5 Replies Latest reply on Aug 27, 2012 10:14 AM by david

    Trunk errors on the NetVanta 644

    unified Past_Featured_Member

      Hello,

       

      What would cause the following errors:

      2012.08.09 08:21:42 TM.T02 0 IsdnTrunkManager::getIsdnChannel - TA = 19, IsdnChannel is NULL

      2012.08.09 08:21:42 TM.T02 20 IsdnTmStateIdling - clear trunk appearance

      2012.08.09 08:21:42 SB.CALL 178033 RTP resource unavailable or SDP negotiation failed. Call from (XXXXXXXXXX) to (XXXXXXXXXX).

       

      These errors happened while the call was coming out the PRI routing to a VoIP gateway.

      This call never reach the VoIP gateway (verified by a packet trace that I keep running).

       

      Any help will be greatly appreciated.

       

      Thank you,

        • Re: Trunk errors on the NetVanta 644
          david Employee

          Unified,

           

          We may have a B channel allocation or resource issue of some kind.  I would first verify that both sides of the PRI are set to use the same number of B channels.  If you set the resource selection to circular, can you determine if the failures only occur when specific B channels are requested?  We would likely need to see the following debug output for a given test call.

           

          debug voice verbose

          debug isdn L2-formatted

          debug sip stack message

           

          Thanks!

          David

            • Re: Trunk errors on the NetVanta 644
              unified Past_Featured_Member

              Hi David,

               

              I have thousands of calls going through the 644 daily.

              This only happened once.

                • Re: Trunk errors on the NetVanta 644
                  david Employee

                  unified,

                   

                  With that call volume, we bring in the possibility both systems trying to access the same B channel at the same time.  This is typically called glare.  The best prevention is to make sure that each device prefers a different end of the B channels.  In Adtran CLI syntax, this would be "resource-selection linear ascending" on one unit and "resource-selection linear descending" on the other side of the PRI.  Given this configuration, the potential for glare really only exists when there is only one free B channel. 

                   

                  If you get a complete debug capture, we would be glad to take a closer look.

                   

                  Thanks,

                  David