- FreeSWITCH-users - FreeSWITCH Mailing Lists

FreeSWITCH powered Application Server
by Thomas Quintana 07 Jan '14

07 Jan '14

Hi Guys, I have been an avid user of FreeSWITCH for quite sometime and have been fortunate enough to attend a ClueCon convention. Recently, I had the need to use Python for building communications software but could not find any solutions that were 100% async on top of ESL and that really exposed the full power of the platform in an intuitive manner. This has led me to create an actor based framework coupled with a declarative FSM library for building FreeSWITCH powered Python apps. I'm still polishing the project but it is now in a very usable state. I hope it may be as useful to others as FreeSWITCH has been to me. Thank You! https://github.com/thomasquintana/freepy Best, Thomas Quintana P.S. Sorry for the lack of creativity in the name.

2 3

Strangely High CPU Usage on Idle FS server
by Anthony Matejcich B.Com 06 Jan '14

06 Jan '14

HI guys Relatively new to the FS world and have exhausted research online. Running FS FreeSWITCH Version 1.2.10+git~20130624T024915Z~8fd4994b0c (git 8fd4994 2013-06-24 02:49:15Z) On VMWare virtual machine with Debian6 and specs of 2 CPU with 1 GB RAM, 50GB storage Currently only running one client with very very low call usage until we have tested the FS server. Total of 30 calls and day maximum and basically only running a simple IVR set up with 7 digit keypress prior to sending to Lua script for a simple calculation and then delivery to mobile or landline in Australia. System has been great for 6 months but issue arose around the 13th Dec 13 - dmtf entries are being duplicated. A single kepyress appears to register 2 of the same digit and is recorded as such in the CLI notes. Only happens about 50% of the time. Added a CPU to the mix and it may have assisted with this becoming about 30% of the time showing the issue now. But that seems crazy to require CPU's given it is such a small set up. Reviews showed high CPU Usage via top command information. CPU usage for the FS process is sitting between 14 and 22%! (constantly fluctuating) For the idle machine/with no calls coming in. Thus %wa is fluctuating between 1 and 8% constantly. There is no load on the server - but we did look at master logs and update core.db to the default <param name="core-db-name" value="/dev/shm/core.db" /> Very strange given the success over the first 6 months. Can anyone help please with information or where to go? Would be greatly appreciated. Anthony

2 2

Switching from RPM to GIT
by Callum Guy 06 Jan '14

06 Jan '14

I have a simple Monday morning query - if i've installed FreeSWITCH using RPM's and have now decided to remove the application and use the git head - is there a simple way to do this? I'm pretty much a novice so far as Linux administration so my approach would be as follows: List the packages: rpm -qa | grep -i freeswitch freeswitch-1.2.14-1.el6.x86_64 freeswitch-application-snom-1.2.14-1.el6.x86_64 freeswitch-appl.... ... ... (x30) Remove them all one at a time using rpm -e freeswitch-1.2.14-1.el6.x86_64 Proceed with GIT install. Is there a cleaner/easier approach? Thanks.

4 5

dynamic change conference profile in conference.conf.xml
by Shijun Bian 06 Jan '14

06 Jan '14

Hi all, I just be in trouble in the conference setting. I am building up a freeswitch cluster to server >100,000 users and all these users will be in thousands conference room. because the user base is growing day by day , so I have to dynamically add freeswtich node if necessary. But the ip address of the conference room is specified in the profile as the "domain" field in the conference.conf.xml which is loaded when the freeswitch started . now the problem is I can not restart all the freeswitch nodes the add a new profile into the conference.conf.xml when a new IP is be added into the cluster. so I wonder is it possible that I could add a new conference profile dynamiclly? I have been in trouble and google around for days but still can't find a way to solve it. Please help, thanks in advance. -- *Best Regards* *ShiJun Bian*

