I have still the same issue with the docker image. When I try to start it, it seems to just stops because of the STUN server.
Starting Meetecho Janus (WebRTC Gateway) v0.2.2
Checking command line arguments…
Debug/log level is 4
Debug/log timestamps are disabled
Debug/log colors are enabled
Adding ‘vmnet’ to the ICE ignore list…
Using 172.17.0.2 as local IP…
Token based authentication enabled
Initializing ICE stuff (Full mode, ICE-TCP candidates disabled, IPv6 support disabled)
STUN server to use: stun.l.google.com:19302
ICE handles watchdog started
Testing STUN server: message is of 20 bytes
[FATAL] [ice.c:janus_ice_set_stun_server:770] No response to our STUN BINDING test
[FATAL] [janus.c:main:3554] Invalid STUN address stun.l.google.com:19302
So in order to make it work you definitely need to set a STUN server in the janus conf, it does not have to be the google one it can be any STUN server as long as you can have access to, please ensure that you can actually reach the STUN server from the Janus box (webrtc docker image or manual installation of janus)
Tried Firefox (54), Chrome (57) and Mattermost desktop for linux (3.7.0).
Yes.
Each time.
All we get is the “There was a problem connecting the video call.” error. We can see our selves in the window but there is no video or audio with the other side.
I am also facing the same problem. I was able to install the docker succesfully and Webrtc is configured on the server as well as the End user profile. But i am getting this error There was a problem connecting the video call.
Used both docker preview for WebRTC and my own janus setup. I didn’t setup STUN/TURN
I have mattermost 4.0.2 on Debian 8 usng Postgresql running via systemd service. Mattermost is behind a proxy which do SSL termination
I tried both Chorme and Firefox on Ubuntu 16.04 and Windows10 (latest available version)
THe issue reproduce for every user each time a call is started
I also checked logs for all entities
In browser console nothing appears
In mattermost logs (debug) I see many call but not error. I see also some /api/v3/users/websocket:webrtc
request
I janus logs I see the token request successfull
Got an admin/monitor HTTP POST request on /admin…
[transports/janus_http.c:janus_http_admin_handler:1483] … Just parsing headers for now…
[transports/janus_http.c:janus_http_headers:1734] Host: PUBLIC_HOSTNAME:7889
[transports/janus_http.c:janus_http_headers:1734] User-Agent: Go-http-client/1.1
[transports/janus_http.c:janus_http_headers:1734] Content-Length: 179
[transports/janus_http.c:janus_http_headers:1734] Content-Type: application/json
[transports/janus_http.c:janus_http_headers:1734] Accept-Encoding: gzip
[transports/janus_http.c:janus_http_admin_handler:1590] … parsing request…
Processing POST data (application/json) (179 bytes)…
[transports/janus_http.c:janus_http_admin_handler:1640] – Data we have now (179 bytes)
[transports/janus_http.c:janus_http_admin_handler:1590] … parsing request…
Processing POST data (application/json) (0 bytes)…
[transports/janus_http.c:janus_http_admin_handler:1645] Done getting payload, we can answer
{“admin_secret”:“XXXXXXXXXXXXXXXXXXXXXXXXXXX”,“janus”:“add_token”,“token”:“eWRjNWdnc3JwN2Z0M2JlZzE0Y3FpaThvOG8=”,“transaction”:“9tyi93jkmjdjzge9gtbngg61ae”}
Forwarding admin request to the core (0x7fd220001e60)
Got an admin API request from janus.transport.http (0x7fd220001e60)
Transport task pool, serving request
My setup is as follows
janus installed on a public host, no proxy, no nat
mattermost running on a public host behind reverse proxy (ssl termination)