Users cannot enter meeting, 'You are the only one in the meeting'

Install & Config
#1

Hi. I installed Jitsi on a virtual Ubuntu machine hosted with an static IP adress in my provider’s data center. Installation worked fine, I have a working SSL certificate for the subdomain on which Jitsi should run.

When I try to use the link to the meeting I recently opened on my first client machine also on my second machine (as if I transmitted it with an eMail I type it into the URL line of the browser) the room opens immediately, camera and mic running. But on both machines I am told that I am alone inside the meeting. The second machine cannot enter the meeting.

I’m not sure where to start for debugging. Is this a Jicofo issue?

My configuration:
Server
Apache2 webserver with VirtualHost subdomain incl. working SSL certificate (letsencrypt certificate generated with CertBot) on an Ubuntu 18.04 LTS 64bit server with static IP.

Clients
2 Linux Mint Clients with Firefox and alternatively Chromium browsers.

Jicofo, Prosody and Videobridge2 running, all necessary ports allowed by firewall.

Is there any additional information you need for helping?

Marco

#2

Welcome to the forum.

Yes, this definitely sounds like a Jicofo issue but using anything but an auto-generated Letsencrypt certificate comes with its own caveats. You may in fact not be in actual meetings, but just viewing your own video feed in each client. Check out the Advanced configuration section of the Quick Install Guide; perhaps that might help to solve your problem.

#3

OK, thanks. I will check the Advanced configuration section again. Maybe I missed something in there.

Additionally here a part out of the jicofo.log. Could this be helpful?

java.lang.OutOfMemoryError: unable to create new native thread
at java.lang.Thread.start0(Native Method)
at java.lang.Thread.start(Thread.java:717)
at java.util.concurrent.ThreadPoolExecutor.addWorker(ThreadPoolExecutor.java:957)
at java.util.concurrent.ThreadPoolExecutor.processWorkerExit(ThreadPoolExecutor.java:1025)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1167)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
Jicofo 2020-11-26 15:57:47.741 FATAL: [43] org.jitsi.xmpp.component.ComponentBase.log() Ping timeout for ID: LuGo4-28579
Jicofo 2020-11-26 15:57:47.839 FATAL: [43] org.jitsi.xmpp.component.ComponentBase.log() Ping timeout for ID: LuGo4-28580
Jicofo 2020-11-26 15:57:57.742 FATAL: [43] org.jitsi.xmpp.component.ComponentBase.log() Ping timeout for ID: LuGo4-28583
Jicofo 2020-11-26 15:57:57.840 FATAL: [43] org.jitsi.xmpp.component.ComponentBase.log() Ping timeout for ID: LuGo4-28584
Jicofo 2020-11-26 15:58:06.708 FATAL: [10750] org.jitsi.jicofo.Main.log() An uncaught exception occurred in thread=Thread[Jicofo Scheduledpool-3-thread-10685,5,main]

I replaced the German word ‘SCHWERWIEGEND’ with ‘FATAL’. Maybe this word is usually not used in the English logfile.

#4

What are the resources? CPU, RAM…?

And the services status

systemctl status prosody
systemctl status jicofo
systemctl status jitsi-videobridge2
systemctl status jitsi-videobridge
#5

Virtual Processor vCores 4
RAM 8GB

systemctl status prosody
prosody.service - LSB: Prosody XMPP Server
Loaded: loaded (/etc/init.d/prosody; generated)
Active: active (running) since Thu 2020-11-26 18:19:05 CET; 10min ago
Docs: man:systemd-sysv-generator(8)
Process: 647 ExecStart=/etc/init.d/prosody start (code=exited, status=0/SUCCESS)
Tasks: 1 (limit: 60)
CGroup: /system.slice/prosody.service
└─658 lua5.1 /usr/bin/prosody
But with this error:
portmanager: Error binding encrypted port for https: No key present in SSL/TLS configuration for https

systemctl status jicofo
jicofo.service - LSB: Jitsi conference Focus
Loaded: loaded (/etc/init.d/jicofo; generated)
Active: active (running) since Thu 2020-11-26 18:18:54 CET; 12min ago
Docs: man:systemd-sysv-generator(8)
Process: 213 ExecStart=/etc/init.d/jicofo start (code=exited, status=0/SUCCESS)
Tasks: 59 (limit: 60)
CGroup: /system.slice/jicofo.service
└─235 java -Xmx3072m -XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=/tmp -Djdk.tls.ephemeralDHKeySize=2048 -Dconfig.file=/etc/jitsi/jicofo/jicofo

systemctl status jitsi-videobridge2
jitsi-videobridge2.service - Jitsi Videobridge
Loaded: loaded (/lib/systemd/system/jitsi-videobridge2.service; enabled; vendor preset: enabled)
Active: active (running) since Thu 2020-11-26 18:18:54 CET; 13min ago
Process: 396 ExecStartPost=/bin/bash -c echo $MAINPID > /var/run/jitsi-videobridge/jitsi-videobridge.pid (code=exited, status=0/SUCCESS)
Main PID: 395 (java)
Tasks: 46 (limit: 65000)
CGroup: /system.slice/jitsi-videobridge2.service
└─395 java -Xmx3072m -XX:+UseConcMarkSweepGC -XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=/tmp -Djdk.tls.ephemeralDHKeySize=2048 -Dconfig.file=

There is no service called videobridge. Only videobridge2 is installed.

#6

Did you restart the services after the error? Their status seem good but there is an error in the log file.

the output for

free -m

Do you connect to Jitsi using the IP address or the host address?
Is Jitsi configured using the IP address or the host address?

#7

Yes, I restarted all the services. Same output.
free -m:
total used free shared buffers cached
Mem: 8192 837 7256 2 98 7352
Swap : 0 0 0

I connect to Jitsi with the host adress subdomain.domain.de. I configured this hostname during install process of Jitsi.

The error shown from prosody:
portmanager: Error binding encrypted port for https: No key present in SSL/TLS configuration for https port 5281
There’s a key and a certificate configured inside prosody file ‘subdomain.domain.de.cfg.lua’ which aren’t the certificates I installed with letsencrypt. I think these files are for internal communication and don’t have to match with my own certificates, right? I did not find port 5281 in any documentation. The webserver should communicate with prosody over port 5280. Could this be the problem?

#8

You should check without restarting after the error to check if any service is crached or not…
The binding message is OK

#9

‘out of memory’ :slight_smile:
here is what I have on a LXD container:
Tasks: 293 (limit: 2329)
somewhat better, isn’t it ?
maybe check your /etc/systemd/system.conf ? I am not sure of the number of places this particular problem could come from.

#10

I updated the system.conf to the values mentioned in the Quick install guide:
Tasks: 47 (limit: 65000)
I already did the same a couple of days ago before restoring my virtual server to the recent snapshot. Back then this had no effect. But I will check it again when I’m back in my office.
The ‘out of memory’ error is still shown:
CGroup: /system.slice/jicofo.service
\u2514\u25005704 java -Xmx3072m -XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=/tmp -Djdk.tls.ephemeralDHKeySize=2048 -Dconfig.file=/etc/jitsi/jicofo/jicofo.conf

#11

This is not the error, an argument for the running process

#12

OK, I tried it again with increasing of the task limit. And now the connection (at least between two clients) works.

#13

I have a similar problem with a pop - up message about low memory. Updated system. the Conf didn’t fix the situation. I can’t figure out how to increase the task limit?

#14

You must modify the /etc/systemd/system.conf file:

DefaultLimitNOFILE=65000
DefaultLimitNPROC=65000
DefaultTasksMax=65000

Usually these lines are commented out. Remove the # and insert these values.