Jitsi meet keeps disconnecting for 2 or more users in a room

Hi, I am fairly new to Jitsi meet and I have installed Jitsi meet in my server running on Ubuntu18 via quick installation process. When I try open jitsi room on two devices, it seems to work for about 10 seconds and then it keeps disconnecting even if there are only 2 participants.

This is the error message that I get on my browser console:
[modules/RTC/BridgeChannel.js] <l._send>: Bridge Channel send: no opened channel.

This is my config from /etc/jitsi/videobridge/sip-communicator.properties

And these are my ufw rules:

22/tcp ALLOW Anywhere
443/tcp ALLOW Anywhere
8443/tcp ALLOW Anywhere
10000/udp ALLOW Anywhere
80/tcp ALLOW Anywhere
4443/tcp ALLOW Anywhere
5347/udp ALLOW Anywhere
5347 ALLOW Anywhere

I’m guessing the server is running over NAT since I get a 10.xx.xx.xx ip when I do ifconfig.

you could check if the config is taken in account with
sudo grep SinglePortUdpHarvester /var/log/jitsi/jvb.log | tail -n 1
you should get back the address you get with ip a
if you don’t get a line having a date > last restart, there is a problem in your config file.
Check if your jvb is listening with
sudo ss -tapnu | grep java | grep 10000
you should get one (or possibly two line
if there is nothing check if jvb is running at all with ps fauxww | grep jvb

if everything looks good, check if port 10000 is really working.
Use nc.
On the server:

sudo systemctl stop jitsi-videobridge2
nc -l 10000 -u

On a workstation:

echo "123" | nc -u <your public address> 10000

you should get back ‘123’ on the server console.


I had this issues yesterday when I setup my Jitsi server.
Did you change your private and public IPs on the last two lines for this post, or have you not set them yet?
You should have your NAT_HARVESTER_LOCAL_ADDRESS=10.xx.xx.xx
and your NAT_HARVESTER_PUBLIC_ADDRESS set to your wan address.

1 Like

I think there is a problem with my 10000 UDP port.

echo "123" | nc -u <your public address> 10000

When I ran this, nothing happened.

Yes that I did. Did you fix it by any chance?

Then your network does not seem to be setup completely. Check the NAT & firewall config on all the steps in the link between workstation and server.

1 Like

Finally got it working. There was a problem with my 10000 udp port.
I had to open it through my aws account while creating the server. Previously I had only opened the ports after doing ssh through the server with command ufw allow.

Thank you very much!