June 2018 - FreeSWITCH-users - FreeSWITCH Mailing Lists

Wrong IP in ACK
by Sergey Safarov 21 Jun '18

21 Jun '18

Are anybody experiences issue when for some of calls ACK send to other IP. http://prntscr.com/jtxkbz in this case FS in bypass media with two profiles (1 to internet, 2 intranet). INVITE is send to DNS name that is resolved to 1. 10.0.9.32 2. 10.0.9.33 3. 10.0.9.34 4. 10.0.9.35 FS 1.6.19 Contact on 200 message is correct SIP/2.0 200 OK Via: SIP/2.0/TCP 10.0.9.51:12000;rport=39049;branch= z9hG4bKBH60NQjSZj36a From: "Anonymous" <sip:+12022553656@10.0.9.51>;tag= rH5cBQarr559D To: <sip:13013688630@metropolis.ack.one>;tag=UDv7UXv3B53Da Call-ID: b0e4d7e1-e79e-1236-8cb9-02420a000933 CSeq: 123998414 INVITE Contact: <sip:13013688630@10.0.9.33:11000;transport=tcp> User-Agent: proxy.ackone.com Allow: INVITE, ACK, BYE, CANCEL, OPTIONS, MESSAGE, INFO, UPDATE, REGISTER, REFER, NOTIFY, PUBLISH, SUBSCRIBE Now for future investigate FS upgrader to 1.6.20 and enabled siptrace and loglevel Sergey Safarov

2 6

Enterprise/Production Quality?
by Geoff Mina 20 Jun '18

20 Jun '18

Is anyone out there actually using FS successfully in an enterprise environment? We have deployed a handful of servers in an extremely simple configuration to allow standard SIP infrastructure to communicate with WebRTC clients. We run ~150 concurrent users per host and we can’t go a week without something in the core of Sofia failing. We have seen hung profiles that simply don’t respond to REGISTER requests (tried every suggested tweak to no avail) as well as seeing FS hang every call in a RINGING state without actually ringing the end client. These seem like pretty fundamental components we are struggling with. Our use case seems quite simple - yet the software (1.6.19) seems like it has never even seen a production deployment. Anyone out there have a drastically different experience?

10 14

Re: [Freeswitch-users] DNS cache on invite/register
by Marcin Kowalczyk 20 Jun '18

20 Jun '18

Hi, I agree they should have shared nonce, but still I think FS should use same IP for initial and challenged invite unless TTL for domain is not expired. Is there a way to force it to such behaviour? I've tried with fs_path but seems this does not work for challenged invite neither Somebody posted bug FS-8820 on same case but with different carrier. Regards Date: Tue, 19 Jun 2018 14:26:53 -0400 > if they are doing nonces on a shared realm, they should be sharing those > nonces across their cluster. If they are not, every time it hits a diff > member of the cluster you will get a stale nonce and have to go through > that juggle again. This is just a side effect of how sip auth works. > > > On Jun 19, 2018, at 12:50 PM, Marcin Kowalczyk <kowalma(a)gmail.com> wrote: > > Hi, > > I'm having problem setting-up trunk with provider who does have muliple > sip-proxies with not shared nonce db. > > sip.freeconet.pl resolves to 3 diffrent IP and FS does not cache/remeber > IP that send back request. > > So we have a case > FS->P1: Invite > P1 -> FS -> 407 + nonce > FS does again DNS lookup and > FS->P2 Invite with nonce from P1 > P2->FS 407 > FS fails call. > > Is there a way to force FS to send all messages within dialog to one IP > only ? > > Regards > > >

