I've recently deployed Exchange 2013 in our organisation (upgraded from 2010). With our 2010 environment I had 3 mailbox servers in the DAG, 2 at head office (SITE1) and one at our DR site (SITE2). With this being a DR site the bandwidth usage was quite
high throughout the day.
I have tried to recreate this setup with 2013 but, from the articles I've read, each site with a mailbox server role also requires the client access role. This is where the problem seems to start!
Due to the high bandwidth usage (not just Exchange) to the DR site the ECP webpage becomes extremely slow at generating pages, property windows, etc. I have seen this issue reported by other users in a similar situation. I'm guessing it's because all Exchange servers need to send data and it's having to wait longer to receive the required data from SITE2.
We also get issues with Outlook becoming unresponsive. Users at SITE1 appear to have no issues, SITE2 I'm not aware of any reported issues. But other sites appear to select whichever client access server they like.
I can cope to a point with the ECP slowness but is there any way we can prevent clients from selecting the SITE2 CAS over SITE1?
Would setting the DR site CAS role setting 'isOutOfService' to True (until required in a DR situation) make a difference or would this cause more issues as then technically there is no CAS role in SITE2. My other thought was to somehow add a cost to the CAS, but I'm not sure if that would be possible.
Thanks for any help!
I have tried to recreate this setup with 2013 but, from the articles I've read, each site with a mailbox server role also requires the client access role. This is where the problem seems to start!
Due to the high bandwidth usage (not just Exchange) to the DR site the ECP webpage becomes extremely slow at generating pages, property windows, etc. I have seen this issue reported by other users in a similar situation. I'm guessing it's because all Exchange servers need to send data and it's having to wait longer to receive the required data from SITE2.
We also get issues with Outlook becoming unresponsive. Users at SITE1 appear to have no issues, SITE2 I'm not aware of any reported issues. But other sites appear to select whichever client access server they like.
I can cope to a point with the ECP slowness but is there any way we can prevent clients from selecting the SITE2 CAS over SITE1?
Would setting the DR site CAS role setting 'isOutOfService' to True (until required in a DR situation) make a difference or would this cause more issues as then technically there is no CAS role in SITE2. My other thought was to somehow add a cost to the CAS, but I'm not sure if that would be possible.
Thanks for any help!