- FreeSWITCH-users - FreeSWITCH Mailing Lists

Re: [Freeswitch-users] rtp-timeout-sec VS media_timeout
by David P 18 Mar '21

18 Mar '21

Although I didn't see that adding... media_timeout=300000 ...to our conference settings have any effect after restarting FS10.5 this way... fs_cli -x 'fsctl shutdown elegant restart' ...I do see it having an effect now. I suspect there's been a reboot since then. FYI On Sun, Mar 14, 2021 at 1:00 AM < freeswitch-users-request(a)lists.freeswitch.org> wrote: > > ---------- Forwarded message ---------- > From: David P <davidswalkabout(a)gmail.com> > To: FreeSWITCH Users Help <freeswitch-users(a)lists.freeswitch.org> > Cc: > Bcc: > Date: Sat, 13 Mar 2021 16:15:55 +1300 > Subject: Re: [Freeswitch-users] rtp-timeout-sec VS media_timeout > Using FS10.5 installed from packages on Debian Stretch... > > I wanted to replicate the behavior we had with <param name="rtp-timeout-sec" > value="300"/> so that we'll hangup calls in which leg B of a conference has > sent no media for 5 mins. > > I edited our dialplan like this... > > <action application="conference_set_auto_outcall" > > data="['conference_member_flags=endconf,jitterbuffer_msec=5p:100p,media_timeout=300000']sofia/gateway/...deleted..."/> > > and restarted like this... > > fs_cli -x 'fsctl shutdown elegant restart' > > But after 8 mins in such a call there was no hangup. >

2 1

Will fail2ban work for this?
by Steven Schoch 17 Mar '21

17 Mar '21

I just set up a new FreeSWITCH system on my home network, and set a forward for port 5080 to connect to Flowroute. While I'm debugging some call routing stuff, my logs are getting overrun with stuff like this: 2021-03-16 15:52:02.267501 [NOTICE] switch_channel.c:1118 New Channel sofia/external/7750@<my IP> [2de89b87-cd07-4c0f-b9fb-3da8e5a68d37] 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:585 (sofia/external/7750@<my IP>) Running State Change CS_NEW (Cur 1 Tot 7822) 2021-03-16 15:52:02.267501 [DEBUG] sofia.c:10280 sofia/external/7750@<my IP> receiving invite from 80.94.93.12:62635 version: 1.10.5 -release-17-25569c1631 64bit 2021-03-16 15:52:02.267501 [DEBUG] sofia.c:7326 Channel sofia/external/7750@<my IP> entering state [received][100] 2021-03-16 15:52:02.267501 [DEBUG] sofia.c:7336 Remote SDP: v=0 o=- 81921704 81921704 IN IP4 0.0.0.0 s=pplsip c=IN IP4 0.0.0.0 t=0 0 m=audio 7628 RTP/AVP 100 6 0 8 3 18 5 101 a=rtpmap:100 speex/16000 a=rtpmap:101 telephone-event/8000 a=fmtp:101 0-11 a=alt:1 1 : DF50DC48 0000001F 0.0.0.0 7628 2021-03-16 15:52:02.267501 [DEBUG] sofia.c:7739 (sofia/external/7750@<my IP>) State Change CS_NEW -> CS_INIT 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:604 (sofia/external/7750@<my IP>) State NEW 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:585 (sofia/external/7750@<my IP>) Running State Change CS_INIT (Cur 1 Tot 7822) 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:628 (sofia/external/7750@<my IP>) State INIT 2021-03-16 15:52:02.267501 [DEBUG] mod_sofia.c:93 sofia/external/7750@<my IP> SOFIA INIT 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:40 sofia/external/7750@<my IP> Standard INIT 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:48 (sofia/external/7750@<my IP>) State Change CS_INIT -> CS_ROUTING 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:628 (sofia/external/7750@<my IP>) State INIT going to sleep 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:585 (sofia/external/7750@<my IP>) Running State Change CS_ROUTING (Cur 1 Tot 7822) 2021-03-16 15:52:02.267501 [DEBUG] switch_channel.c:2332 (sofia/external/7750@<my IP>) Callstate Change DOWN -> RINGING 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:644 (sofia/external/7750@<my IP>) State ROUTING 2021-03-16 15:52:02.267501 [DEBUG] mod_sofia.c:154 sofia/external/7750@<my IP> SOFIA ROUTING 2021-03-16 15:52:02.267501 [DEBUG] switch_core_state_machine.c:236 sofia/external/7750@<my IP> Standard ROUTING 2021-03-16 15:52:02.267501 [INFO] mod_dialplan_xml.c:637 Processing 7750 <7750>->900442037697855 in context public I thought fail2ban was designed for stuff like this, but I don't see any auth attempts here (I set "log-auth-failures" to "true"). These are coming in a bit faster than 1 per second. It appears they are dialing random extensions. How can I make them stop? -- Steve

