We have identified the following scenario:
- Enterprise pool at central location & backup location
- Edge pool at central location & backup location
- Standard Edition servers at remote sites, associated with Edge pool at central location
We have found that when the Edge servers at the central site are unavailable, outbound calling across all locations fails. I understand there is a dependency on the Edge server, since the Lync client uses it to check the bandwidth policy service before
it can place the call. If no Edge server is defined, the Lync client uses the front-end pool.
My concern is that if a Standard edition server loses connectivity to the Edge pool, I lose outbound dial. This isn't a DR situation; simply having the Edge servers down prevents outbound dial. I don't see this mentioned anywhere very prominently
(or at all), and would like to know if there's a way force a client to use its front end rather than Edge for the bandwidth policy check? It seems that our options for site resiliency require NOT associating an Edge server with the site, unless
it is locally available. Is this correct?
TIA
Kevin Clark | blog.blimeyuc.com