2 1

dynamic account and anonymous registration
by Jimmy Chang 06 Jan '14

06 Jan '14

Hi, I'm planning to build a public WebRTC web page. I'd like to use sipml5 for WebRTC client. Users can make WebRTC video calls from this client to FS registration agents without any login processes. I want to implement this WebRTC client without a explicit register. My problem is how to do that. Here are my three thoughts and questions. 1) WebRTC clients generate dynamic accounts for public users and use these accounts to register to FS. I wonder that if mod_xml_curl can fullfill my requirement. Or how can I implement dynamic account in FS? 2) WebRTC clients use a anonymous registration. I have seen FS support multiple registrations, so I create a anonymous user named "public_user" and allow it multiple registrations. I set <param name="multiple-registrations" value="true"/> in sip profile. Does that fulfill my requirement? 3) sipml5 make calls without any registrations. Is it possible? Does any sipml5 guru know how to do that? Regards, Jimmy --- 由於 avast! 防毒軟體已開啟防護，所以此封電子郵件將不受病毒及惡意軟體的侵害。 http://www.avast.com

3 3

Git v1.2.17 ca447fe 2014-01-02 - Polycom IP-330 outbound calls hang on answer
by Francis 05 Jan '14

05 Jan '14

Debian Wheezy 7.3 32bit Version 1.2.17 -n20140102T213421Z-1~wheezy+1git ca447fe 2014-01-02 11:34:22Z 32bit Phone firmware: 3.3.5 Configured as EXT 1000 Incoming calls work fine, but outgoing calls, either to local extensions or though a gateway hang up the instant the call is answered by the remote party. The IP-330 shows the call as answered and in progress and there is a split second of audio before it drops, but the actual call has dropped and the other party is idle. Calling three different endpoints has the same effect: Grandstream GXP2124 (extension 1002) Zoiper communicator on Android Sip trunk to PSTN A capture in wireshark shows the following sequence immediately after the call is established: 3963 8.986119000 10.0.0.124 10.0.0.119 SIP/SDP 1017 Request: UPDATE sip:1000@10.0.0.119:5060, with session description 3963 8.986119000 10.0.0.124 10.0.0.119 SIP/SDP 1017 Request: UPDATE sip:1000@10.0.0.119:5060, with session description 3994 9.002869000 10.0.0.119 10.0.0.124 SIP 443 Status: 481 Call Leg/Transaction Does Not Exist 3996 9.003466000 10.0.0.124 10.0.0.119 SIP 687 Request: BYE sip:1000@10.0.0.119:5060 4029 9.019642000 10.0.0.119 10.0.0.124 SIP 440 Status: 481 Call Leg/Transaction Does Not Exist .... at which point, a BYE is sent to the b leg and the call terminates, with the IP-330 left dangling with an empty call leg, not realizing that nobody is listening anymore. This is what I gleaned from fs_cli: 2014-01-05 18:13:08.904735 [DEBUG] sofia.c:5779 Channel sofia/internal/1000(a)hpbx.nest entering state [calling][0] 2014-01-05 18:13:08.904735 [DEBUG] switch_rtp.c:927 [ zrtp utils]: Send <HELLO> ssrc=1538954628 seq=54771 size=144. Stream 40:CLEAR:START 2014-01-05 18:13:08.924723 [DEBUG] switch_core_session.c:1016 Send signal sofia/internal/1000(a)hpbx.nest [BREAK] 2014-01-05 18:13:08.924723 [DEBUG] switch_core_session.c:1016 Send signal sofia/internal/1000(a)hpbx.nest [BREAK] 2014-01-05 18:13:08.924723 [DEBUG] switch_core_session.c:1016 Send signal sofia/internal/1000(a)hpbx.nest [BREAK] 2014-01-05 