June 2010 - FreeSWITCH-users - FreeSWITCH Mailing Lists

Dropped Call During VM Email Delivery?
by Jerry Richards 14 Jun '10

14 Jun '10

We getting dropped calls recently. They seem to appear in the vicinity where mod_voicemail is delivering a voice mail Email (at least that's where we've seen them). Below, I show a snippet from the Freeswitch log that shows two calls (on extensions 1028 and 1063) getting simultaneously disconnected. Do you know what might be the cause? 2010-06-14 08:59:02.495697 [DEBUG] mod_voicemail.c:2394 Deliver VM to 1036(a)192.168.72.141 2010-06-14 08:59:02.500634 [DEBUG] ozmod_sangoma_boost.c:1174 1:23 STATE [DOWN] 2010-06-14 08:59:02.500634 [WARNING] sangoma_boost_client.c:221 TX EVENT (N): CALL_STOPPED_ACK:(86) [w1g23] Rc=0 CSid=0 Seq=766 2010-06-14 08:59:02.500634 [DEBUG] zap_io.c:1388 channel done 1:23 2010-06-14 08:59:04.500353 [DEBUG] switch_ivr_bridge.c:478 sofia/internal/1028@192.168.72.141:5060 ending bridge by request from read function 2010-06-14 08:59:04.500353 [DEBUG] switch_ivr_bridge.c:478 sofia/internal/1063@192.168.72.141:5060 ending bridge by request from read function 2010-06-14 08:59:04.500353 [DEBUG] switch_core_session.c:643 Send signal sofia/internal/1028@192.168.72.141:5060 [BREAK] 2010-06-14 08:59:04.500353 [DEBUG] switch_core_session.c:643 Send signal sofia/internal/1063@192.168.72.141:5060 [BREAK] 2010-06-14 08:59:04.500353 [DEBUG] switch_ivr_bridge.c:565 BRIDGE THREAD DONE [sofia/internal/1028@192.168.72.141:5060] 2010-06-14 08:59:04.500353 [DEBUG] switch_ivr_bridge.c:585 Send signal OpenZAP/1:1/18007297580@g1 [BREAK] 2010-06-14 08:59:04.500353 [DEBUG] switch_ivr_bridge.c:478 sofia/internal/1032@192.168.72.141:5060 ending bridge by request from read function 2010-06-14 08:59:04.500353 [DEBUG] switch_core_session.c:643 Send signal sofia/internal/1032@192.168.72.141:5060 [BREAK] 2010-06-14 08:59:04.500353 [DEBUG] switch_ivr_bridge.c:565 BRIDGE THREAD DONE [sofia/internal/1032@192.168.72.141:5060] 2010-06-14 08:59:04.500353 [DEBUG] switch_ivr_bridge.c:585 Send signal OpenZAP/1:3/19403232644@g1 [BREAK] 2010-06-14 08:59:04.500353 [DEBUG] switch_ivr_bridge.c:565 BRIDGE THREAD DONE [sofia/internal/1063@192.168.72.141:5060] 2010-06-14 08:59:04.500353 [DEBUG] switch_ivr_bridge.c:585 Send signal OpenZAP/1:2/14253491003@g1 [BREAK] 2010-06-14 08:59:04.501359 [DEBUG] switch_utils.c:631 Emailed file [/tmp/mail.12765311428da7] to [chuck.wichser(a)teotech.com] 2010-06-14 08:59:04.501359 [DEBUG] mod_voicemail.c:2562 Sending message to chuck.wichser(a)teotech.com Best Regards, Jerry

1 0

major problem with latest version
by Christian Löschenkohl 14 Jun '10

14 Jun '10

hello i expirience a big problem with the latest version (pulled from git). inbound sip calls don't get answered (i do answer them with the answer app). does anybody expirience the same issue? br -- Ing. Christian Löschenkohl Technische Leitung, Forschung & Entwicklung VoIP xpirio Telekommunikation & Service GmbH Lakeside B04 9020 Klagenfurt Austria T +43 (0) 5 77 11 - 1000 F +43 (0) 5 77 11 - 1002 E christian.loeschenkohl(a)xpirio.com

5 14

LED-Controlling on Snom phones. Missing Subject header
by Helmut Kuper 14 Jun '10

14 Jun '10

Hello, I try to control the LEDs on Snom 370 phones using LUA and event type SEND_MESSAGE. The phone receives this message: MESSAGE sip:2850@85.16.245.213:1051 SIP/2.0 Via: SIP/2.0/UDP 85.16.246.6;rport;branch=z9hG4bKeer7H9cFa85DN Max-Forwards: 70 From: <sip:2850@85.16.246.6>;tag=yH2p954XZZ1tj To: <sip:2850@85.16.246.6> Call-ID: 2a6ed9ba-ef6d-122d-3d83-00144fe6e332 CSeq: 131979517 MESSAGE Contact: <sip:mod_sofia@85.16.246.6:5060> User-Agent: FreeSWITCH-mod_sofia/1.0.head-17097:17188M Allow: INVITE, ACK, BYE, CANCEL, OPTIONS, MESSAGE, UPDATE, INFO, REGISTER, REFER, NOTIFY, PUBLISH, SUBSCRIBE Supported: timer, precondition, path, replaces Content-Type: application/x-buttons Content-Length: 10 k=1 c=on and Snom acknowledges this with 200 OK. Unfortunately nothing happens. Snom's howto (http://wiki.snom.com/Category:HowTo:LED_Remote_Control) describes a message like this: MESSAGE sip:44@172.20.25.101:2048;transport=tls;line=44qsudyt SIP/2.0 Via: SIP/2.0/TLS 172.20.25.102:5061;branch=z9hG4bK-2909fe081f8466c9836c3b673f284842;rport From: <sip:44@localhost>;tag=17820 To: <sip:44@localhost> Call-ID: umceg9jb@pbx CSeq: 35160 MESSAGE Max-Forwards: 70 Contact: <sip:172.20.25.102:5061;transport=tls> Subject: buttons Content-Type: application/x-buttons Content-Length: 7 k=3 l=42 Main difference is the subject header containing "buttons" (despite of the body) Is there a way to add subject header as above to the message generated by FS? regards Helmut