6 6

Faxing with FreeSWITCH
by bruce＠sqls.net 16 Mar '21

16 Mar '21

Hello! I'm working on setting up a FreeSWITCH/Flowroute fax server. The server is connected without any nat on a 20MB fiber internet connection. I've built some lua scripts that allow me to receive and send faxes and log all the faxes to a sqlite database. I even have a management script that can be called from the fs_cli to add fax2email routes and adjust settings. It's still a pretty rough implementation but overall it's working fairly well but I have two main questions that I hope someone could help me with. 1) I would like to detect "voice calls" or maybe I want to detect fax calls? There's a dialplan function to listen for and detect a fax caller but I'm having trouble implementing it in my lua script. It seems it's a non-blocking function, maybe? Because it just walks right over that line instead of waiting and listening for any fax tones. I thought it would block the script for x seconds and listen for a fax tone and return a result based on what it found. Can someone help me accomplish something similar to that? 2) So far, by far... my most common error is "The call dropped prematurely". I'm not sure how to start debugging the error to figure out what's causing it and how (if I can) solve it. I do have pcap's of the calls from pcapsipdump and I also have the freeswitch log of each call (my lua script pulls and saves that for each fax call). I've opened the pcap files up in wireshark but I don't know what I'm looking for and I think that might be over my head at the moment. Can anyone give me any help on what this error is normally caused by or what I could look for and/or test? Here are a couple of example logs that ended with "The call dropped prematurely". http://bmts.us/faxlogs/FAX-2e913a50-3a4f-4abc-9882-a323a97e5138.log http://bmts.us/faxlogs/FAX-59d02d9d-c4ba-44c2-8edd-a1c9142c7c1c.log

6 11

Opus packet loss
by Terry C 15 Mar '21

15 Mar '21

Hi, We use FreeSwitch 1.10.5 with Opus audio for clients. We target voice quality to be wideband. The config settings are below. We also set a jitter buffer in the dialplan. Is there a downside to increasing the packet loss percent to 40-50% other than using more bandwidth? Is there a voice quality impact if you set it too high (what would be considered too high)? Thanks. <param name="use-vbr" value="1"/> <param name="use-dtx" value="1"/> <param name="complexity" value="10"/> <param name="packet-loss-percent" value="30"/> <param name="keep-fec-enabled" value="1"/> <param name="use-jb-lookahead" value="1"/> <param name="maxaveragebitrate" value="40000"/> <param name="maxplaybackrate" value="16000"/> <param name="sprop-maxcapturerate" value="16000"/> <param name="adjust-bitrate" value="1"/> </settings>

