December 2013 - FreeSWITCH-users - FreeSWITCH Mailing Lists

H323
by Richard Mace 17 Dec '13

17 Dec '13

HI All, Where should the file h323.conf.xml be located and how does FreeSWITCH know where's it's config files are and which ones to use? Thanks Richard

2 1

clock sync troubles
by Javier Aristizábal 16 Dec '13

16 Dec '13

Hello folks, I would like to know if FreeTMD will allow two different clocks to be used for two different ISDN cards? The BRI running on the NTT (telco) clock and the PRI card running on its own hardware clock then pass an incoming call on the BRI card over FreeTDM to the PRI card. Would this avoid clock sync troubles? Regards -- Javier Aristizábal

2 5

Is it possible to play a wav file to all legs at the same time?
by Ali Pey 16 Dec '13

16 Dec '13

Hello, I would like to play an announcement before call recording starts. I tried uuid_broadcast, but it plays the file on a-leg first and then after that it plays the file on the b-leg. It does not play the file on both legs at the same time. Is there a way to play a file on all legs simultaneously? Thanks, Ali Pey

1 1

Building mod_managed - error with swig.cs
by Erik M. Devane - Comms Guy 16 Dec '13

16 Dec '13

All, It's been eighteen days since I moved my production server from Windows to CentOS, as I was recommended, and I couldn't be happier with how stable the system has been. Now I'm trying to migrate some tools that I wrote for Windows over to CentOS, and I'm having difficulty building mod_managed. What I want to do is to send an array of bytes over TCP whenever an incoming call is bridged or unbridged. Initially, I used the excellent ESL dll and the managed wrapper, and I tried using lua but support for sending byte arrays is limited. I decided that mod_managed would allow me to reuse some of my interface code, and so I followed instructions in the wiki to build mod_managed, using the same 1.2.15 tarball that I'm currently running. The error I'm getting is this: making all mod_managed swig.cs(27328,27): error CS1519: Unexpected symbol `event' in class, struct, or interface member declaration swig.cs(27328,29): error CS1519: Unexpected symbol `{' in class, struct, or interface member declaration swig.cs(27329,9): error CS1519: Unexpected symbol `{' in class, struct, or interface member declaration swig.cs(27330,55): error CS1519: Unexpected symbol `(' in class, struct, or interface member declaration swig.cs(27330,64): error CS1519: Unexpected symbol `,' in class, struct, or interface member declaration swig.cs(27330,86): error CS1519: Unexpected symbol `(' in class, struct, or interface member declaration swig.cs(27330,92): error CS1519: Unexpected symbol `)' in class, struct, or interface member declaration swig.cs(27332,4): error CS1525: Unexpected symbol `get' swig.cs(27333,6): error CS1525: Unexpected symbol `IntPtr' swig.cs(27333,68): error CS1525: Unexpected symbol `(' swig.cs(27334,45): error CS1525: Unexpected symbol `)' swig.cs(27334,59): error CS1514: Unexpected symbol `new', expecting `.' or `{' swig.cs(27334,57): error CS1530: Keyword `new' is not allowed on namespace elements swig.cs(27334,60): error CS1525: Unexpected symbol `switch_event' Compilation failed: 14 error(s), 0 warnings Opening swig.cs in my development environment shows that it is, indeed, using reserved names for methods - renaming 'event' to 'fsevent' stops the C# compiler from complaining and mod_managed builds. Thinking that I was just using old code, I checked the git tree, there's an updated swig.cs, but it also uses events and params keywords, and also fails to build. Does anyone know what I'm missing? Erik

