January 2017 - FreeSWITCH-users - FreeSWITCH Mailing Lists

ICTFax Version 3.7.4 released
by Tahir Almas 28 Jan '17

28 Jan '17

Pleased to update about the release of ICTFax Version 3.7.4 using freeswitch based ICTCore communication framework http:/ictcore.org Regards *Tahir Almas* Managing Partner ICT Innovations http://www.ictinnovations.com Leveraging open source in ICT

1 1

Dial Plan tidbit for a HOTLINE automatic dial
by Sean Devoy 27 Jan '17

27 Jan '17

I have never posted here with information, only enquiries. Based on a posting I made a few years ago, no one had a solution "A HOTLINE" that dials automatically. Essentially, I learned FS gets no events for a phone going off hook so it is really a phone function not a switch function (on VOIP devices). I have recently discovered a dial plan to achieve it. It may be obvious to many or have been posted here before, but I don't see it in any quick searches. Cisco/Linksys phones and ATAs conform to the standards specified by MGCP and MEGACO. I suspect most venders do, so, this is likely to work for most everyone. There is a dial plan feature "<: xxx >" that allows you to insert additional digits when sent to the switch but not be required in the dial plan match. Seven digit dialing in the U.S. is a good example. To insert 1 + area code (410 for example), use "<:1410>[2-9]xxxxxx S0" Note also the S0 which will send to the switch immediately if matched. Extending that concept to somewhat of an extreme "S0 <:18005551212>" will dial that inserted number as soon as the phone is taken off hook, Another option is to extend the timeout to allow the user to dial, but dial the hot line if the user does nothing. "S3 <:18005551212>" I found it interesting and helpful. Hope it helps someone else to0. If it is old news, sorry for wasting your time, Sean

1 0

should FS put/remove the call on hold on update request
by Antonio Silva 27 Jan '17

27 Jan '17

Hi all, i've seem some dialog with avaya device and i'm not sure if the behaviour is the correct from fs on update request. I have: FS <--------> remote end invite-----> <---- trying <----- 200ok <----- invite (a=sendonly) 200ok (a=recvonly) ----- > <----- update (a=sendrecv) (1)200 ok ----> <---- bye ack ----> in (1) FS do not change the call flow and it stays in hold. This is the correct behaviour? should it process the update and remove the call from hold, responding wih 200 ok (a=sendrecv) ? I could file a jira if you think is a bad behaviour.. Normally, instead of the update would expect an invite (a=sendrecv)... so i'm pretty sure that it could be a bug in the avaya device. Thanks, António

2 4

Bridge with SIP URI Passed Through
by Lewis Green 27 Jan '17

27 Jan '17

Hi, I have a FreeSwitch setup where the extensions in the system can be either remote PBX’s or remote extensions. If a call comes from the outside trunk to the FreeSwitch box it has the DDI in the SIP Request URI, we then match it in a dial plan and bridge it. This works fine for the extensions remotely but the remote PBX’s are not getting the DDI passed through in the bridged path of the call. Is there any way to pass this through? Thanks

2 1

call recording wav vs mp3
by Tihomir Culjaga 27 Jan '17

27 Jan '17

Hello, im wondering what is the performance impact if FS records calls in mp3 directly opposed to recording in pcm? does anyone do it in production ?

2 2

Calls getting stuck in playback application playing a tone_stream
by Luke Wahlmeier 27 Jan '17