18:13:08.924723 [DEBUG] switch_core_session.c:1016 Send signal sofia/internal/1000(a)hpbx.nest [BREAK] 2014-01-05 18:13:08.924723 [DEBUG] switch_core_session.c:1016 Send signal sofia/internal/1000(a)hpbx.nest [BREAK] 2014-01-05 18:13:08.944752 [DEBUG] sofia.c:5779 Channel sofia/internal/1000(a)hpbx.nest entering state [terminating][487] 2014-01-05 18:13:08.944752 [NOTICE] sofia.c:6623 Hangup sofia/internal/1000(a)hpbx.nest [CS_EXECUTE] [ORIGINATOR_CANCEL] 2014-01-05 18:13:08.944752 [DEBUG] switch_channel.c:3183 Send signal sofia/internal/1000(a)hpbx.nest [KILL] 2014-01-05 18:13:08.944752 [DEBUG] switch_core_session.c:1351 Send signal sofia/internal/1000(a)hpbx.nest [BREAK] 2014-01-05 18:13:08.944752 [DEBUG] switch_ivr_bridge.c:633 BRIDGE THREAD DONE [sofia/internal/1000(a)hpbx.nest] 2014-01-05 18:13:08.944752 [DEBUG] switch_ivr_bridge.c:658 Send signal sofia/internal/sip:1002@10.0.0.120:5060 [BREAK] 2014-01-05 18:13:08.964735 [DEBUG] switch_ivr_bridge.c:633 BRIDGE THREAD DONE [sofia/internal/sip:1002@10.0.0.120:5060] 2014-01-05 18:13:08.964735 [DEBUG] switch_ivr_bridge.c:658 Send signal sofia/internal/1000(a)hpbx.nest [BREAK] 2014-01-05 18:13:08.964735 [NOTICE] switch_ivr_bridge.c:721 Hangup sofia/internal/sip:1002@10.0.0.120:5060 [CS_EXCHANGE_MEDIA] [NORMAL_CLEARING] 2014-01-05 18:13:08.964735 [DEBUG] switch_channel.c:3183 Send signal sofia/internal/sip:1002@10.0.0.120:5060 [KILL] 2014-01-05 18:13:08.964735 [DEBUG] switch_core_session.c:1351 Send signal sofia/internal/sip:1002@10.0.0.120:5060 [BREAK] 2014-01-05 18:13:08.964735 [DEBUG] switch_core_state_machine.c:481 (sofia/internal/sip:1002@10.0.0.120:5060) State EXCHANGE_MEDIA going to sleep 2014-01-05 18:13:08.964735 [DEBUG] switch_core_state_machine.c:415 (sofia/internal/sip:1002@10.0.0.120:5060) Running State Change CS_HANGUP 2014-01-05 18:13:08.964735 [DEBUG] switch_core_state_machine.c:679 (sofia/internal/sip:1002@10.0.0.120:5060) State HANGUP 2014-01-05 18:13:08.964735 [DEBUG] mod_sofia.c:500 sofia/internal/sip:1002@10.0.0.120:5060 Overriding SIP cause 480 with 487 from the other leg 2014-01-05 18:13:08.964735 [DEBUG] mod_sofia.c:506 Channel sofia/internal/sip:1002@10.0.0.120:5060 hanging up, cause: NORMAL_CLEARING 2014-01-05 18:13:08.964735 [DEBUG] switch_ivr_bridge.c:1524 sofia/internal/sip:1002@10.0.0.120:5060 skip receive message [UNBRIDGE] (channel is hungup already) 2014-01-05 18:13:08.964735 [DEBUG] switch_ivr_bridge.c:1527 sofia/internal/1000(a)hpbx.nest skip receive message [UNBRIDGE] (channel is hungup already) 2014-01-05 18:13:08.964735 [DEBUG] switch_core_session.c:2810 sofia/internal/1000(a)hpbx.nest skip receive message [APPLICATION_EXEC_COMPLETE] (channel is hungup already) 2014-01-05 18:13:08.964735 [DEBUG] switch_core_state_machine.c:478 (sofia/internal/1000(a)hpbx.nest) State EXECUTE going to sleep 2014-01-05 18:13:08.964735 [DEBUG] switch_core_state_machine.c:415 (sofia/internal/1000(a)hpbx.nest) Running State Change CS_HANGUP 2014-01-05 18:13:08.964735 [DEBUG] mod_sofia.c:558 Sending BYE to sofia/internal/sip:1002@10.0.0.120:5060 2014-01-05 18:13:08.964735 [DEBUG] switch_core_state_machine.c:48 sofia/internal/sip:1002@10.0.0.120:5060 Standard HANGUP, cause: NORMAL_CLEARING Version 1.2.17 git 7712626 2013-12-27 20:52:28Z 64bit works fine.