1 0

Recording skipping silence on outbound call recording
by Andrew Keil 20 Jun '18

20 Jun '18

To FreeSWITCH Users, I managed to find and fix the issue and thought I would provide my solution here since it may help someone else. Prior to the recording I set the following inside my outbound Lua script: session:setVariable("record_fill_cng","1400") Note: 1400 matches the waste_resources default setting inside the switch_ivr_record_file (switch_ivr_play_say.c) This fixed the issue and now my recording contains the silence periods correctly (which is what I was after). I also noticed that since the Lua destination script that my outbound was calling contains periods of session:sleep(...), when waiting for DTMF, and it was this period that was not being recorded. Now with this fix everything is fine and those periods of session:sleep(...) are matched inside my outbound recorded audio file. Andrew --- My original e-mail to FreeSWITCH Users --- To FreeSWITCH Users, I am attempting to make a recording of an outbound call using the originate API call which starts a Lua script which then simply records to disk exactly what it hears. I just noticed an interesting issue (or feature) where the recording skips any silence. What I am aiming for is for no silence to be skipped inside the recording, if someone can provide some settings for me to try I would be more than happy to experiment further. The settings for the Lua session:recordFile are length of recording: 60 seconds; silence threshold: 300; silence seconds: 60 seconds {essentially termination on silence is disabled since the silence seconds = recording seconds} I also tried with silence threshold = 0 and silence is still skipped inside the recording. I also tried the setting inside the appropriate profile: <param name="suppress-cng" value="true"/> and this made no difference, silence is still skipped inside the recording. My setup is one single FreeSWITCH 1.6.20 (production version) box with a simple internal gateway setup to land on the same FreeSWITCH IVR (just for testing). API command: originate {origination_uuid=121e20a6-1245-439a-8527-53efab009334-out-1,origination_caller_id_number=FLOODOUT,call_timeout=15,ignore_early_media=true,return_ring_ready=true,suppress_cng=true}sofia/gateway/visytel-pc-ivr/01111111111 FLOODOUT The SDP inside my log is similar to this for the outbound and inbound (Note: The outbound call is performing the recording): 121e20a6-1245-439a-8527-53efab009334-out-1 Local SDP: 121e20a6-1245-439a-8527-53efab009334-out-1 v=0 121e20a6-1245-439a-8527-53efab009334-out-1 o=FreeSWITCH 1529274701 1529274702 IN IP4 192.168.15.15 121e20a6-1245-439a-8527-53efab009334-out-1 s=FreeSWITCH 121e20a6-1245-439a-8527-53efab009334-out-1 c=IN IP4 192.168.15.15 121e20a6-1245-439a-8527-53efab009334-out-1 t=0 0 121e20a6-1245-439a-8527-53efab009334-out-1 m=audio 25994 RTP/AVP 9 8 0 101 121e20a6-1245-439a-8527-53efab009334-out-1 a=rtpmap:9 G722/8000 121e20a6-1245-439a-8527-53efab009334-out-1 a=rtpmap:8 PCMA/8000 121e20a6-1245-439a-8527-53efab009334-out-1 a=rtpmap:0 PCMU/8000 121e20a6-1245-439a-8527-53efab009334-out-1 a=rtpmap:101 telephone-event/8000 121e20a6-1245-439a-8527-53efab009334-out-1 a=fmtp:101 0-16 121e20a6-1245-439a-8527-53efab009334-out-1 a=silenceSupp:off - - - - 121e20a6-1245-439a-8527-53efab009334-out-1 a=ptime:20 121e20a6-1245-439a-8527-53efab009334-out-1 a=sendrecv 6da97c7c-1052-4f78-8dde-83901147a0c1 2018-06-18 15:44:55.596296 [DEBUG] sofia.c:7094 Remote SDP: 6da97c7c-1052-4f78-8dde-83901147a0c1 v=0 6da97c7c-1052-4f78-8dde-83901147a0c1 o=FreeSWITCH 1529274701 1529274702 IN IP4 192.168.15.15 6da97c7c-1052-4f78-8dde-83901147a0c1 s=FreeSWITCH 6da97c7c-1052-4f78-8dde-83901147a0c1 c=IN IP4 192.168.15.15 6da97c7c-1052-4f78-8dde-83901147a0c1 t=0 0 6da97c7c-1052-4f78-8dde-83901147a0c1 m=audio 25994 RTP/AVP 9 8 0 101 6da97c7c-1052-4f78-8dde-83901147a0c1 a=rtpmap:9 G722/8000 6da97c7c-1052-4f78-8dde-83901147a0c1 a=rtpmap:8 PCMA/8000 6da97c7c-1052-4f78-8dde-83901147a0c1 a=rtpmap:0 PCMU/8000 6da97c7c-1052-4f78-8dde-83901147a0c1 a=rtpmap:101 telephone-event/8000 6da97c7c-1052-4f78-8dde-83901147a0c1 a=fmtp:101 0-16 6da97c7c-1052-4f78-8dde-83901147a0c1 a=silenceSupp:off - - - - 6da97c7c-1052-4f78-8dde-83901147a0c1 a=ptime:20 6da97c7c-1052-4f78-8dde-83901147a0c1 Kind Regards, Andrew Keil

