Automatic Switchover (ASO)

IMPORTANT: Check the latest release notes before you attempt to use the Peer Persistence feature.

Automatic Switchover (ASO) is an optional, but highly recommended, feature that can be used with synchronous replication to enable automatic failure recovery. When ASO is used in conjunction with synchronous replication, the feature is known as HPE Peer Persistence.

The HPE Peer Persistence feature enables synchronous replication with ASO, which allows your arrays to recover automatically and non-disruptively from storage based failures.

NOTE: The HPE Peer Persistence feature is designed to enable automatic recovery failure in the event of storage failures. The feature will not prevent site-to-site failures that result from host level failures.

ASO currently works on array groups that consist of two arrays. When ASO is enabled and a switchover occurs, the partner that serves I/O for a volume collection is switched, which non-disruptively reverses the direction of the synchronous replication.

ASO facilitates automatic failure recovery through the use of a Witness daemon that can be run on an independent host (or in a separate VM) that can communicate with the group leader and the backup group leader.

NOTE: The Witness does not need to be on the same subnet as the array group, but its IP address must be routable from the array group’s management network. The Witness can be installed in a different data center or even in a cloud environment as long as the round-trip time (RTT) between the Witness and the two arrays is 250 milliseconds or less. Multiple methods exist to deploy the Peer Persistence Witness. If the Witness is installed in a different data center or cloud environment and HPE Storage Support is needed to resolve an issue with the Witness, consider that if HPE Storage Support cannot access the data center, cloud environment, or infrastructure, troubleshooting may be limited.

When the Witness detects that an array is unavailable, the ASO process performs a handover of affected synchronous replication volume collections and seamlessly maintains the availability of group services and of synchronously replicated volumes.

To enable ASO, you must perform one of the following procedures:
  • Install and configure the Witness software on a CentOS V7.9.x. client.
  • Run the Witness in a separate VMware based Virtual Machine.
After the Witness is configured, you can then use the array OS GUI or CLI to set up the Witness and enable Automatic Switchover between the arrays in the group.
NOTE: The ASO check box is selected by default; however, ASO is not actually enabled unless the Witness is configured. For example, you might go to AdministrationAvailability in the array OS GUI and see that Enable is checked for Automatic Switchover, but the fields for the Witness are blank. In this case, ASO is not enabled. You must both configure the Witness and check the Enable box for ASO to be configured.
Tip: You can use the following group command to enable or disable ASO from the CLI.
group --edit --auto_switchover_enabled {yes|no}

For information about enabling, deploying, and testing the HPE Peer Persistence feature, including possible ASO failure scenarios, refer to HPE Peer Persistence Deployment Considerations in the HPE InfoSight documentation portal.