4 3

Re: [Freeswitch-users] rtp-timeout-sec VS media_timeout
by David P 13 Mar '21

13 Mar '21

Using FS10.5 installed from packages on Debian Stretch... I wanted to replicate the behavior we had with <param name="rtp-timeout-sec" value="300"/> so that we'll hangup calls in which leg B of a conference has sent no media for 5 mins. I edited our dialplan like this... <action application="conference_set_auto_outcall" data="['conference_member_flags=endconf,jitterbuffer_msec=5p:100p,media_timeout=300000']sofia/gateway/...deleted..."/> and restarted like this... fs_cli -x 'fsctl shutdown elegant restart' But after 8 mins in such a call there was no hangup. ---------- Forwarded message ---------- From: Dragos Oancea <dragos(a)freeswitch.org> To: FreeSWITCH Users Help <freeswitch-users(a)lists.freeswitch.org> Cc: Bcc: Date: Thu, 4 Mar 2021 11:04:26 +0200 Subject: Re: [Freeswitch-users] rtp-timeout-sec VS media_timeout For media timeout there are the following chan vars: media_timeout, media_hold_timeout, media_timeout, media_hold_timeout_video, media_hold_timeout_audio, media_timeout_audio . They are in milliseconds, not seconds like rtp-timeout-sec . <param name="media_timeout" value="300000"/>

1 0

Un-Subscribing on 481
by Joseph Barrero 12 Mar '21

12 Mar '21

Hi, everyone. I recently noticed that Freeswitch continues sending NOTIFY messages even after receiving a 481 from my proxy after not being able to deliver the messages to the UA. This seems to happen when the UA disconnects from the Internet briefly and registers again before the previous subscription expires. Below is the code I found in sofia.c that seems to handle the 481 response. if (status == 481 && sip && !sip->sip_retry_after && sip->sip_call_id && (!sofia_private || !sofia_private->is_call)) { char *sql; sql = switch_mprintf("delete from sip_subscriptions where call_id='%q'", sip->sip_call_id->i_id); switch_assert(sql != NULL); sofia_glue_execute_sql(profile, &sql, SWITCH_TRUE); nua_handle_destroy(nh); } Assuming that the variable sofia_private->is_call evaluates to true when NOTIFY messages are sent during a call, it seems to curtail the removal of the SUBSCRIPTION even after receiving a 481 from the proxy. Is there a drawback to removing a subscription on a 481 response during a call? Would it make sense to simply remove the check if there is a call in place? Is checking ofia_private->is_call there for another purpose? Thanks, Joe Barrero

2 1

Record Frequency Problems
by R G 12 Mar '21

12 Mar '21

Hi all, i've got a weird bug with the record_session function. I added he record_session function into my dialplan and set the record_sample_rate to 16000. When i use Google Chrome the record is fine and the wave file has a 16khz Sample Rate. When i use Mozilla Firefox the Sample Rate is at 32khz. my additions to the dialplan: <action application="set" data="RECORD_READ_ONLY=true"/> <action application="set" data="record_sample_rate=8000"/> <action application="record_session" data="/var/freeswitch/meetings/${strftime(%Y-%m-%d-%H-%M-%S)}_${call_uuid}.wav"/> Kind regards, Robert

1 0

No audio on bridged call
by Mircea Botoca-Huh 12 Mar '21

12 Mar '21

5 9

Re: [Freeswitch-users] using mod_unimrcp to transcribe voicemails
by David P 12 Mar '21

12 Mar '21

One of the ClueCon 2020 talks was by someone who uses mod_unimrcp and has a github project showing its FS integration, iirc.

3 3

Odd RTP skew behavior
by Marc Bernard 12 Mar '21

12 Mar '21

Hello Peeps, I often notice the following behavior on calls between two FreeSwitch servers: https://contattafiles.s3.us-west-1.amazonaws.com/getinfinity/4WUXPK9ibKr018c /Pasted%20Image%3A%20Jan%2029%2C%202021%20-%2010%3A05%3A07am Telco <==> FS Node 1 <== X ==> FS Node 2 <==> Callee UA Caller UA <==> FS Node 2 <== X ==> FS Node 1 <==> Telco We're not doing proxy media or bypass media. There are other calls at the same time that does not have this behavior, so I doubt it is caused by network issues. Does anyone have an idea of what could cause this ? Appreciate any help. Most Kindly, Marc

3 4