January 2012 - FreeSWITCH-users - FreeSWITCH Mailing Lists

ring tone is heard for 2 sec before busy tone on Lync client
by Manjiri Deshpande 06 Jan '12

06 Jan '12

Hi, I am using freeswitch 107 on windows platform I am working for a product where lync client can call a phone registred with Lync server. When Phone is set to DND mode and lync client calls that phone,I could hear ring tone for abt 2 sec and then busy tone comes. The behvaiour should be ,only busy tone should be heard. I removed "ringback" and "ignore_early_media" from dialplan. Please find dial plan , freeswitch logs and Wireshark capture attached with this mail. From wireshark it is clear that no 180 or 183 messages are flowing.But still from FS logs it could be figure out that,FS is entering into "RINGING" mode. Thanks, Manjiri Deshpande

1 0

Session timeout
by Peter Steinbach 06 Jan '12

06 Jan '12

I have enabled session timers in my sofia internal profile the following way: <param name="enable-timer" value="true"/> <param name="session-timeout" value="120"/> However I do not see any Re-Invite or Update packets sent from Freeswitch to my User Agents. What am I doing wrong? Peter -- With kind regards Peter Steinbach Telefaks Services GmbH mailto:lists (att) telefaks.de Internet: www.telefaks.de

3 2

Re: [Freeswitch-users] Encrypted clients to unencrypted destination ... possible ?
by Bob Smith 05 Jan '12

05 Jan '12

Hello Peter, I will have another go at this over the next couple of days and let you know ! Bob >Hello Bob, > >sure, this does work. I have a couple of phones working with TLS and >SRTP and they all communicate to the outside world through Freeswitch >via SIP and RTP. You have to ensure though that Freeswitch is not is >proxy or bypass media mode. > >Best regards >Peter

1 0

Encrypted clients to unencrypted destination ... possible ?
by Bob Smith 05 Jan '12

05 Jan '12

Hello List, To save me much head-banging and time-wasting, could someone please tell me whether it is possible to configure Freeswitch so that an endpoint can register over TLS, but can route out to and receive calls from the big bad unencrypted outside world ? I've got a basic DDI config at the moment that I can receive calls on fine when endpoint registers as unencrypted. But the moment I turn on TLS, the DDI no longer rings. I've followed the instructions here http://wiki.freeswitch.org/wiki/SIP_TLS to enable TLS (only enabling TLS on the internal Sofia obviously). By the way, it would be great if someone could update those instructions with more detail as to how to get things working with a commercial certificate, as I had no luck making it work, so there's obviously a trick that's missing, and had to resort to a self signed one. Thanks Bob

2 1

bridging 2 SIP destinations from PBX behind NAT
by Michal Zubáč 05 Jan '12

05 Jan '12

Hi. I encountered problem with bridging two SIP calls. My FreeSwitch is behind NAT, I'm not sure if endpoint switches are, but I think it doesn't matter now. Situation looks like this: SIP1 <--> NAT <--> PBX <--> NAT <--> SIP2 SIP1 is caller switch, SIP2 is destination switch so call is initiated like this: SIP1->PBX->SIP2 I tested both endpoints separately using ISDN endpoint and they worked. Like this ISDN1->PBX->SIP2 & SIP1->PBX->ISDN2 But when I connected SIP1 & SIP2 together, call is estabilished, but no audio was going through our firewall. Both audio streams from SIP1 and SIP2 were filtered out by our firewall, because Freeswitch (PBX) wasn't sending any initial packets to SIP endpoints, so no NAT holes were created. It look strange to me. I expected FreeSwitch to at least send some "empty" RTP packets to SIP endpoints as soon as call estabilishment is confirmed on SIP channel. But Freeswitch doesn't do it and only sends ringing indication. I had problems in scenario, where SIP2 endpoint doesn't ring and immediately answers the call. Call was estabilished, but audio was stuck in our firewall, because Freeswitch haven't initiated any RTP communication yet and was probably waiting for something. I managed to solve this problem by adding <action application="set" data="instant_ringback=true"/> between my <action application="ring_ready" /> and <action application="bridge" /> commands in dialplan. Now it works, but I'm feeling kind of uncertain with this solution, because I don't really understand why. What was Freeswitch waiting for, before I added instant_ringback? Why can't FreeSwitch prepare NAT holes (as described before) as soon as SDP are interchanged? Thanks for clarification. Regards -- Michal Zubac ComGate Interactive s.r.o. Prague Marina Office Center Jankovcova 1596/14a 17000 Praha 7, Czech Republic

2 1

help with wait_for_silence
by Hector Geraldino 05 Jan '12

05 Jan '12

Hi, I'm trying to use the wait_for_silence application to stop recording audio from a channel, and I need to use it not from a FS dialplan but from an ESL application. So far I can execute the wait_for_silence application using 'execute wait_for_silence arguments' with no issues. I can even see the result of the execution of this command on the fs_cli console: 2012-01-05 19:04:40.266001 [DEBUG] switch_ivr.c:576 sofia/internal/5512(a)192.168.8.11 Command Execute wait_for_silence(200 15 15 5000) EXECUTE sofia/internal/5512(a)192.168.8.11 wait_for_silence(200 15 15 5000) The problem is that I don't know if this is working or not, as it is not raising any event that can be received by the application. I don't see any event related to this application (or to detect silence using any other mechanism) on the wiki Events List page [http://wiki.freeswitch.org/wiki/Event_list] so I'm not sure it is usable in an socket outbound mode. Resuming: is it possible to detect the silence on a channel and then perform an action in an ESL application, using wait_for_silence app or any other mechanism? Thanks, Hector

