Configuration Considerations and Limitations for VPLS and VLL MCT

Draft comment: chturner
<<<SME Please review and advise regarding RSVP and BGP information. Does this remain, or is it obsolete?

Hitless ISSU is not supported. Before starting ISSU, issue the shutdown clients command on the PE where the ISSU is planned to gracefully move the traffic to the MCT peer. Similarly, use the force-standby or no deploy command for the PW CCEP before starting ISSU.
Statistics are not supported.
For VPLS MCT, consider the following:
- Configuring a cluster peer as a BD peer impacts data traffic.
- You can use the shutdown clients command to shutdown traffic on one node before a software upgrade. After you issue this command, all PWs are put into standby mode.
- shutdown clients brings down all CCEP interfaces. Other nodes attempt client-isolation logic, and you may see the Strict behavior.
- Logical-interface shutdown brings down the admin state of the CCEP LIF interface if the parent port is an MCT client interface and does not trigger a DF re-election.
- Protection for a PW link failure is not supported. Active forwarding paths does not occur between the nodes.
For VLL MCT, consider the following:
- Cross-connect is used instead of VSI.
- MAC learning is not required.
- BUM traffic handling is not required.
- The MCT role does not depends upon endpoint as well as the PW redundancy status.