[SOLVED!] JVB Error on boot, Won't start

Hi, just booted my server and tried to access my jitsi installation. Couldn’t get into a meeting. I tried restarting services and I got the error:

" Job for jitsi-videobridge2.service failed because the control process exited with error code.
See “systemctl status jitsi-videobridge2.service” and “journalctl -xe” for details"

I checked JVB log and this is what I have:

2020-09-04 00:01:00.205 INFO: [97] [confId=1fab6de88376fe8b gid=87813 stats_id=Lila-mqI componentId=1 conf_name=a4663ab3c7c64ef2 ufrag=qi061ehb1o0u0 name=stream-2297088a epId=2297088a local_ufrag=qi061ehb1o0u0] MergingDatagramSocket$SocketContainer.runInReaderThread#770: Failed to receive: java.net.SocketException: Socket closed

2020-09-04 00:01:00.662 INFO: [177] [confId=1fab6de88376fe8b gid=87813 stats_id=Darius-g0Y componentId=1 conf_name=a4663ab3c7c64ef2 ufrag=babsd1ehb3o7sv name=stream-239d1e7a epId=239d1e7a local_ufrag=babsd1ehb3o7sv] MergingDatagramSocket$SocketContainer.runInReaderThread#770: Failed to receive: java.net.SocketException: Socket closed

2020-09-04 00:01:02.150 INFO: [84] [confId=1fab6de88376fe8b gid=87813 stats_id=Antonia-sg4 componentId=1 conf_name=a4663ab3c7c64ef2 ufrag=8m1s31ehb1nq60 name=stream-186a6539 epId=186a6539 local_ufrag=8m1s31ehb1nq60] MergingDatagramSocket$SocketContainer.runInReaderThread#770: Failed to receive: java.net.SocketException: Socket closed

2020-09-04 00:01:03.404 INFO: [102] [confId=1fab6de88376fe8b gid=87813 stats_id=Colby-5ut componentId=1 conf_name=a4663ab3c7c64ef2 ufrag=99kgv1ehb1oi9i name=stream-c9171593 epId=c9171593 local_ufrag=99kgv1ehb1oi9i] MergingDatagramSocket$SocketContainer.runInReaderThread#770: Failed to receive: java.net.SocketException: Socket closed

2020-09-04 00:01:05.834 INFO: [72] [confId=1fab6de88376fe8b gid=87813 stats_id=Barton-25z componentId=1 conf_name=a4663ab3c7c64ef2 ufrag=9prtp1ehb1npm3 name=stream-7cadbccc epId=7cadbccc local_ufrag=9prtp1ehb1npm3] MergingDatagramSocket$SocketContainer.runInReaderThread#770: Failed to receive: java.net.SocketException: Socket closed

2020-09-04 00:01:16.945 INFO: [109] [confId=1fab6de88376fe8b gid=87813 stats_id=jibri componentId=1 conf_name=a4663ab3c7c64ef2 ufrag=fekr61ehb1oud7 name=stream-c33ed35b epId=c33ed35b local_ufrag=fekr61ehb1oud7] MergingDatagramSocket$SocketContainer.runInReaderThread#770: Failed to receive: java.net.SocketException: Socket closed

2020-09-04 00:01:16.946 INFO: [107] [confId=1fab6de88376fe8b gid=87813 stats_id=jibri componentId=1 conf_name=a4663ab3c7c64ef2 ufrag=fekr61ehb1oud7 name=stream-c33ed35b epId=c33ed35b local_ufrag=fekr61ehb1oud7] MergingDatagramSocket$SocketContainer.runInReaderThread#770: Failed to receive: java.net.SocketException: Socket closed

2020-09-04 00:01:36.995 INFO: [61] [confId=1fab6de88376fe8b gid=87813 stats_id=Selena-nJG componentId=1 conf_name=a4663ab3c7c64ef2 ufrag=v6691ehb1npet name=stream-db8bc467 epId=db8bc467 local_ufrag=v6691ehb1npet] MergingDatagramSocket$SocketContainer.runInReaderThread#770: Failed to receive: java.net.SocketException: Socket closed

2020-09-04 00:01:36.996 INFO: [58] [confId=1fab6de88376fe8b gid=87813 stats_id=Selena-nJG componentId=1 conf_name=a4663ab3c7c64ef2 ufrag=v6691ehb1npet name=stream-db8bc467 epId=db8bc467 local_ufrag=v6691ehb1npet] MergingDatagramSocket$SocketContainer.runInReaderThread#770: Failed to receive: java.net.SocketException: Socket closed

2020-09-04 00:02:06.822 INFO: [296] [confId=1fab6de88376fe8b gid=87813 conf_name=a4663ab3c7c64ef2] Conference.updateStatisticsOnExpire#626: expire_conf,duration=5224,has_failed=false,has_partially_failed=false

2020-09-04 16:50:12.077 INFO: [37] org.ice4j.ice.harvest.MappingCandidateHarvesters.initialize: Initialized mapping harvesters (delay=79ms). stunDiscoveryFailed=false

The only thing I’ve changed in the system from the last successful run was to grant my username ownership of the /var/log/jitsi/ directory so I could access the logs therein. I’m wondering if this issue arose from that. Any help would be appreciated. Thanks.