2 1

WebRTC no media sometimes
by Seven Du 16 Dec '13

16 Dec '13

Hi, So I findally got some time trying WebRTC. I noticed that some clients has no media when using WebRTC. I using the latest master with jssip 3.0. While my chrome(30.0.1599.101) has no problem all the time, other chromes elsewhere seems has no luck. e.g. 31.0.1650.63 and 33.0.1737.0 canary. I listed 3 channels below, looks like only my channel has 'AES_CM_128_HMAC_SHA1_80' and other channels has no that and no media. Also I see no auto stun for a long time from the log. Before trying more, is any hint it's a Chrome version problem or anything else? the no audio stun for a long time log has anything related to the client or only related to server side? Any configurable params? Thanks. L16,48000,768000,opus,48000,0,srtp:sdes:AES_CM_128_HMAC_SHA1_80,webRTC,1000@115.29.169.114,,ACTIVE,,,,,, L16,48000,768000,opus,48000,0,,webRTC,1000(a)115.29.169.114,,ACTIVE,,,,,, L16,48000,768000,opus,48000,0,,webRTC,1000(a)115.29.169.114,,ACTIVE,,,,,, 3 total. 2013-12-13 19:18:24.459945 [ERR] switch_rtp.c:718 No audio stun for a long time! 2013-12-13 19:18:35.479939 [ERR] switch_rtp.c:718 No audio stun for a long time! -- Seven Du http://www.freeswitch.org.cn http://about.me/dujinfang http://www.dujinfang.com Sent with Sparrow (http://www.sparrowmailapp.com/?sig)

5 8

Authenticated users still using public context
by Nathan Samson 16 Dec '13

16 Dec '13

Hi, I fear I am going to ask a very stupid and/or commong question, but after searching the internet I stil lcoudln't find the answer. My Setup * I have users in the directory (see an example configuration at the bottom of this mail) * Authenticated users should be able to use the default Dialplan * I want my DID provider to be able to make calls to me, it should use the public Dialplan / context. My DID provider is mydivert but that seems to be irrelevant. * I only have one SIP Profile (external) which has as default context public. All my users are connecting from the internet and not from the local network, so the internal profile would not make sense (or am I wrong?) My problem is that calls from nathan(a)sip.radarwerk.be are NOT using the default context, but are using the public context. 2013-12-15 15:04:53.206937 [DEBUG] switch_core_state_machine.c:117 sofia/external/nathan(a)sip.radarwerk.be Standard ROUTING "2013-12-15 15:04:53.206937 [INFO] mod_dialplan_xml.c:558 Processing nathan <nathan>->1999 in context public" The user nathan in the directory has user_context set to default, so I do not get why this is happening. See below for my configuration. [root@sip freeswitch]# cat directory/default/nathan.xml <include> <user id="nathan" number-alias="1200"> <params> <param name="password" value="**"/> </params> <variables> <variable name="toll_allow" value="domestic,international,local"/> <variable name="user_context" value="default"/> <variable name="effective_caller_id_name" value="Nathan Samson"/> <variable name="effective_caller_id_number" value="1200"/> <variable name="outbound_caller_id_name" value="Nathan Samson"/> <variable name="outbound_caller_id_number" value="1200"/> </variables> </user> </include> Anyone an idea why the user_context is ignored? I am happy to provide my complete configuration. But basically I am using freeswitch-config-rayo 1.2.16 from the CentOS 6 repositories with some small modifications (mainly adding mydivert gateway, my users and my custom extensions in the default directory). Regards, Nathan

3 4

Gateway Registration Issues
by Aloysius Lloyd 15 Dec '13

15 Dec '13

Hi All, I am trying to connect IRISTEL ( CLEC ) in Canada and FreeSWITCH. I have the following gateway configuration <include> <gateway name="wwc_test"> <param name="auth-username" value="username"/> <param name="username" value="14161231234" /> <param name="password" value="password" /> <param name="proxy" value="sbc1.iristel.net" /> <param name="realm" value="iristel.net" /> <param name="expire-seconds" value="120"/> <param name="register" value="true"/> <param name="retry-seconds" value="30"/> <param name="ping" value="25"/> </gateway> </include> The above configuration is not working. I have the following error in cli Error. *2010-05-13 08:34:15.987465 [ERR] sofia_reg.c:1501 14161231234 Registration Failed with status Conflict [409]. failure #1* ==== Also carrier saying there is conflict in the contact field.Here is how they receive now. *Contact: <sip:gw+test_iristel@10.20.30.1:5080 ;transport=udp;gw=test_iristel>.* But the contact field should be like below Contact: <14161231234@10.20.30.1:5080; Any suggestions? Thanks in advance. Lloyd

