April 2012 - FreeSWITCH-users - FreeSWITCH Mailing Lists

Dialplan/PHP esl variable problem.
by Alex Crow 09 Apr '12

09 Apr '12

Hi all, I have a dialplan that calls a PHP script through CURL: <extension name="test" continue="false"> <condition field="destination_number" expression="^669$"> <action application="answer"/> <action application="sleep" data="1000"/>  <action application="curl" data="http://localhost/provision/index.php?action=test&uuid=${uuid}&exten=${calle…" /> <action application="sleep" data="1000"/> <action application="log" data="INFO ${curl_response_data}"/> <action application="hangup" data="NORMAL_CLEARING"/> </condition> </extension> (notice the play_and_get_digits is commented out) and in PHP I connect via ESL via this function and code: function saySomething($uuid) { $sock = new ESLconnection('localhost','8021','ClueCon'); sleep (2); $res = $sock->execute("play_and_get_digits","4 11 1 5000 # ivr/ivr-please_enter_the_phone_number.wav ivr/ivr-that_was_an_invalid_entry.wav target_num \d+", "$uuid"); //$foo = $uuid . ' target_num'; sleep (2); $res=$sock->sendRecv("api uuid_getvar $uuid target_num"); $ret = $res->getBody(); $sock->disconnect(); //echo $ret; return ($ret); } if ($action == 'test') { $res = saySomething($_REQUEST['uuid']); echo $res; } However if I run the play_and_get_digits from the PHP the subsequent uuid_getvar returns _undef_. If I comment out the play_and_get_digits in PHP and uncomment it in the dialplan, the uuid_getvar *does* return the digits pressed. Is there something I'm doing wrong here? Thanks Alex -- This message is intended only for the addressee and may contain confidential information. Unless you are that person, you may not disclose its contents or use it in any way and are requested to delete the message along with any attachments and notify us immediately. "Transact" is operated by Integrated Financial Arrangements plc Domain House, 5-7 Singer Street, London EC2A 4BQ Tel: (020) 7608 4900 Fax: (020) 7608 5300 (Registered office: as above; Registered in England and Wales under number: 3727592) Authorised and regulated by the Financial Services Authority (entered on the FSA Register; number: 190856)

5 21

tone detect
by dinh thuc 09 Apr '12

09 Apr '12

Hi I want to test tone detect in Freeswitch as: detect tone and events I want to create virtual: send and get tone depend on events I have read http://wiki.freeswitch.org/wiki/Misc._Dialplan_Tools_tone_detectand http://wiki.freeswitch.org/wiki/Mod_spandsp#start_tone_detect Do you know the way to establish freeswitch to get events tone decect ? (that mean is vitual tone to test in freeswitch) B& R

1 0

Fwd:
by Manjiri Deshpande 08 Apr '12

08 Apr '12

http://www.caliearthquakekits.com/entrance.php?jkypilo=237&xuhutohu=85

1 0

Need help with FreeTDM config
by David George 07 Apr '12

07 Apr '12

I am writing an application that uses FreeTDM with a Digium TE-220 Dual T1 card. I started by looking at the testanalog sample and found that I also needed to add ftdm_global_configuration after ftdm_global_init. I am now able to place a call, but I get a segfault early on referencing sigmsg->channel. I am using an Adtran TSU600e for a channel bank and have a mixture of FXS and FXO cards in it. Right now I am just testing with a phone plugged into an FXS port configured for TANDEM_LS. The T1 is configured AMI/D4 for channelized T1 with robbed bit signalling. The T1 is up and I have no alarms. There isn't a lot of information on the FreeSWITCH Wiki or on the Sangoma FreeTDM Wiki for T1 RBS. There is one mention that I should use channel type of fxo-channel or fxs-channel for RBS ( http://wiki.freeswitch.org/wiki/FreeTDM ). I saw another place that says to use em-channel, but it may be specific to running Dahdi on Sangoma cards ( http://wiki.sangoma.com/FreeSWITCH-Dahdi-Mode ). I have tried both configurations, but have gotten further with the fxo-channel config than with the em-channel config. With the channel set to fxo-channel I get a segfault when I place a call and the first sigmsg->channel is referenced. I can open a Jira and add backtraces, but I wanted to check if I probably had a configuration issue first. E&M configuration I tried: [span zt tsu600e] trunk_type => EM em-channel => 1-4 ;fxo-channel => 1-4 ;fxs-channel => 5-6 ;fxo-channel => 9-12 FXO config I tried: [span zt tsu600e] fxo-channel => 1-4 Channel 1-4 is really an FXS card. I am using the reverse like I used to with Asterisk. Not sure if it is correct or not. Thanks, David -- David George Hark Technologies

