locked
Response groups and the dreaded audio delay RRS feed

  • Question

  • We are experiencing mixed results with response groups.  Users answering RGS calls are getting anywhere from a .5 to 3 second delay from when they answer to when the call is actually connected and audio is flowing.  The delay seems random, .5 seconds would be acceptable but 3 seconds is really pushing it...  All of these users are using a fully updated Lync client as their platform.  The lync server, media gateway and RGS clients are all located within the same physical site so network latency is not the problem.  Not to mention non-RGS calls connect almost instantly.  What could I start checking if this could be a possible configuration issue?
    Monday, November 7, 2011 7:27 PM

All replies

  • Hi,

    there's nothing much to check in Configuration as long as your getting the call to an agent. You can try with changing the agent login method (formal, informal). and also check in the same workstation a non RGS user signed in, if the problem exists. if not then you can rule out the workstation.

    Try running Sip Stack in Logging too and check wthere you can find any errors in the log. also you can use Wireshak on the mediation server and see what's happening in real time. chack and confirm whether it works when you dial the RGS Line URI from internal Lync Client and how it behaves.

    Thamara. 

    Tuesday, November 8, 2011 2:41 AM
  • This behaviour is inherent in the product unfortunately. The delay you're experiencing (the "pip" sound) is the Response Group service finding an agent and then when it does, cancelling the request to all other agents in the queue.

    To do this it must find all users in the queue and all their signed in endpoints and notify them of an incoming call, so this generates a lot of SIP traffic, resulting in the behaviour you're seeing.


    Justin Morris | Consultant | Modality Systems
    Blog - www.justin-morris.net
    Twitter: @jm_deluxe
    If this post has been useful please click the green arrow to the left or click "Propose as answer"

    Tuesday, November 8, 2011 2:49 PM
  • Thank you both.  Justin, that is pretty much what I was assuming.  My next test will be to try changing the routing method to something other than Parallel.  Then it wouldn't have to cancel the request to the other agents, but in the end Parallel is best suited for our company.
    Tuesday, November 8, 2011 5:19 PM
  • try something like Longest Idle. it works well for me.

    Thamara.

    Tuesday, November 8, 2011 5:40 PM
  • Hi tpullins,

    This issue occurs because the RGS takes several seconds to establish a connection between the caller and the RGS agent. Please refer to KB2517739.

    In addition, please also check if your previous post is helpful.

    • Edited by Noya Lau Wednesday, November 9, 2011 2:48 AM
    Wednesday, November 9, 2011 2:46 AM
  • Hi tpullins,

    This issue occurs because the RGS takes several seconds to establish a connection between the caller and the RGS agent. Please refer to KB2517739.

    In addition, please also check if your previous post is helpful.


    That article is referring to Lync Phone Edition, which I am not using.  I even have that hotfix installed already.  Yes my previous thread was because I was getting a huge 7+ second delay, luckily I figured that one out.  Now the delay is a bit more reasonable, but still very annoying.
    Wednesday, November 9, 2011 5:07 AM