Planning for High-Availability

When adding high-availability to a Lync Server Edge Server deployment, an organization must make a decision about how it will provide load-balancing features. The only options available for load-balancing Edge Servers are to use Lync’s native DNS load-balancing feature or to leverage a hardware load-balancing solution.

DNS load balancing seems like an attractive feature at first because it requires very little configuration, but does have some minor limitations. Organizations should review these limitations and then make a decision about whether a hardware load balancer is required. The main limitation of DNS load balancing is that it does provide automatic failover for some features or legacy endpoints, including these:

• Endpoints running previous versions of the Office Communicator client

• Federated organizations running Office Communications Server 2007 R2 or previous

• Public IM connectivity

• Exchange Server 2007 and 2010 Unified Messaging

This doesn’t mean these features won’t work; it simply means they are not aware of DNS load balancing and the fact that they could leverage a second server if the first DNS record returned is not responding. So in a scenario in which both servers are online there will be no difference, but during an outage these features will not automatically fail over. Administrators might need to manually remove some DNS entries to prevent clients from connecting to a failed server.

Many organizations still elect to use DNS load balancing despite the minor limitations addressed here due to the deployment simplicity and cost savings.