Intro

Lync DNS load balancing is a software solution you can use for the SIP / media traffic for Front End pools, Edge Server pools, Director pools, and stand-alone Mediation Server pools

You still need to also use hardware for HTTPS traffic from clients over ports 443 and 80

Additionally, if you are using Exchange UM, only Exchange 2010 SP1 or latest service pack has built-in support for Lync Server 2010 DNS load balancing

(Following image : reference topology with high availability and a single data center )

How it Works

a) The front-end servers register their fully qualified domain name (FQDN) as A records in DNS.

b) When the Enterprise pool is created, the pool FQDN is registered to return from DNS the list of IP addresses of all the front-end servers.

The difference with DNS round robin is that the client receives a list of ALL the ip related to ALL the front-end servers and choses one in a random manner.

In Lync each front-end server in a pool has a completely independent registration database but the client has a predefined registrar calculated by a hash value of the user’s SIP URI (that’s a unique value).

So tipically a client contacts the wrong registrar that redirects the client to the right registrar where it successfully registers.

Lync Server uses the hash algorithm we talked above to determine which front-end server the client will primarily connect to, and also the order of failover–for every front-end server in the pool.

The hash algorithm is based on the maximum number of servers in the pool (10).

That’s why you have a limit of ten servers in a single pool.

If a front-end server is down and the client tries to connect to it you have two possible outcomes :

• The initial server chosen from the DNS response is unavailable. Using DNS load balancing, the client will then simply attempt another server from the list until a successful connection is received. From here, a registration is attempted by using the previously mentioned logic.
• The primary registrar is unavailable. In this scenario, the server that provides the redirect will be fully aware of the states of other servers in the pool and will then redirect the client to connect to the first available backup registrar front-end server that is determined by the hash list of the user’s SIP URI.

When a server fails, the physical registrar sequence is updated to show the server as unavailable and shared amongst all surviving servers by using a server-server heartbeat. Any users who would primarily connect to the failed server are redirected to the next server in their logical registrar sequence and are then connected in backup mode.

Now, at some point in the future, the server will be recovered, returning the physical registrar sequence back to its original state.

Server Draining

A new feature called server draining enables you to take a server offline without any loss of service to users. When a server is drained it stops taking new connections and calls. These new connections and calls are routed through other servers in the pool.

If you put a FE server in draining mode, all clients will have to reconnect but active calls and conferencing session should remain active until there sessions end

When all existing sessions have ended, the server is ready to be taken offline.

DNS load balancing and Lync client are requirements for server draining.

References

http://technet.microsoft.com/en-us/library/gg398634.aspx

http://blogs.technet.com/b/ucedsg/archive/2010/10/12/lync-dns-load-balancing-and-server-draining.aspx

http://blogs.technet.com/b/nexthop/archive/2011/05/25/dns-load-balancing-in-lync-server-2010.aspx

http://voipnorm.blogspot.it/2011/03/lync-server-draining.html

Other Languages

This article is available in other languages, including Italian.

Load Balancing con DNS in Lync Server 2010