27 Jan '17

I am having issues with calls getting stuck in the playback application. This seems to happen only when the call is executing javascript with mod_v8 and the javascript calls some kind of sound playback on the channel. We are currently on freeswitch 1.6.12-b91a0a6 on 64bit linux Debian Jessie, Linux kernel 4.4.6. Here is what a stuck call looks like: bef33c54-66f2-4183-b281-35b5b9ade65f,inbound,2017-01-25 13:05:03,1485374703,sofia/external/###########@###.##.##.##,CS_EXECUTE,###########,###########,192.168.24.77,###########,playback,tone_stream://v=-7;>=10;+=.1;%(750,0,293.6,370.0,440.0);v=-7;>=5;+=.1;%(1500,250,293.6,370.0,440.0),XML,public,PCMU,8000,64000,PCMU,8000,64000,,laxvoip6,,,,ACTIVE,,,,,,,###########,###########,192.168.24.77,###########,XML,public The most basic form of the javascript that causes this looks like this: if ( session.ready() ) { session.answer(); var dtmf = session.getDigits(30, "*", 20*1000); var otherUUID = getDTMF(dtmf); session.execute("playback", "tone_stream://v=-7;>=10;+=.1;%(750,0,293.6,370.0,440.0);v=-7;>=5;+=.1;%(1500,250,293.6,370.0,440.0)"); } 99.9% of the calls that go through this path work fine, but about 1 in 1000 get stuck in this state. And once this happens all messages from the far side are ignored, including the BYE, and the media never times out even though there is no data being transmitted anymore. I have been able to get a core dump and this looks like the offending threads bt: #0 0x00007f0187361a9d in read () at ../sysdeps/unix/syscall-template.S:81 81 ../sysdeps/unix/syscall-template.S: No such file or directory. (gdb) bt #0 0x00007f0187361a9d in read () at ../sysdeps/unix/syscall-template.S:81 #1 0x00007f0187ef25b6 in _timerfd_next (timer=0x1bbdeb8) at src/switch_time.c:474 #2 timer_next (timer=0x1bbdeb8) at src/switch_time.c:838 #3 0x00007f0187e27e78 in switch_core_timer_next (timer=<optimized out>) at src/switch_core_timer.c:74 #4 0x00007f00fc1f07f4 in read_stream_thread (thread=0x51, obj=0x7f00fc1364a0) at mod_local_stream.c:445 #5 0x00007f018811a840 in dummy_worker (opaque=0x1bbea40) at threadproc/unix/thread.c:151 #6 0x00007f018735b064 in start_thread (arg=0x7f00fc137700) at pthread_create.c:309 #7 0x00007f0186a3362d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111 (gdb) thread 46 [Switching to thread 46 (Thread 0x7f00db373700 (LWP 9663))] #0 0x00007f0187361e03 in recvfrom () at ../sysdeps/unix/syscall-template.S:81 81 ../sysdeps/unix/syscall-template.S: No such file or directory. (gdb) bt #0 0x00007f0187361e03 in recvfrom () at ../sysdeps/unix/syscall-template.S:81 #1 0x00007f01881184b5 in apr_socket_recvfrom (from=from@entry=0x7f015d2eb8c0, sock=0x7f015d2ebb30, flags=0, buf=0x7efd6a066cf0 "\200", len=0x7f00db36e230) at network_io/unix/sendrecv.c:155 #2 0x00007f0187e0f829 in switch_socket_recvfrom (from=0x7f015d2eb8c0, sock=<optimized out>, flags=<optimized out>, buf=<optimized out>, len=<optimized out>) at src/switch_apr.c:946 #3 0x00007f0187ea6198 in do_flush (force=<optimized out>, bytes_in=139644540165635, bytes_in@entry=88, rtp_session=<optimized out>) at src/switch_rtp.c:5111 #4 rtp_common_read (rtp_session=0x7efd6a05eb58, payload_type=0x7efd6a066cf0 "\200", payload_type@entry=0x7f015d2dc984 "", pmapP=0x4020, pmapP@entry=0x7f015d2dc9a8, flags=0x7f0187361e03 <recvfrom+51>, flags@entry=0x7f015d2dc998, io_flags=2) at src/switch_rtp.c:6861 #5 0x00007f0187ea6cef in switch_rtp_zerocopy_read_frame (rtp_session=0x7efd6a05eb58, frame=0x7f015d2dc940, io_flags=2) at src/switch_rtp.c:7304 #6 0x00007f0187e5b023 in switch_core_media_read_frame (session=0x7f015d2d32d8, frame=0x7f015d2dc2e8, flags=1563281800, stream_id=-2026496509, stream_id@entry=0, type=SWITCH_MEDIA_TYPE_AUDIO) at src/switch_core_media.c:2184 #7 0x00007f016b1830a5 in sofia_read_frame (session=0x7f015d2d32d8, frame=0x7f00db370760, flags=2, stream_id=0) at mod_sofia.c:1018 #8 0x00007f0187e45b2d in switch_core_session_read_frame (session=0x7f015d2d32d8, frame=0x7f00db370760, flags=16416, stream_id=-2026496509) at src/switch_core_io.c:175 #9 0x00007f0187ed133a in switch_ivr_play_file (session=0x7f015d2d32d8, fh=0x7f00db370f40, file=0x7f00db370f00 "\300\004\363h\001\177", args=0x7f0187361e03 <recvfrom+51>, args@entry=0x7f00db370f00) at src/switch_ivr_play_say.c:1822 #10 0x00007f0168f2a7b3 in playback_function (session=0x7f015d2d32d8, data=<optimized out>) at mod_dptools.c:2831 #11 0x00007f0187e4036a in switch_core_session_exec (session=0x7f015d2d32d8, application_interface=0x1abb0b8, arg=0x7efd69adf478 "tone_stream://v=-7;>=10;+=.1;%(750,0,293.6,370.0,440.0);v=-7;>=5;+=.1;%(1500,250,293.6,370.0,440.0)") at src/switch_core_session.c:2784 #12 0x00007f00e3bdc1c8 in FSSession::ExecuteImpl (this=0x7efd5d9ca280, info=...) at src/fssession.cpp:1391 #13 0x00007f00e3be03cd in FSSession::Execute (info=...) at ./include/fssession.hpp:125 #14 0x00007f00e347a619 in v8::internal::FunctionCallbackArguments::Call(void (*)(v8::FunctionCallbackInfo<v8::Value> const&)) () from /usr/local/freeswitch/lib/libv8.so #15 0x00007f00e3495b08 in v8::internal::Builtin_HandleApiCall(int, v8::internal::Object**, v8::internal::Isolate*) () from /usr/local/freeswitch/lib/libv8.so Has anyone seen this before? Let me know if there is any other information needed. Thanks Luke

