By default, the switchback setting of a resource is intelligent. This means that once the failover occurs for that resource from Server A to Server B, the resource remains on Server B until another failure or until an administrator intelligently switches the resource to another server. Thus the resource continues to run on Server B even after Server A returns to service. Server A now serves as a backup for the resource.
In some situations, it may be desirable for a resource to switch back automatically to the original failed server when that server recovers. SIOS Protection Suite offers an automatic switchback option as an alternative to the normal intelligent switchback behavior described above. This option can be selected for individual resource hierarchies on individual servers. If automatic switchback is selected for a resource hierarchy in the In-Service-Protected (ISP) state running on a given server and that server fails, the resource hierarchy is failed over to a backup system; when the failed server recovers, the hierarchy is automatically switched back to the original server.
- If using data replication (DataKeeper), you must choose intelligent switchback. Automatic switchback is not supported.
- Checks for switchback are only made either when LifeKeeper starts or when a new server is added to the cluster; they are not performed during normal cluster operation.
- SIOS Protection Suite never performs an automatic switchback from a higher priority server to a lower priority server.
If there is a dependency between two resources with different switchback strategies, the switchback strategy of the parent resource takes precedence.
Post your comment on this topic.