Hi guys,
I have two geographic sites, setup as separate central sites with a front end EE pool in each central site. Each central site has a single pool.
Each site has it's own associated edge server.
Everything is working as expected. If a user is moved to a different pool then everything works which is great.
I'm now at the point of testing DR, where I have pool pairing setup. Lync failed over to the other pool as expected, and after moving the CMS and invoking failover to move the users all services are restored and life is good....
Other than federation!
The scenario is that Pool A / Site A (paired with Site/Pool B) goes offline. Everything fails over to Pool B, and the user that was originally on Pool A is all happy and working on Pool B. If I look into the configuration information on the Lync 2013 client I can see that the client is set to use the Edge server in Pool B.
If the user tries to federate externally it works - I can IM, see presence and chat, even do calls. It's all good.
However if an external party tries to federate with the user then it fails. Can't see the presence, nothing. It only works if the internal user initiates the chat.
I have DNS setup so that _sipfederationtls._tcp.domain.com and _sip._tls.domain.com each have two SRV records each. The lowest priority points to Site A Edge Server, the higher priority points to Site B Edge Server.
I even changed DNS so that the priority is reversed, with Pool B being the first one tried. (Remember that Pool B is now the only operation pool / site).
Can anyone shed any light on this please? Only thing I can think of doing is changing the site federation route assignment in topology builder for the failed site (Site A) to use the Edge server in Site B. However if I do this, what servers do I need to re-run Lync Deployment? Any of them? The only existing Edge server (running in Site B), or all FE servers and the Edge?
Thanks - Steve