WAAS Central Manager fault tolerance procedures

[Clifford McGlamry]

If a central manager is being run in a warm standby mode (one primary, one in standby), and it becomes necessary to promote the standyb to primary, my understanding is that the system then restarts itself.

Once this happens, does the primary change its configuration so that it takes over the IP address of the formerly primary central manager? If not, I'm curious how all the devices out there know to talk to it.

This isn't addressed in any of the documentation I've seen (interaction with the rest of the WAAS devices to promotion of standby central manager to primary). If it doesn't change its IP address, it would seem you'd have to log into every device and repoint them to the new primary central manager.

Would appreciate responses from anyone having experience with this. I can't find clear documentation on this anywhere.

[tblancha]

When the primary is still primary and the secondary CM registers to it, then the primary tells the application accelerators about the secondary address. It also tells the secondary all config changes. When the primary is demoted or removed (one or the other is required) and the secondary is manually promoted, then the remote WAE's know to go to the secondary address. Note that even if the secondary is promoted to primary and the remotes switch to it, the show cms info will still show the primary's IP address. So, the accelerators will only allow one IP address of the primary and this will never change. They are told about the secondary automatically. There should never be 2 primary devices since it is a manual process.