1 0

vm-keep-local-after-email vs vm-delete-file?
by Iruwen 07 Apr '12

07 Apr '12

Hi, is vm-delete-file deprecated in favor of vm-keep-local-after-email? I can only find the latter in the wiki now (http://wiki.freeswitch.org/wiki/Mod_voicemail#vm-keep-local-after-email) Are they supposed to do the same? The recorded mp3 files aren't removed for me, anybody else experience that? Regards

2 1

core dump on sip profile, fs version 2012-04-05
by Nestor A Diaz 06 Apr '12

06 Apr '12

Hi guys, i have a core dump generated on freeswitch everytime i restart a new profile i have created: freeswitch@internal> version FreeSWITCH Version 1.1.beta1 (git-c41a16d 2012-04-05 14-28-01 -0500) freeswitch@internal> sofia profile internal-voice flush_inbound_reg reboot Socket interrupted, bye! [ERROR] fs_cli.c:1320 main() Error Connecting [Socket Connection Error] [INFO] fs_cli.c:1326 main() Retrying The full gdb trace is here (thanks to moises silva for the tip): http://pastebin.com/d0uRnDrs I updated from FreeSWITCH Version 1.0.head (git-9be51d5 2012-01-21 13-45-21 -0500) to this version thinking it will fix the problem, but the problem remains. the profile internal-voice is just the same internal profile with the following things changed: # diff internal.xml internal-voice.xml 1c1 < <profile name="internal"> --- > <profile name="internal-voice"> 18c18 < <X-PRE-PROCESS cmd="include" data="internal/*.xml"/> --- > <X-PRE-PROCESS cmd="include" data="internal-voice/*.xml"/> 79c79 < <param name="context" value="public"/> --- > <param name="context" value="internal"/> 89c89 < <param name="rtp-ip" value="$${local_ip_v4}"/> --- > <param name="rtp-ip" value="$${voice_ip_v4}"/> 91c91 < <param name="sip-ip" value="$${local_ip_v4}"/> --- > <param name="sip-ip" value="$${voice_ip_v4}"/> where voice_ip_v4 is defined in vars.xml as follows: <X-PRE-PROCESS cmd="set" data="voice_ip_v4=172.17.0.20"/> <X-PRE-PROCESS cmd="set" data="voice_net_v4=172.17.0.0/24"/> voice_ip_v4 is no other thing than a vlan interface for voip phones. By proof and error i found that if i rename from internal-voice to another name, i.e.voice.xml then the sip profile doesn't core dump agin, a name problem ? i know that is difficult to believe but that's what i am experiencing, before upgrading gdb show me a problem on sofia_glue.c On another freeswitch server i still have the problem of the event trap it change without having changed the ip address, like the following: 2012-03-22 20:34:22.353275 [INFO] mod_sofia.c:5155 EVENT_TRAP: IP change detected 2012-03-22 20:34:22.353275 [INFO] mod_sofia.c:5156 IP change detected []->[] [fe80::225:90ff:fe3a:2e1a]->[::1] I don't use the internal-ipv6 profile at all, so maybe the best thing i can do is to disable it. I am not sure what triggers the above behavior but when it has been triggered then the vlan profile i use for the phones can't be started again, if i do a : `netstat -unlp` the related port is in use by the freeswitch process but the profile is down. it writes a lock on the profile: 2012-03-22 20:34:24.253327 [DEBUG] sofia.c:2121 Write lock internal but no 'unlock' like it does with the other profiles. So i finally moved all the phones to the data vlan and no problem. Another problem i have found is when i do a `show calls` and there is no calls the system show some calls, i found this only one time. No space problem, Any explanation ? Overall i am happy with my first freeswitch experience, it keeps on processing from 30 to 40 calls concurrently on business hours from sip devices and a sangoma pri card with echo cancelation in conjunction with a sangoma transcoder card and a supermicro server. Having working with asterisk before and sick with locks on the queue function i think freeswitch matters the transition from Asterisk, i have some things to master but i am on the way .... Slds. -- Nestor A. Diaz Ingeniero de Sistemas Tel. +57 1-485-3020 x 211 Cel. +57 316-227-3593 Tel. SIP: sip:211@tiendalinux.com Email/MSN: nestor(a)tiendalinux.com http://www.tiendalinux.com/ Bogota, Colombia