3 11

help about a system with login logout
by babak yakhchali 14 Jun '10

14 Jun '10

Hi I need to implement a system in which users should login and logout to be able to place or receive calls using cisco 7941 ip phones on freeswitch (first I tried to somehow program one of soft keys as login and logout shortcut but I couldn't do that). Any suggestions would be appreciated. thanx

6 12

Anti-tromboning in FreeSWITCH?
by Ron Avriel 14 Jun '10

14 Jun '10

Hi, Is there any way to implement Anti-tromboning/Anti-Hairpinning/Media Release (http://en.wikipedia.org/wiki/Anti-tromboning) in FreeSWITCH? My scenario is similar to image in link above. Currently, when user A calls B I get two calls and media passing twice through FS. Is there any way for media to pass directly between A and B? Thanks, Ron

5 7

position of members in the fifo
by Abubacker siddiq 14 Jun '10

14 Jun '10

Dear all, I am working in fifo using outbound sockets in that my requirement is to say the queue number for the waiting customer for every '5' seconds, I tried this with fifo_position variable but the issue is when ever the customer calls has been answered and hangup then also the queue number is not decremented , is it a correct behaviour ? I can do this using fifo list and fifo list_verbose by parsing its xml structure , but I want to know whether we have a easy way to achieve this. Thanks in advance !

1 0

uuid_media hangs
by peely 14 Jun '10

14 Jun '10

Hi, I'm trying to use the uuid_media command to control the media path after a call has been established. I can't seem to get it to respond wither through the console or through ESL. Whenever I issue a uuid_media {uuid} or uuid_media off {uuid} the console or connection just hangs, even when I have debug output I get nothing. I've tried this with both bridged and originated calls. I have enable-3pcc set to true on the profile. Effectively I am trying to set up an IVR through ESL then connect the caller on to a destination, once connected I want to push the media away. Looking at the SIP coming in from the far end I see invite and update is allowed so in theory this should be supported by the remote party. Could somebody please let me know what the conditions are to be able to use uuid_media? Thanks, Neil. -- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/uuid-media-hangs-tp5145657p51… Sent from the freeswitch-users mailing list archive at Nabble.com.

7 16

updated sqlite3?
by Helmut Kuper 14 Jun '10

14 Jun '10

Hello, today I played around with lua scripts using lua-sqlite3 module. I found that the delete statement of the sqlite3 version (3.1.3?) FS is using is not able to handle LIMIT in DELETE statements. Also the centos sqlite3 cli tool (v3.3.6) doesn't do. I found that sqlite3 version 3.6.23 does the trick as long as you use SQLITE_ENABLE_UPDATE_DELETE_LIMIT during compile time. So any chance to get FS sqlite sources updated? As long as not, can I riskless update FS's sqlite source to 3.6.23 and recompile FS? regards Helmut

2 1

Non-Blocking Music on Hold
by Jon Davies 14 Jun '10

14 Jun '10

Hi There I'm attempting to port an application I've written in CCXML/VXML to FreeSwitch/javascript and finding it a very enjoyable experience. I've hit the limits of the documentation and my knowledge and was wondering if there were any clever people on the list who can assist me with one little part. My scenario is: Make an outbound call and play it MOH Make a 2nd outbound call If the 2nd call connects, take the 1st call off MOH and bridge them together If the 2nd call fails for whatever reason, take the 1st call off MOH and play it a 'sorry' message. My problem is, all my attempts to play MOH to the first call results in blocking, meaning I dont ever get to originate my 2nd call. I've tried firstcall.streamFile("music.wav"); and firstcall.execute( "fifo", "myqueue in undef 'music.wav'" ); but both block until I hang up. Anyone got any insight as to how I can achieve this scenario in a non-blocking manner? I'm using javascript. -- Regards Jon Davies

2 1

Full NAT bypass solution with STUN/ICE/FS but without TURN: is possible?
by Fabio Pietrosanti (naif) 14 Jun '10

14 Jun '10

Hi all, i need to setup a server that handle VoIP connection in a way that always works in the most efficient way respect to latency and NAT traversal. FS provide very good NAT detection system but unfortunately when it fail, it just fail and does not have any kind of fallback mechanism. If FS do bypass-media and the client cannot communicate because the phone call is lost. If FS do proxy-media, the clients can always communicate (over UDP) but there is added latency. So we cannot rely only on FS NAT detection system and need also some client side system. Now, in theory the best approach would be not to touch FS but just use client-side system based on ICE/STUN with a TURN server-side component as a fallback. However there's no widely available TURN server, or at least TURN is not widely and well diffused just now. I am wondering whether it may be possible to make a setup where: - Client try to establish peer to peer connectivity with STUN/ICE (so FS is doing bypass-media not touching the SDP) - If they cannot work, will go in fallback to FS (but with proxy-media) instead than going via TURN So using the FS B2BUA proxy-media feature as alternative to TURN server, but letting clients to dynamically try to connect directly one each other by using the ICE methods that are dynamic. Does it seems something feasible? Fabio Pietrosanti

3 3