I'm facing some interesting problems.
My helpdesk contacted me and advised that a user is facing the following problem:
- Calls to a bridge and being muted after 30 minutes he has to drop the call and call again.
1- Changed the number he was dialing. He was dialing to a number in Singapore and changed to a number in USA the problem persisted.
2- Upgraded the IOS from the Gateway (3825) it was running the c3825-spservicesk9-mz.124-23.bin changed to c3825-spservicesk9-mz.124-25c.bin problem persisted
3- Made the same call using a diferent Gateway using E1 from a another provider. It is a 3825 running c3825-spservicesk9-mz.124-23.bin problem persisted
4- Made a call and sniffed that. When the problem occurs I can see RTCP packted being exchanged from Gateway to Phone normal but RTP packtes from the Gateway to the Phone stops and RTP packtes from Phone to Gateway pass.
I'm going to open a SR in TAC from this but I was wondering if someone has a clue on that.
I will add that the problem in intermitent and doing a call right now and until now no problem detected. Message was edited by: Rafael Chavantes
Are you using any kind of NAT that might be timing out after 30 minutes? What kind of ip addressing and routing structure do you have in place?
Have you run any diagnostics or looked at the DSP status on the gateway? The 30 minutes is a bit odd but you could potentially have a bad DSP...is the 30 minutes consistent or estimated?
Please rate helpful posts!
The DSP seems ok. The 30 minutes is very well estimated the times I saw this after 30 minutes the call started to get some cuts and a few seconds later it muted.
There is a command in order to check the DSP integrity?
This happens on a CME I have exactly, with a SIP provider.
It was suggest to change some sip paramenter, but that did not helped in my case.
You should be able to find the relevant post in this forum.
So no SIP anywhere? What type of handsets? E1 is a direct 'traditional' cicuit to the service provider?
Paolo is on the right track here, I think. I have also had the same problem with a certain SIP service provider. I was able to work around the issue by disabling or extending the SIP Session Expire Timer on my (non-Cisco) system. Not everyone implements the SIP spec exactly the same way and that is what can cause issues such as this.
People after some troubleshooting I'm starting to think that the problem is located at the remote end.
I mean for some reason the bridge concentrator stops sending me RTP packets.
Dou you think that can be the problem? Basically after 30 minutes I continue to receive RTCP packets but no RTP from the gateway to the phone.
The doesn't drops it remaisn up but with no sound coming in.
This "bridge concentrator" is Off Net, correct? Which means that the voice path is established over the PSTN? Which means that bridge concentrator isn't terminating/generating RTP, correct? I just want to clarify because some folks started going down the SIP road and you said you were using E1.
You say the gateway is sending RTCP but not RTP. Have you checked to see if you have VAD configured on the voip dial-peer? VAD would behave in the exact same way from a network analyze perspective. Now I suspect this is not the issue because of the long process you have in play to troubleshoot this. But, it could be a clue. Have you checked/researched the bug toolkit for any DSP related defects. You said it was intermittent in the OP. I have seen DSP issues appear intermittent to callers until it was determined that the problem was a single DSP that was only hit when call volume hit a certain percentage. Which means I have to ask if you have been able to determine any trends with this issue, particularly time of day?
Please remember to rate helpful responses and identify
The bridge concentrator is off-net running over the PSTN. My guess is that the bridge concentrator after 30 minutes stops generating RTP packets.
Basically what happens is that after 30 minutes I can see RTP packets going out from the IP Phone to the gateway but cannot see RTP coming back from the gateway to the IP Phone, RTCP packets are all the time normal going and coming back.
No vad at all.
Did not look for DSP bugs yet.
After some troubleshootings and looking some logs I could see that most of the time when users call this bridge after 30 minutes the connection gets mute and they have to call again, yes I could check that in about 10 calls 1 work well the other 9 the problem occurs.
No particularly time of the day, the first impression is that this doesn't matter.
There is a way to check DSP integrity?
IT Infrastructure Analyst, Information Technology
Rod. Presidente Dutra 2660 | 21535-900, RJ | Brazil
P: 55 21 2472 7518 | F: 55 21 2471 4129