2 2

Lua script not working in Freeswitch
by Joli Martinez 05 Jan '14

05 Jan '14

Hello, I have a script that connects to a mysql db. The script works perfect when I type "lua <scriptname>", but when I have Freeswitch all the script I get the following error message. I have changed my script to now call a python script to read the database and I am experiencing the same result. If I call the script from the command line it works but Freeswitch throws this error. Please help, Thanks, 2014-01-03 18:52:28.381720 [ERR] mod_lua.cpp:199 /usr/local/freeswitch/scripts/testCallerID.lua:31: attempt to concatenate global 'pythonResult' (a userdata value) stack traceback: /usr/local/freeswitch/scripts/testCallerID.lua:31: in main chunk

5 7

Fail2ban wiki update
by Jim Lynch 04 Jan '14

04 Jan '14

In attempting to setup a working fail2ban installation, I have been in communications with the maintainers of fail2ban. They have taken an interest in supporting FreeSWITCH and have asked for comments on https://github.com/fail2ban/fail2ban/pull/552. The only information they have is what I've been able to give them from my admittedly sparse knowledge of FS error messages. At their suggestion I have deleted the sample configuration on the wiki and replaced it with a link to a configuration on their site. It is a small improvement to the deleted configuration. If that information is incorrect or incomplete please let them (or me) know. I'll relay any responses. Thanks, I hope I haven't overstepped my bounds. Jim.

1 0

iLBC and spandsp codecs not available in stock install
by Francis 02 Jan '14

02 Jan '14

Hi, When I do a show codecs in fs_cli, none of the spandsp codecs show up, even though GSM and G722 are enabled in vars.xml and mod_spandsp is enabled in modules.conf.xml. The module was compiled and installed. I'm having the same issue with iLBC. This isn't consistent with all modules however, since speex works just fine. Any ideas? Francis

4 7

strange behavior in setting origination_caller_id_number
by Jun Sun 02 Jan '14

02 Jan '14

I have a very simple call plan to allow internal extension to call external PSTN numbers. See below. <extension name="domestic.example.com"> <condition field="${toll_allow}" expression="domestic"/> <condition field="destination_number" expression="^(\d{11})$"> <action application="set" data="origination_caller_id_number=11111111"/> <action application="set" data="effective_caller_id_number=${outbound_caller_id_number}"/> <action application="set" data="effective_caller_id_number=${outbound_caller_id_number}"/> <action application="set" data="effective_caller_id_name=${outbound_caller_id_name}"/>  <=== NOT WORK  <=== NOT WORK <action application="bridge" data="[origination_caller_id_number=111111]sofia/gateway/idt/$1"/> <=== WORKS!! </condition> </extension> There a couple of things strange here: 1. I have to put origination_caller_id_number inside data field of "bridge" application in order for the dialing out to happen. Why putting it outside as channel variable does not work? 2. Actually because bridge command is for leg-B, really it should be "effective_caller_id_number" not "origination_caller_id_number". Is that right? I'm using IDT express as SIP carrier. I don't know if they changed something or I changed configuration in freeswitch. Now it seems I have to set origination_caller_id_number if I want to use IDT for outbound calls. I'd really appreciate your help. Cheers. Jun

5 13