When 2 or more participants, all disconnect after ~30 seconds

Hello i have a fresh installed jitsi.

This ist what i did:
apt install apache2 gnupg2 certbot apt-transport-https python-certbot-apache curl npm nodejs git

touch /etc/apt/sources .list.d /jitsi-stable .list

wget -qO - https: //download .jitsi.org /jitsi-key .gpg.key | apt-key add -

sh -c "echo 'deb https://download.jitsi.org stable/' > /etc/apt/sources.list.d/jitsi-stable.list"

apt update

certbot certonly -d DOMAIN --apache

apt install jitsi-meet

In front of it, there is a ReverseProxy(opnsense).
I did there this:
Port 4443 TCP
Port 10.000 - 20.000 UDP
and ofcourse 443 for the site is open.

Now when we want to join. All is ok.
After ~30seconds with min. 2 people, we disconnect both at the same time.
With 3 it is the same.

In the js console on client:
The last 2 errors come when we disconnect.
The errors before comes every several seconds.
2020-08-06_13-09

In jicofo-log this is the only warning:
Jicofo 2020-08-06 13:11:32.992 WARNUNG: [30] org.jitsi.jicofo.JitsiMeetConferenceImpl.log() No jingle session yet for visiblecorruptionssoundoverall@DOMAIN

in jvb.log there is no warning or error.

@demancho
I searched days in this forum and on google. But there is no solution for my problem. Please help :slight_smile:

@tplugge,

Have you tried the tip by @emrah, found here: Tip: how to check UDP/10000 connectivity?

If so, what were the results?

Check out this reply also:

Hey,
thanks for the reply.

I tested it now:
Serversite:
interface: enp0s3 (192.168.1.0/255.255.255.0)
filter: ( udp port 10000 ) and ((ip || ip6) || (vlan && (ip || ip6)))
match: is accessable

Clientsite:
Here there is no output

So now the test works.

But the problem is the same.
When we are 2 we see us and can talk but dc after ~30 seconds.
When we are 3 we dont see us and we cant talk and dc after ~30 seconds.

Still seems as a videobridge problem. What is the output?

systemctl status jitsi-videobridge2.service

Is there something wrong on jvb and jicofo logs?

Maybe you need to customize NAT_HARVESTER in the sip config according to your environment.

I have 3 Jitsi Server on 2 it works on the third not.

Every instance is on an other network.
The Service ist on.

I have configured this:
org.ice4j.ice.harvest.NAT_HARVESTER_LOCAL_ADDRESS=HERE_IS_MY_LOCAL_IP
org.ice4j.ice.harvest.NAT_HARVESTER_PUBLIC_ADDRESS=HERE_IS_MY_PUBLIC_IP

Like is said in the first Post there is nothing in the logs with warning or error.

Sounds like have a working config to research and compare against. I would start by comparing the configuration files of the working server with non-working. You’ll find something that’s different.

This also concerns me. There should be something useful. Nothing may mean your not actually hitting the server you think you are?

Hey i tested it alot.

The configurations of all 3 Jitsi’s is the same.

Only the Firewall is not the same.
Which Ports do we need ?
TCP 4443
10000-20000 UDP
are open

Hi tplugge

lets say the network kind of like below, you may try the follow check, and hope can fix the problem.


Network


(1) check file /etc/hosts ,make sure the FQDN point to 127.0.0.1

127.0.0.1 localhost
127.0.0.1 meet meet.a.com

(2) check file /etc/jitsi/videobridge/sip-communicator.properties , make sure NAT_HARVESTER configured.

org.ice4j.ice.harvest.NAT_HARVESTER_LOCAL_ADDRESS=192.168.1.250
org.ice4j.ice.harvest.NAT_HARVESTER_PUBLIC_ADDRESS=10.1.1.100

(3) check file /etc/turnserver.conf , make sure external-ip and allow-peer-ip configed.

external-ip=10.1.1.100/192.168.1.250
allowed-peer-ip=192.168.1.250

(4) reboot jitsi-meet server and testing

good luck :grinning:

James Chen