How to setup redunancy for internal relay hosts

We are working on an Exchange 2003 to 2010 migration.  The customer wants redundancy wherever possible.  We are planning two Transport servers and I understand shadow redundancy for messages coming in from the edge transport.  These transport servers will also be used for internal relay from various scanners and other machines need relay service.

My question is this.  What can I do to build in some redundancy for these various machines needing internal relay.  I can only put one relay server address into most of these machines.  Can I use some sort of network load balancer to accomplish redundancy?  Is there some more accepted method of accomplishing redundancy for internal relay machines?

Tx



Bill
LVL 1
labdunnAsked:
Who is Participating?
 
Simon Butler (Sembee)Connect With a Mentor ConsultantCommented:
If you want redundancy everywhere then you will need a load balancer.
The load balancer will allow you to have redundancy on the CAS connections via an RPC CAS Array, HTTPS traffic and also SMTP traffic. The DNS address will point to the load balancer.
You can also put most load balancers in to a pair so they are redundant as well.
Look at Kemp and JetNexus as the two main players in this space.

Configure everything with unique name spaces though - so SMTP traffic uses smtp.example.com, the CAS array outlook.example.local, HTTPS services webmail.example.com etc.

You do NOT need to have the CAS array address on the SSL certificate.

Simon.
0
All Courses

From novice to tech pro — start learning today.