We have implemented WAAS in an enterprise network. At HQ, we have 2 WAEs managed by a Central Manager and all branches are running WAE modules on 2800 series routers. All branch WAEs are registered on Central Manager at HQ. All is working well.
Now, when we are trying to implement same WAAS topology (2 WAEs with 1 CM) at DR site so that if HQ site is completely down and DR sites becomes active, our branch routers should work with Central Manager at DR.
The issue is how would it work? All branch WAEs are registered to an IP address of HQ CM. So, how would it work with DR CM in case of disaster?
Basically, when your WAAS setup is up and running, you don't need the CM for normal optimization to take place (except for secure store etc).
So first discussion you'll need to take, is how long time will your "disaster" last, and can you live without a CM for that time being.
If the answer to this is that you'll need a CM at the DR site, then you would set up a standby CM at the DR site (remember same WAAS version as primary CM).
The standby will get the configuration synchronized from the primary CM.
If the primary CM disapears (for some reason) you will need to perform a manual procedure to "promote" the standby to primary - it is not automatic.
The procedure(s) for the different scenarios is described here :
BTW : All you WAE should register to the same (primary) CM so you don't end up with two WAAS management "domains".
Thanks Finn for your reply.
I have understood that DR CM should be configured as Standby. What about IP addressing?
Normally, Branch WAEs are registered on Primary CM that has 2 DC WAEs. When DR CM is activated manually, it has saperate 2 DR DC WAEs so what will happen with IP addressing and cached data?
Normally : the primary has one address, the standby has another. They can be same subnet/VLAN or different - but not same IP-adress !
When the standby takes the primary role, it sends out information to all the WAE, because it basically has a synchronized copy of the CMS database. Therebye the WAE knows about the "new" CM. ... at least this is to my understanding.
Regarding your question about IP-addresses (of the WAE) and cached data :
These two things are completely independant.
If the HQ or DR WAE sees the traffic, and is the optimization peer), it will cache the traffic.
So the content of the cache is not dependant of the IP-address on the WAE itself.
All 4 WAEs should register to the same (primary) CM - thus the two WAE's at the DR site should NOT have the same IP addresses as the two at the main site (HQ).
So normally you'll have all 4 up and running (I assume, unless the DR site is a really cold site), they just wont see any traffic, as the servers aren't running there, but in the main site.