3 3

How to use Particular Interface
by Murugan Pandian 26 Jan '17

26 Jan '17

Hi, I have two interface in my server server one for Internet(eth0) another one for SIP Trunk(eth1) when i start freeswitch it always bing eth0 and eth1 ,I need freeswitch bind only one Interface eth1,It's possible in freeswitch ? Regard's Murugan Pandian

2 2

Random number generation
by Chris Young 26 Jan '17

26 Jan '17

Hello all, On the wiki, there is an example of using mod_expr to generate a random number:  <action application="set" data="rand_val=${expr(randomize(&x);ceil(random(0,100,&x)))}"/> <action application="log" data="INFO Random value is ${rand_val}"/> This works well the first time but if the same dialplan instructions are executed subsequently, the same number is produced each time. Is this by design? Is there any other way to seed the random number generator so it will produce different results, as randomize() doesn't seem to be doing quite what I thought it would? Many thanks, Chris Chris Young Software Engineer [cid:image7482a0.PNG@dc00f514.4fb88205] t: +44 118 943 9249 e: chris.young(a)enghouse.com<mailto:chris.young@enghouse.com> w: www.enghouseinteractive.co.uk<http://www.enghouseinteractive.co.uk> [cid:image6c1ba6.PNG@b8963a35.449174f5] Enghouse Interactive (UK) Ltd is a company registered in England and Wales. Registered number: 04230977. Registered office: Imperium, Imperial Way, Reading, Berkshire, RG2 0TD

4 11

Increase response time for sip request
by devang nathwani 25 Jan '17

25 Jan '17

Hello, I want to increase response time, Like how log FreeSWITCH will wait for response on a SIP request before sending a new request? Is there any params, does ptime will work for that? any other link or suggestion?

2 1

Freeswitch H26x negotiation problem
by Sina Aghasi 25 Jan '17

25 Jan '17

Hi I using FS 1.6 . I've compiled it source codes. There is no problem when doing a video call using VP8 codec, but when I restrict the clients to use H263 or H264 codecs, no video is shown. After investigating FS logs and network packets, i found out that clients request H263 or H264 but FS does not negotiate it! Following lines show that FS only compare H264 with VP8: 2017-01-25 05:58:07.132032 [DEBUG] switch_core_media.c:4843 Video Codec Compare [H264:108]/[VP8:99] 2017-01-25 05:58:07.132032 [DEBUG] switch_core_media.c:4843 Video Codec Compare [H264:99]/[VP8:99] Although H264 is loaded as "show codec" output is: ... codec,GSM,mod_spandsp codec,H.261 Video (passthru),mod_h26x codec,H.263 Video (passthru),mod_h26x codec,H.263+ Video (passthru),mod_h26x codec,H.263++ Video (passthru),mod_h26x codec,H.264 Video (passthru),mod_h26x ... codec,VP8 Video,CORE_VPX_MODULE codec,VP9 Video,CORE_VPX_MODULE 30 total. This problem occurs in video conference call and point to point call. Any solution?

3 2