2 2

FW: Inbound ESL .RecvEventTimed(0) Blocking?
by Michael Lutz 06 Apr '12

06 Apr '12

Hi, Can someone please give me an answer, I am problaby not the first to notice. I need to know why (and how to avoid) ESLevent RecvEventTimed(0) being blocking and keeps waiting for event to arrive, as it should not. It seems to ignore any form of time you pass as parameter and just always wait for an event. Btw, I am on Windows. Thanks! Mike. From: Michael Lutz Sent: donderdag 5 april 2012 22:19 To: 'FreeSWITCH Users Help' Subject: Inbound ESL .RecvEventTimed(0) Blocking? Hi Guys, Can anybody please help me, I am using .RecvEventTimed(0) on a inbound ESL connection to check if events are there. According to specs, this should be non-blocking and as the wiki says "used for polling". However, the code just seems to stop until it gets an event. I am listening with a filter for a specific uuid like: ESLevent eslEvent = fsConnection.SendRecv("event plain ALL"); eslEvent = fsConnection.SendRecv("filter Unique-ID " + thisCall.callId); . Some code. . eslEvent = fsConnection.RecvEventTimed(0); But it just waits for an event on the RecvEventTimed(0); When I generate an event (for example pressing a dtmf) is continues. Is there some extra thing I need to do? Thanks, Mike.

3 2

windows alternative for <X-PRE-PROCESS cmd="exec" ...
by babak yakhchali 06 Apr '12

06 Apr '12

Hi I'm trying to setup a centralized dialplan for multiple instances of freeswitch. I tried <X-PRE-PROCESS cmd="exec" data="wget -qO - http://www.example.com/dialplan.xml" /> but it is not implemented on windows yet. is there any similar alternative? thanx

10 12

fs_cli error
by Jagadish Thoutam 06 Apr '12

06 Apr '12

I am getting below error when i try to fs_cli freeswitch version : FreeSWITCH Version 1.1.beta1 (git-c41a16d 2012-04-05 14-28-01 -0500) [root@localhost freeswitch]# ./fs_cli [ERROR] fs_cli.c:1320 main() Error Connecting [Socket Connection Error] Usage: ./fs_cli [-H <host>] [-P <port>] [-p <secret>] [-d <level>] [-x command] [-t <timeout_ms>] [profile] -?,-h --help Usage Information -H, --host=hostname Host to connect -P, --port=port Port to connect (1 - 65535) -u, --user=user@domain user@domain -p, --password=password Password -i, --interrupt Allow Control-c to interrupt -x, --execute=command Execute Command and Exit -l, --loglevel=command Log Level -q, --quiet Disable logging -r, --retry Retry connection on failure -R, --reconnect Reconnect if disconnected -d, --debug=level Debug Level (0 - 7) -t, --timeout Timeout for API commands (in miliseconds)

3 3

originate
by David Villasmil 05 Apr '12

05 Apr '12

Hello guys, i'm trying to execute an originate originate sofia/external/669448337(a)1.2.3.4 0013058883456 +OK a27c762c-7f4a-11e1-9269-f52ca8b46747 2012-04-05 20:10:32.893284 [ERR] switch_core_io.c:929 Codec PROXY PASS-THROUGH encoder error! WHY do i see that error? codecs are fine both on FS and and the endpoint... I originally want to originate and then generate dtmf to the leg... if anyone can give me a help, i would appreciate it thanks! David

2 2