5 6

mod_xml_curl creating too many open files
by Tim St. Pierre 15 Dec '13

15 Dec '13

Hello, I'm having a problem where FS seems to exhaust it's open file limit after about 7 hours of operation. Once this happens, the database and curl connections no longer function, and I usually lose the ability to get a CLI connection. Existing calls usually stay up, until I kill -9 and launch again. I'm using FreeSWITCH Version 1.2.8+git~20130403T221701Z~79be96aa8e on Centos 2.6.18-348.el5 This machine had been running solid for at least 90 days without a hitch - I had almost hit 500K sessions, when this started happening two days ago. I have changed very little on the machine - the odd dialplan route, and nothing that seems to coincide with that time. It looks like CURL is not closing sockets properly, as netstat shows between 2000 - 4000 http connections in TIME_WAIT state. The number slowly creeps up, then down a little, then up some more until it all comes crashing down. ls /proc/15742/fd|wc -l currently returns about 400. It was 300 an hour ago. When it gets to 1024, I'm going to have to restart again. Any suggestions as to where I should look? -Tim

4 12

FreeSWITCH 1.2.16 Released!!
by Ken Rice 13 Dec '13

13 Dec '13

Greetings FreeSWITCHers The FreeSWITCH Dev team is pleased to announce the Release of FreeSWITCH 1.2.16! This is a maintenance release. Updates included: SpanDSP and Faxing code to address outstanding bugs, updates to sofia and libsofia to address possible memory leak, and many other issues addressed. Source tarball is available at http://files.freeswitch.org/freeswitch-1.2.16.tar.bz2 Packages for Debian and RHEL/Centos available via the FreeSWITCH deb and yum repos! -- Ken http://www.FreeSWITCH.org http://www.ClueCon.com http://www.OSTAG.org irc.freenode.net #freeswitch Twitter: @FreeSWITCH

2 1

Unable to send peer-to-peer messages with mod_sms
by Iskren Hadzhinedev 13 Dec '13

13 Dec '13

Greetings! I installed freeswitch from the yum repository on a clean centos 6.5, loaded mod_sms and tried to send a text message between two clients. I use extensions 1001 and 1002 for testing. Had to change the default chatplan to "public" and when I send a message from one client to the other the debug console prints out: -- 2013-12-13 11:48:20.545501 [INFO] mod_sms.c:336 Processing text message 1002->1001 in context public Chatplan: 1001 parsing [public->demo] continue=false Chatplan: 1001(a)172.16.1.10 Regex (PASS) [demo] to(1001(a)172.16.1.10) =~ /^(.*)$/ break=on-false Chatplan: 1001(a)172.16.1.10 Action reply(Hello, you said: ${_body}) 2013-12-13 11:48:20.545501 [DEBUG] sofia_presence.c:224 Can't find registered user 1002(a)172.16.1.10 2013-12-13 11:48:20.567453 [WARNING] sofia_presence.c:4523 Not sending message to ourselves! -- and the demo reply doesn't show up on the sending device. When I invoke "sofia_contact 1001" from fs_cli, however I get this: sofia/internal/sip:1001@172.16.1.10:54302;ob Invoking "sofia_contact 1000", which is not used at the moment returns this: error/user_not_registered So, why can't sofia_presence find the registered user when it's listed correctly and how do I fix this? Thank you in advance. P.S. Before installing from the RPM repository I compiled it from scratch following the installation guide, but I had the exactly same problem. Kind regards, Iskren Hadzhinedev

2 3