1 0

DNS cache on invite/register
by Marcin Kowalczyk 19 Jun '18

19 Jun '18

Hi, I'm having problem setting-up trunk with provider who does have muliple sip-proxies with not shared nonce db. sip.freeconet.pl resolves to 3 diffrent IP and FS does not cache/remeber IP that send back request. So we have a case FS->P1: Invite P1 -> FS -> 407 + nonce FS does again DNS lookup and FS->P2 Invite with nonce from P1 P2->FS 407 FS fails call. Is there a way to force FS to send all messages within dialog to one IP only ? Regards

2 1

gateway selection by prefix
by Francesco Facco de Lagarda 19 Jun '18

19 Jun '18

I have 3 gateways: Telecom Italia on an analogue adapter MessageNet (sip provider) Skype How can I select gateway by a prefix THAT THEN GETS DROPPED before Dialing.. Example I would like to have All calls beginning with 9 Sent to MessageNet , DROPPING THE "9" All calls beginning with 7 sent to Skype, dropping the "7" Otherwise use telecom

2 1

Sending SIP Messages from LUA
by Ronnie Beck 19 Jun '18

19 Jun '18

Hi All, I am trying to write a LUA script which will slot into a chatplan whose goal is to simple take a simple text message (SIMPLE MESSAGE) and then distribute it to multiple recipients. I wrote a script which takes some arguments (sender, recipieant list and the body) then would generate new SIP Messages to be sent to each of the recipient. The code I use to generate the message (which is more or less the example given in mod_sms for sending an SMS, not exactly what I want to do): local event = freeswitch.Event("CUSTOM", "SMS::SEND_MESSAGE"); event:addHeader("proto", "global"); event:addHeader("dest_proto", "sip"); event:addHeader("from", source); event:addHeader("to", domain.."/"..recipient); event:addHeader("type", "text/plain"); event:addHeader("skip_global_process", "true"); event:addBody( body ); event:fire(); This gives the error: [WARNING] sofia_presence.c:221 Not sending to local box for XXXX(a)10.11.12.13 [ERR] sofia_presence.c:272 Chat proto [sip] from [<sip:YYYY@10.11.12.13>;tag=uUQudJKEl1SKjq8wdjqwi68768] to [XXXX(a)10.11.12.13] Some message text Nobody to send to: Profile internal But the message actually arrives successfully. This is fine but I would prefer not to have lots of error messages. So I tried to find a more correct way to do this. I tried the following (based on what I could find in freeswitch 1.6.20 source code): local event = freeswitch.Event("SWITCH_EVENT_SEND_MESSAGE"); event:addHeader("user", recipient ); event:addHeader("host", domain ); event:addHeader("profile", domain ); event:addHeader("subject", "SIMPLE MESSAGE"); event:addHeader("content-type", "text/plain"); event:addHeader("from", source); event:addBody( body ); event:fire(); But this gives the same error. How can I send a simple SIP Message with LUA to a locally registered user? Many thanks, Aaron

