multiple failovers will increase the global cluster name to be too long

[gottumuk]

Each time, we do a global cluster failover, we keep appending date to the global cluster name, which increases the length of the name without any upper bound. This makes it unmanageable