September 2021 - FreeSWITCH-users - FreeSWITCH Mailing Lists

Reload TLS certificate
by Andrew Cassidy 13 Sep '21

13 Sep '21

Hi guys, Is there a way to reload a TLS certificate without restarting the whole service? Kind regards, -- *Andrew Cassidy BSc (Hons) MBCS* Managing Director 0330 44 55 960 <+443304455960> andrew(a)cassidyweb.co.uk www.cassidyweb.co.uk

2 1

Re: [Freeswitch-users] How to detect close of verto tab?
by David P 11 Sep '21

11 Sep '21

> > Thanks, Ken. I'd like to know about any such timers to handle cases where > users don't try to re-enter the conference and only the virtual agent's > channel has media. > > > From: Ken Rice <krice(a)freeswitch.org> > To: FreeSWITCH Users Help <freeswitch-users(a)lists.freeswitch.org> > Cc: > Bcc: > Date: Thu, 9 Sep 2021 22:27:38 -0500 > Subject: Re: [Freeswitch-users] How to detect close of verto tab? > Thats actually a design feature, the whole intent os of someone > accidentally navigates away they can come right back and resume their call. > > there may be some timers you can adjust. > > Sent from my iPhone > > > On Sep 9, 2021, at 20:19, David P <davidswalkabout(a)gmail.com> wrote: > > > > > > Our config of FS 10.5 doesn't detect when a verto user closes the tab > (or directs the tab to a different url, etc) and the session runs on the > server for many minutes (while our virtual agent tries to engage the user, > thus defeating our media timeout). > > > > Is there a best practice for detecting this per-channel so it can be > used to end the conference? > > > > Note that browser event beforeunload cannot be relied upon, especially > in mobile browsers, as a trigger for hanging-up on the verto end. >

1 0

info not being executed?
by David Villasmil 11 Sep '21

11 Sep '21

Hello all, I'm facing a weird symptom... I have a dialplan something like so: <extension name="get_all_calls_with_vm"> <condition field="destination_number" expression="(.*)"> <action application="info"/> <action application="set" data="fromProvider=${sip_h_X-Provider}"/> ... Everything is getting executed BUT "info"... i've never seen this, anyone has any ideas why this might happen? Regards, David Villasmil email: david.villasmil.work(a)gmail.com phone: +34669448337

2 1

How to detect close of verto tab?
by David P 10 Sep '21

10 Sep '21

Our config of FS 10.5 doesn't detect when a verto user closes the tab (or directs the tab to a different url, etc) and the session runs on the server for many minutes (while our virtual agent tries to engage the user, thus defeating our media timeout). Is there a best practice for detecting this per-channel so it can be used to end the conference? Note that browser event beforeunload cannot be relied upon, especially in mobile browsers, as a trigger for hanging-up on the verto end.

2 1

Default configuration?
by Antony Stone 09 Sep '21

09 Sep '21

Hi. Doing a standard package-managed installation of Freeswitch under Debian using https://freeswitch.org/confluence/display/FREESWITCH/Debian+10+Buster I see that there are 20 internal extensions 1000 - 1019 already defined, and I've successfully registered a couple of phones to 1000 and 1001. Am I right in thinking that the default dialplan is supposed to allow calls between internal extensions, or do I have to configure something else for this to work? My setup is: - FreeSwitch installed under Debian 10 Buster on my home network 192.168.78.0/24 - two telephones configured manually to register to the FreeSwitch server, also on the same network 192.168.78.0/24 Result: - both telephones register to FreeSwitch successfully - dialling from one extension to the other causes the second extension to ring, and can be answered, but the SDP contents of the Invite requests from FreeSwitch to each telephone tell them both to send audio to the *external address* of my Internet router, not either the internal address of my FreeSwitch server, or the internal address of the other telephone. Obviously this doesn't work. This happens whether or not I start FreeSwitch with the "-nonat" option. So, I'm thinking there must be something simple that I'm failing to understand about the default configuration of FreeSwitch, because I thought that calling from one internal extension to the other was about the simplest test I could perform. Can anyone point me at "how to get started" for a simple scenario such as this? Thanks, Antony. -- Schrödinger's rule of data integrity: the condition of any backup is unknown until a restore is attempted. Please reply to the list; please *don't* CC me.

