A single node in a four-node MetroCluster can be taken down for maintenance by performing an HA takeover of that node. MetroCluster provides site-level disaster recovery, but routine maintenance of one controller is handled through the local HA pair so that the partner continues serving the node's storage and data remains online. A MetroCluster switchover is a site-level operation and would be excessive for planned maintenance of one node. Disabling HA would remove the very protection required to keep data online. Aggregate relocation of CFO aggregates alone is not the correct maintenance workflow because the HA takeover process handles controller ownership and service continuity for the node. The support engineer must choose the least disruptive protection mechanism that matches the failure or maintenance domain. For a single controller event inside a four-node MetroCluster, local HA takeover is the correct action. References/topics: MetroCluster maintenance, HA takeover and giveback, controller service operations, site-level switchover, and nondisruptive availability design.
Contribute your Thoughts:
Chosen Answer:
This is a voting comment (?). You can switch to a simple comment. It is better to Upvote an existing comment if you don't have anything to add.
Submit