Lync Server Support Home
Top Lync Solutions RSS Feed
These short videos focus on specific tasks and show you how to accomplish them for Microsoft Lync Server 2010.
Check out the most comprehensive, actively managed UC blog roll in the known universe, your one-stop source for links to over 100 of the very best UC blogs. Here you will also find weekly blog highlights and a feed for a dozen of the top blogs.
Microsoft Lync Server 2010 communications software allows users to connect to each other remotely. This chapter describes the most common remote access scenarios, and then provides detailed call flow diagrams, traces, and related ms-diagnostics codes. Overviews of the basic scenarios are provided, giving context for how users make various connections. You can download this and other Microsoft Lync Server 2010 Resource Kit chapters from the Download Center.
Authors: Randy Wintle, Dustin Hannifin
Publication date: August 2011
Product version: Lync Server 2010, Lync 2010
The upcoming Microsoft Lync Server 2010 Resource Kit book will provide in-depth technical content about Lync Server 2010. The book will focus on professionals who want to understand more about how the product works internally. You can download this and other Microsoft Lync Server 2010 Resource Kit chapters from the Download Center.
Internals (technical details of remote user access capabilities) are discussed in depth in this chapter and figure prominently in the scenarios. An introduction to the Interactive Connectivity Establishment (ICE) protocol, Simple Traversal Underneath NAT (STUN), and Traversal Using Relay NAT (TURN) implementations in Lync Server are provided.
Lync Server 2010 allows users to access instant messaging (IM) and presence remotely by using the Lync Server Access Edge service. The scenarios in this chapter describe remote IM and presence in remote user access, federation, and public IM connectivity.
The most common call flow scenarios that involve remote user access are described in detail and as follows:
Keywords: Presence, IM, instant messages, Enterprise Voice, peer to peer, conferencing, PSTN, gateway, remote access, federated, user, partner,
Great job on this chapter Randy and Dustin. You both rock!
Very nice material, I've learnt a lot from it. But I've found some figures confusing and not in sync with the scenario (e.g. peer-to-peer call flow). It would be also good to see IPs on the network diagrams. A bit more details about how candidate IP-port pairs are obtained would be good to.
Attila, thanks for the feedback. Just sent a note to the authors on this.
Hi Attila, can you comment with some more specific details with errors in the diagrams?
As for your comment about allocation:
Ice will look at all available IPv4 addresses available on the client. Using these addresses in addition to the default port range, or administrator configured port range, the client will randomly allocate a pair of ports in that range per call. In addition, the client will also discover the closest public ip address which is being used for nat purposes and allocate ports there (these are the reflexive or home ip addresses). For the relay addresses, the client will send an allocate request to the av edge with the information provided in the initial MRAS response. The edge will validate the client credentials, then randomly allocate ports not current in use in the defined edge port range. The client combines all of this and sends with that initial INVITE. The callee will follow the same process and respond with a 200 OK. I think the chapter outlines the rest. If not, please feel free to ask any more questions, thanks!
Hi Randy, on page 37: "In this scenario, two remote Lync 2010 users...", but diagram shows one Lync user inside.
Thanks for the explanation. For me the most mysterious are the ports on the external IP of the NAT device, since those are out of control. I know that using STUN (at least how it was before), you can figure out the behavior of the NAT device: so you make a "hole" by sending a packet to the STUN server and then you test from a 1st and 2nd IP of the STUN server if that "hole" lets in packages from what IP and port combinations.
I also know that now STUN is something more than just figuring out the NAT type. Are the STUN messages themselves testing and opening the ports for media?
It is also strange to me that internal users will use external IP of edge if relaying is needed with an external participant. It means, that internal clients will use the (site) local Internet access to connect the edge server. Doesn't it spoil the CAC?
I'm studying these because I'm experiencing random problems with Polycom HDX and RMX devices, especially when Lync clients are outside. Anyway, I keep learning and hopefully understand everything better soon.