2 2

SIP attacks from 188.161.101.73
by Karol 05 Jan '12

05 Jan '12

Anyone else see this?? http://pastebin.freeswitch.com/18078

6 6

How to troubleshoot "503 Maximum Calls In Progress"?
by Chris Fowler 05 Jan '12

05 Jan '12

Hi, On our Production FreeSWITCH box: FreeSWITCH Version 1.0.head (git-4c435ec 2011-03-14 11-54-08 -0500) UP 0 years, 14 days, 6 hours, 28 minutes, 44 seconds, 474 milliseconds, 751 microseconds 12519 session(s) since startup 2 session(s) 0/30 1000 session(s) max min idle cpu 0.00/100.00 At 14:30:37 a reloadxml command was issued. At 14:30:43 the box started rejecting all calls with 503 error. At 14:38 it started working again. Looking at the code I see three conditions can trigger this behavior: if (sess_count >= sess_max || !sofia_test_pflag(profile, PFLAG_RUNNING) || !switch_core_ready()) { nua_respond(nh, 503, "Maximum Calls In Progress", SIPTAG_RETRY_AFTER_STR("300"), TAG_END()); switch_log_printf(SWITCH_CHANNEL_LOG, SWITCH_LOG_CRIT, "No more sessions allowed at this time.\n"); goto done; } Logs show it wasn't lack of available sessions. How can sofia_test_pflag(profile, PFLAG_RUNNING) or switch_core_ready() fail and busy the system? Thoughts? Thx, Chris.

3 2

Sangoma A200 FXO Outgoing Problem
by Ryan V 05 Jan '12

05 Jan '12

Hi, I have configured A200 and able to receive incoming calls. I am unable to dial out. Here are the logs. 2011-12-08 16:39:15.990089 [DEBUG] ftmod_analog.c:62 [s3c1][2:1] Changed state from DOWN to DIALING 2011-12-08 16:39:15.990089 [DEBUG] ftmod_analog.c:439 [s3c1][2:1] ANALOG CHANNEL thread starting. 2011-12-08 16:39:15.990089 [DEBUG] ftmod_analog.c:459 [s3c1][2:1] Initialized DTMF detection 2011-12-08 16:39:15.990089 [DEBUG] ftmod_analog.c:640 [s3c1][2:1] Completed state change from DOWN to DIALING in 0ms 2011-12-08 16:39:15.990089 [DEBUG] ftmod_analog.c:646 [s3c1][2:1] Executing state handler on 3:1 for DIALING 2011-12-08 16:39:21.009503 [DEBUG] ftmod_analog.c:505 [s3c1][2:1] Changed state from DIALING to BUSY 2011-12-08 16:39:21.030044 [DEBUG] ftmod_analog.c:640 [s3c1][2:1] Completed state change from DIALING to BUSY in 20ms 2011-12-08 16:39:21.030044 [DEBUG] ftmod_analog.c:646 [s3c1][2:1] Executing state handler on 3:1 for BUSY 2011-12-08 16:39:21.030044 [DEBUG] ftmod_analog.c:802 [s3c1][2:1] Changed state from BUSY to DOWN 2011-12-08 16:39:21.049928 [DEBUG] ftmod_analog.c:640 [s3c1][2:1] Completed state change from BUSY to DOWN in 20ms 2011-12-08 16:39:21.049928 [DEBUG] ftmod_analog.c:646 [s3c1][2:1] Executing state handler on 3:1 for DOWN 2011-12-08 16:39:21.049928 [DEBUG] ftmod_analog.c:944 [s3c1][2:1] Going onhook2011-12-08 16:39:21.049928 [DEBUG] switch_core_state_machine.c:362 (FreeTDM/3:1/9449905000) Running State Change CS_HANGUP Here is our dial plan <extension name="outgoing_bsnl"> <condition field="destination_number" expression="^9(\d{10})$"> <action application="bridge" data="freetdm/3/a/$1"/> </condition> </extension> I suspect FS is not detecting the dial tone on analogue channel. BTW, outgoing on our digital span works fine. Any suggestions? Thanks, Ryan

5 6

Flite speed and pitch increase after update.
by Brandon McGinty 05 Jan '12

05 Jan '12

Hi everyone. We upgraded Freeswitch on the 30th of December, using the git branch revision from the 28th of December. However, while the conference features improved, the flite package has had its pitch and rate increased for all tts speech. We are using the mod_tts module in our conferences, and it now sounds extrordinarily sped up and pitched up, similar to what one would consider a chipmunk sound. I've gone through the mod_tts sources, the flite sources, the modules directory, the autoload_modules directory, the configs for the conference and flite, and I can't find anywhere how to change the pitch and rate of the synthesizer. I've tried tts-rate and tts-pitch, etc, to no affect. I've also tried google, and noone else seems to be dealing with this strangeness. We are all using soft-phones or pots to sip devices, so the speech output is quite important. Any help you all can provide would be truly be appreciated. Brandon McGinty-Carroll

4 4