pr timeout not ending dtmf capture mode

[Sean H]

Here is a situation where a PR time-out occurred right as the eik/dp was pressed and detected.  Both the time-out event and the EIK event were delivered to SLEE.

WireShark extract:

500 243.086719000 XX -> YY MGCP 155 NTFY 28173 mobicents/asr/3@YY:2427 MGCP 1.0  mgcp.param.observedevents == "AU/oc(rc=328)"

502 243.267512000   ZZ -> YY RTP EVENT 62 Payload type=RTP Event, DTMF Pound # 

503 243.287524000   ZZ -> YY RTP EVENT 62 Payload type=RTP Event, DTMF Pound # 

504 243.307509000   ZZ -> YY RTP EVENT 62 Payload type=RTP Event, DTMF Pound #

505 243.327526000   ZZ -> YY RTP EVENT 62 Payload type=RTP Event, DTMF Pound #

506 243.347495000   ZZ -> YY RTP EVENT 62 Payload type=RTP Event, DTMF Pound #

507 243.347663000 YY -> YY MGCP 165 NTFY 28174 mobicents/asr/3@YY:2427 MGCP 1.0  mgcp.param.observedevents == "AU/oc(rc=100 dc=# pi=0)"

In the situation where a PR instruction is going to event because of a 327/328, subsequent/simultaneous DTMF is still processed by the endpoint even though the instruction is effectively ended.  

Should the following code block be added, to these locations, each time a signal is fired and a failure announcement is not requested?

664

672

676

690

      reset();

      isCompleted=true;

      complete();

---Sean

[oifa.yulian]

Hello
Sorry for late response , it looks i dont get notifications from groups for some reason.
In raws 664 and 676 you should indeed add what you have wrote.
raws 672 and 690 you should replace oc.fire function call with decreaseNa();
Please let me know if it helps , if yes , please open an issue.
Thanks and best regards
Yulian Oifa