The cluster automatically stops the software upgrade process
if any of the nodes fail to upgrade to the new software level.
In this case, any nodes that have already upgraded to
the new software level are downgraded to the original software level.
If a node fails to restart during this downgrade process, the process
is suspended. The following scenarios can cause the downgrade process
to suspend:
- A node (other than the node that is currently upgrading)
is offline, restarted, or has detected an error condition.
- A node fails to update to the new software level.
- A node is deleted while it is in the process of updating.
You must check the error log to determine the reason for
the failure before you attempt to upgrade the cluster again.