Since this seems to be a development version dev issue and not a configuration. At least I think it is. This works fine in 4.2.1 but in 4.4.0- 2011-03-21EDT13:50:34 swift it doesn't have the issue with the same configuration.
The gateway does not try to negotiate a different codec. Everything successfully negotiates for G711u on first try. Other items of note: This is a HA system also. Polycom 335, 560, 670 phones using latest 3.2.4 and 4.3.0 If I shutdown the primary system, the call is maintained through multiple holds using the secondary system. If I shutdown the secondary system and just use the primary system, the symptom exists. Calls placed on hold are disconnected after the second hold placement. The system passes all the DNS tests and I have made sure that all the DNS records are available as listed. Inbound call difference than below. If a call is inbound and placed on hold. MOH is heard immediately, unlike below. After a resume, the MOH is still on (very annoying when talking to someone) If placed on hold again, call is disconnected in 4 seconds. The ~~mh~ in the CDR is active and remains active on the 2811 for 5 minutes. -- Paul Curtis From: Joegen Baclor <[email protected]<mailto:[email protected]>> Reply-To: sipXecs developer discussions <[email protected]<mailto:[email protected]>> Date: Thu, 24 Mar 2011 14:19:54 +0800 To: sipXecs developer discussions <[email protected]<mailto:[email protected]>> Cc: Paul Curtis <[email protected]<mailto:[email protected]>> Subject: Re: [sipx-dev] sipx 4.4.0 and Cisco 2811 Gateways This should be posted in the user forum. I suspect that you are encountering a classic re-invite failure due to codec renegotiations. I would take a look at the packet echange and see if your gateway tries to change codecs between the hold attempts. On 03/24/2011 01:47 PM, Paul Curtis wrote: I am having an issue with version sipXconfig (4.4.0- 2011-03-21EDT13:50:34 swift) where the call is disconnected when put on hold twice. Scenario: We have a cisco 2811 ver IOS15.1 with and ISDN PRI and using it as a sip gateway to SipXecs. User calls in from PSTN Call is answered. Answering user places PSTN call on hold. (MOH is invited but not audio is heard by PSTN user) Answering user resumes call, now MOH is heard along with the voice. Answering user places PSTN call on hold again. MOH is not invited again. 4 seconds later, the call is disconnected. It seems that cisco 2811 disconnects the call because it thinks its a mute call after the second on hold. Also, in SipXecs and on the cisco. The ~~mh~~@ is still active for about 5 minutes. This did not occur on version 4.2.1. Here is the pertinent cisco config: voice call carrier capacity active voice rtp send-recv ! voice service voip allow-connections sip to sip no supplementary-service sip moved-temporarily no supplementary-service sip refer fax protocol t38 version 0 ls-redundancy 0 hs-redundancy 0 fallback pass-through g711ulaw sip bind control source-interface FastEthernet0/0 bind media source-interface FastEthernet0/0 registrar server expires max 3600 min 3600 ! voice class codec 101 codec preference 1 g711ulaw dial-peer voice 5551212 voip huntstop destination-pattern 5035551212 progress_ind setup enable 3 session protocol sipv2 session target sip-server voice-class codec 101 dtmf-relay rtp-nte ip qos dscp cs5 media no vad sip-ua max-forwards 15 retry invite 3 retry response 3 retry bye 3 retry cancel 3 sip-server dns:uws.edu Does anyone have any ideas? -- Paul Curtis _______________________________________________ sipx-dev mailing list [email protected]<mailto:[email protected]> List Archive: http://list.sipfoundry.org/archive/sipx-dev/ _______________________________________________ sipx-dev mailing list [email protected]<mailto:[email protected]> List Archive: http://list.sipfoundry.org/archive/sipx-dev/
_______________________________________________ sipx-dev mailing list [email protected] List Archive: http://list.sipfoundry.org/archive/sipx-dev/
