SIP to PRI Call Loop

[Anonymous]

I am seeing an issue on our Adtran 900 Series where if a customer's PBX does not have a number programed in it's dialplan and it rejects the call the Adtran is sending a 503 Service Unavailable back to our Softswitch, which in turn, routes the calls right back to the Adtran creating a loop. In the past I have deployed hundreds of Adtrans and don't remember having this issue, I am not sure if it is related to software version 10.3.1 or not. Is there some configuration that can be entered on the Adtran to prevent this?

[Anonymous]

Upgrade to R10.3.3

It has fixed a lot of issues for me.

[Anonymous]

Thanks, I will take a look at the release notes and get it in my lab.

[Anonymous]

Cwallace,

I just wanted to check in with you to see if your problem was resolved.  If the firmware upgrade does not resolve the issue, we would likely need to see the exact response the Adtran unit gets on the PRI for the inbound call.  This response can be viewed by capturing the output of "debug isdn L2-formatted" for a test call.  Also, if the PBX is sending a particular cause code, we can sometimes modify the response sent out on the SIP side.  Below is an example of some of the default cause codes that map to a SIP 503.

Feel free to respond to this email with the debug output mentioned above, or with any questions you may have.

Thanks!

David

voice cause-code-map from-pri 101 503

voice cause-code-map from-pri 34 503

voice cause-code-map from-pri 38 503

voice cause-code-map from-pri 41 503

voice cause-code-map from-pri 42 503

voice cause-code-map from-pri 44 503

voice cause-code-map from-pri 46 503

voice cause-code-map from-pri 47 503

voice cause-code-map from-pri 49 503

voice cause-code-map from-pri 50 503

voice cause-code-map from-pri 58 503

voice cause-code-map from-pri 83 503

voice cause-code-map from-pri 84 503

voice cause-code-map from-pri 85 503

voice cause-code-map from-pri 88 503

voice cause-code-map from-pri 95 503

[Anonymous · Accepted solution]

I ended up opening a ticket with Adtran Support earlier this week and the loop was caused by the fact that we have two trunks configured for outbound routing redundancy. One trunk goes to SBC A while the other trunk goes to SBC B. They recommended that we put reject statements in our outbound voice grouped-trunk so that the calls don't get routed back out the other SIP trunk causing the loop. I did some testing and this fixed the issue.