2 1

Memory leak while using BGAPI
by Mody, Darshan (Darshan) 19 Jun '18

19 Jun '18

Hi, We are observing considerable memory leak while using LUA and BGAPI of mod_command.c. Below is the valgrind's output ==21863== 123,170 (45,496 direct, 77,674 indirect) bytes in 517 blocks are definitely lost in loss record 2,370 of 2,398 ==21863== at 0x4C29C23: malloc (vg_replace_malloc.c:299) ==21863== by 0x548707C: switch_event_create_subclass_detailed (switch_event.c:736) ==21863== by 0x11631389: bgapi_exec (mod_commands.c:5151) ==21863== by 0x56EEE8F: dummy_worker (thread.c:151) ==21863== by 0x7677E24: start_thread (in /usr/lib64/libpthread-2.17.so) ==21863== by 0x805734C: clone (in /usr/lib64/libc-2.17.so ==21863== 143,478 (18,392 direct, 125,086 indirect) bytes in 209 blocks are definitely lost in loss record 2,374 of 2,398 ==21863== at 0x4C29C23: malloc (vg_replace_malloc.c:299) ==21863== by 0x548707C: switch_event_create_subclass_detailed (switch_event.c:736) ==21863== by 0x542AADE: switch_core_session_exec (switch_core_session.c:2874) ==21863== by 0x542B108: switch_core_session_execute_application_get_flags (switch_core_session.c:2742) ==21863== by 0x5510C10: CoreSession::execute(char const*, char const*) (switch_cpp.cpp:777) ==21863== by 0x16FED72C: ??? (mod_lua_wrap.cpp:7289) ==21863== by 0x17217323: ??? (in /usr/lib64/liblua-5.1.so) ==21863== by 0x17221E56: ??? (in /usr/lib64/liblua-5.1.so) ==21863== by 0x1721774C: ??? (in /usr/lib64/liblua-5.1.so) ==21863== by 0x17216A6D: ??? (in /usr/lib64/liblua-5.1.so) ==21863== by 0x172178D9: ??? (in /usr/lib64/liblua-5.1.so) ==21863== by 0x1721344C: lua_pcall (in /usr/lib64/liblua-5.1.so) Has some one also observed behavior while using BGAPI? Thanks Darshan

3 4

Pyrun limitations
by Grant Bagdasarian 19 Jun '18

19 Jun '18

Hello, We're currently using the NEventSocket library to control our freeswitch servers for starting python scripts (using a backgroundjob and pyrun py_module_name) which internally use the freeswitch.py library to originate a call. However, we're running into some performance limitations where we can't seem to get more than 5 call setups per second. To quote the documentation (https://freeswitch.org/confluence/display/FREESWITCH/mod_python) "A single python interpreter is spawned at module startup and used for the lifetime of the freeswitch process." "Each thread swaps in its "thread state" before executing python code and then swaps it out when finished. Also during blocking calls into freeswitch, a thread will swap out its thread state in order to not block other threads, and then swap it in after the blocking call to freeswitch has finished." How does this relate to the number of threads freeswitch is allowed to start and run python scripts? If there are hard limitations set, can these be increased? I couldn't find hard numbers in the source files, so this may be controlled somewhere else? Thanks. Grant Bagdasarian Senior Developer cm.com<https://cm.com>

2 3

No T38 SDP in 200 OK after glare
by Douglas Davenport 19 Jun '18

19 Jun '18

After T38 reinvite glare on an inbound call, the provider resends their reinvite after a random delay (they own the call-ID). Freeswitch correctly responds with 200 OK because fax_enable_t38=true however it does not include SDP. Are there any settings that would affect this behavior or is this a bug? I'm aware that bug reports go to JIRA but I'm asking first to know if this is actually a bug or I'm doing something wrong. Here is the call flow. Note the starred 200 OK with no SDP: Provider Freeswitch ──────────┬───────── ──────────┬───────── 12:07:02.067864 │ ─────── INVITE (SDP) ─────> │ 12:07:02.068654 │ <─────── 100 Trying ─────── │ 12:07:03.304619 │ <────── 200 OK (SDP) ────── │ 12:07:03.305554 │ ──────────── ACK ─────────> │ 12:07:06.506575 │ ────── INVITE (T38 SDP)───> │ 12:07:06.718790 │ <────── INVITE (T38 SDP)─── │ 12:07:06.719063 │ ───── 491 Request Pendi ──> │ 12:07:06.879340 │ <─────── 100 Trying ─────── │ 12:07:06.881990 │ <─────────── ACK ────────── │ 12:07:06.943138 │ <──── 491 Request Pendi ─── │ 12:07:06.943356 │ ──────────── ACK ─────────> │ 12:07:07.133006 │ <────INVITE (T38 SDP) ───── │ 12:07:07.133313 │ ──────── 100 Trying ──────> │ 12:07:07.146828 │ ─*************─ 200 OK ─[*NO SDP*]─> │ ***************** 12:07:07.280018 │ <─────────── ACK ────────── │ 12:07:16.012689 │ <─────────── BYE ────────── │ 12:07:16.026995 │ ────────── 200 OK ────────> │

2 3