systems at the recovery site that are mapped to those volumes can then perform read and write operations to the volumes to simulate a disaster failover without affecting the ongoing asynchronous replication from the production to the recovery site. Once the disaster recovery test completes, replication continues as before and recovery volumes return to being synchronized with the production volumes. For more detail on the disaster recovery test feature, refer to the following blog post.
The IBM FlashSystem 25Q4 (9.1.1) release adds the ability to start, monitor and stop a disaster recovery test directly from the GUI for disaster recovery configurations using the async-dr replication policy type. A disaster recovery test can be monitored from either the recovery volume group policies view or the production volume group policies view. However, the user must be logged on to the recovery site and viewing the recovery volume group replication policy in order to start or stop a disaster recovery test.
To start a disaster recovery test and enable temporary read and write access to the recovery volumes, click the 'Start recovery test' button on the recovery volume group replication policy tile.
The disaster recovery test status can be viewed on the same recovery volume group tile. Ensure that applications on mapped recovery site hosts that are utilizing the recovery volumes function as expected.
Once proper validation of the disaster recovery environment is performed, the disaster recovery test can be stopped. The storage administrator can choose to stop the test and make the recovery volumes inaccessible to hosts at the recovery site, or they can choose to stop replication and keep independent access for the volumes at the recovery site.