Hi,
I have created our Exchange 2010 SP2 environment consisting of 2 x Windows Server 2008 R2 SP1 servers running on vSphere 5 and come across an issue regarding the CAS servers and client access to the CAS servers if one of the servers goes down. Both servers have the HT, CAS, MBX and UM roles.
I have also created a DAG on the servers called CompanyEmail.domain.co.uk with 2 NICs, one for normal network traffic and one for the DAG Replication.
I came across this article: http://info.kraftkennedy.com/blog/bid/102113 and realised that I need to create a ClientAccessArray for my CAS servers but I still need to load balance them don't I?
Therefore can I create a NLB cluster on these servers and then create the ClientAccessArray with the same FQDN? Does this have to be different to my DAG FQDN? I could add a 3rd NIC for the NLB heartbeat.
I did read somewhere that I can't create a NLB cluster on these servers that already have a DAG on them. Is that right? And if so, what are my options? Create 2 more servers and make these the CAS servers and use NLB? Or use a hardware load balancer or virtual appllance load balancer?
Please help!
Thanks