2 2

One way video for ~2 min, after that works fine for the rest of the call.
by Nathan Stratton 07 Sep '21

07 Sep '21

I hope someone is able to help me with this, I have been pounding my head for a few days. I am using mod callcenter with video. End user call's into a queue and sees waiting video Agents tryit.jssip.net client rings and agent picks up Both endpoints get two-way audio. End user gets video from agent after about 3 - 4 seconds Agent does not get video from until about ~2 minutes passes After that the call will say up forever, I have tried all the settings I know, but it always takes about ~2 min for the call center agent to see the end user. Running FreeSWITCH-mod_sofia/1.10.6-release-18-1ff9d0a60e~64bit ><> nathan stratton

1 0

mux conference video latency
by Daniel Ohaniuc 07 Sep '21

07 Sep '21

We notice high media latency for mux conferences with Verto clients, all muxed and displayed on the canvas. It works OK, but the latency makes is very difficult to have a real-time dialogue. This is not load related, even with only 1 caller watching himself on the canvas (the mux conference output), it takes almost 1 second to see a local move being reflected on the screen (e.g. if I raise my hand I can see it happening on the screen after almost 1 second ). This is obviously the roundtrip delay, but after discarding the intrinsic network latency (measured to be about 100 ms roundtrip) there seem to be around 800-900 ms added latency. It seems this is being introduced along the buffering/ transcoding/ muxing pipeline.Any suggestions please about what can be the cause and what we should try to reduce the latency? What sort of latency should we expect, what's your experience, has anyone deployed a video conferencing with acceptable latency for bidirectional, real time conversations? Thanks!

2 1

freeswitch_licence_server broken?
by Matthew Grooms 07 Sep '21

07 Sep '21

I'm trying to use a commercial AMD license I just acquired for a server and it doesn't work. Both freeswitch and the license server are running but freeswitch complains about not being able to contact the license server when I run amd_info ... root 25644 0.5 0.7 1256492 29580 ? S<l 11:38 0:01 /usr/local/freeswitch/bin/freeswitch -nc freeswi+ 25733 0.0 0.0 11056 1060 ? S<s 11:38 0:00 freeswitch_licence_server freeswitch@...> amd_info Can't contact licence server. If I kill the freeswitch_licence_server and launch it manually, I get this message instead ... freeswitch@...> amd_info Unrecognised resource AMD/0 Can't contact licence server. [root@... mgrooms]# uname -a Linux voip-dev.cn.bf 3.10.0-1160.36.2.el7.x86_64 #1 SMP Wed Jul 21 11:57:15 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux [root@... mgrooms]# cat /etc/redhat-release CentOS Linux release 7.9.2009 (Core) I'm running this host with both both selinux and firewall disabled. I also contacted supprt@ signalwire but no one is responding. Is the product I just purchased still supported? -Matthew

2 9

Any way to specify multiple audio files for hold_music?
by Nathan Neulinger 07 Sep '21

07 Sep '21

I am aware of using the local_stream option, but unfortunately with the content I'm being provided (voice messages, not just music), starting in the middle of a stream just sounds weird -- hence prefer to specify directly. It'd potentially be fine to start randomly at any wav file in a rotated (not shuffled) set, but starting in middle of single file doesn't work. Related question, if specifying hold_music directly to path to wav, is there any way to still retain the 'pick the best sample rate automatically' functionality? Worst case I just splice the wav files as part of the deployment process, but would be nice to allow for individual replacement. -- Nathan ------------------------------------------------------------ Nathan Neulinger nneul(a)mst.edu Missouri S&T Information Technology (573) 341-6679 System Administrator - Architect (573) 612-1412 System Infrastructure Team Manager

2 2

Identify and configure inbound equipments
by Adrien Martin 07 Sep '21

07 Sep '21

Hello, I currently have a sofia profile with some inbound acl, and try to differentiate inbound calls with more info than just acl. The goal would be to have different SIP/RTP settings and different dialplan processing for several inbound equipment while only having one IP/port for FreeSWITCH (one profile). The variables variable_sip_received_ip and variable_sip_received_port can be used to handle this with an outbound event_socket program, but copying gateways configuration into another program does not look like a good solution. https://freeswitch.org/confluence/display/FREESWITCH/Sofia+Configuration+Fi… says gateways can have inbound variables but i do not manage to make the profile use the gateways settings. When the INVITE is received, FreeSWITCH only logs about acl and not about gateways. How to use the inbound variables of the gateways/how to match inbound gateways? Or is there another way to achieve this?

1 0