Maintenance Operations Overview
In some cases, the HX nodes may need to go through a maintenance window. For example, this may be needed to replace disks, or make modifications to the UCS Service Profiles that require a reboot. The HX maintenance can be performed with no disruption to the HyperFlex workloads.
The maintenance operation can be performed on a single node, multiple nodes, or all nodes in the cluster. When applying the change to all nodes in the cluster, the changes are applied sequentially through a rolling maintenance operation.
In order to perform the maintenance operations, follow the process outlined.
Verify the pre-upgrade validations listed in Cisco HyperFlex Upgrade Guidelines section.
Select a HX node and put it in maintenance mode.
Perform the maintenance operation
Exit maintenance mode
Repeat this process for all hosts in the cluster in a sequence (if applicable: when the maintenance operation must be performed on each node in the cluster