Its possible, that needs to be group jitsi. And you can add your user to that group.

Thanks. So the directory ownership should be “Jitsi”, if I understand you correctly?

Geoup jitsi and user jvb or jicofo

Thank you! Will do now and check.

Thanks so much @damencho! That fixed the JVB error. But now, Jitsi seems to be crashing into that unending “unfortunately something went wrong…” loop

I’m searching the forum now for a possible fix.

@damencho, unfortunately, I’ve not been able to work this out. I’ve gone through all the checks - made sure 10000/udp is actually open, restarted all services e.t.c… Even tried editing sudo nano /sip-communicator.properties as suggested by some, no luck.

There’s nothing in jicofo log (even though status is active and running); here’s my JVB log:

jvb.log (10.0 KB)

What about this?

If you restart jvb does it run? What about js logs in browser?

Yes, jvb runs. Let me get the output for journalctl -xe again. One sec…

Check js console logs to understand why it reloads

That doesn’t sound right. Are your sure it’s running?

Output for journalct -xe

– Support: http://www.ubuntu.com/support

– A start job for unit UNIT has finished successfully.

– The job identifier is 7.

Sep 05 02:13:01 jitsisvr systemd[1251]: Reached target Basic System.

– Subject: A start job for unit UNIT has finished successfully

– Defined-By: systemd

– Support: http://www.ubuntu.com/support

– A start job for unit UNIT has finished successfully.

– The job identifier is 4.

Sep 05 02:13:01 jitsisvr systemd[1251]: Reached target Main User Target.

– Subject: A start job for unit UNIT has finished successfully

– Defined-By: systemd

– Support: http://www.ubuntu.com/support

– A start job for unit UNIT has finished successfully.

– The job identifier is 3.

Sep 05 02:13:01 jitsisvr systemd[1251]: Startup finished in 298ms.

– Subject: User manager start-up is now complete

– Defined-By: systemd

– Support: http://www.ubuntu.com/support

– The user manager instance for user 1000 has been started. All services queued

– for starting have been started. Note that other services might still be starting

– up or be started at any later time.

– Startup of the manager took 298726 microseconds.

Sep 05 02:13:01 jitsisvr systemd[1]: Started User Manager for UID 1000.

– Subject: A start job for unit user@1000.service has finished successfully

– Defined-By: systemd

– Support: http://www.ubuntu.com/support

– A start job for unit user@1000.service has finished successfully.

– The job identifier is 351.

Sep 05 02:13:01 jitsisvr systemd[1]: Started Session 1 of user freddie.

– Subject: A start job for unit session-1.scope has finished successfully

– Defined-By: systemd

– Support: http://www.ubuntu.com/support

– A start job for unit session-1.scope has finished successfully.

– The job identifier is 430.

Sep 05 02:13:03 jitsisvr kernel: [UFW BLOCK] IN=eno2 OUT= MAC=01:00:5e:00:00:01:48:5d:36:71:cc:94:08:00 SRC=192.168.1.1 DST=224.0.0.1 LEN=36

js console log seems to be pointing to a focus problem;

CONFERENCE FAILED: conference.focusDisconnected

Yeah, checked it and jicofo was indeed running. Likely something wasn’t being passed to it so it wasn’t being engaged.

Screen Shot 2020-09-04 at 10.28.13 PM

Check jicofo logs why it disconnects

There’s no fresh log generated for jicofo. I cleared the log earlier, nothing’s been generated since.

So maybe jicofo journal you want to check and probably jicofo cannot access that folder

1 Like

You mean the jicofo.log file or the entire jitsi directory? Isn’t jicofo a member of the jitsi group? I’m thinking jitsi should be the primary group for jicofo - no?

Just tried it: manually added jicofo to the jitsi group (which owns the jitsi folder). Didn’t fix it.

@damencho okay, I’m making progress!!! :smiley:

I made jicofo the specific owner of the file jicofo.log and I was able to get in! I saw a few errors on the js console though, so still monitoring that. But half-yaaaaaaaaaaaay so far!!! :smiley: Hoping for a full-yaaaaaaaaay soon.

Okay @damencho, I think I can confirm it works now. The ‘error’ I saw in the console was just the application looking for authentication. Once I authenticated as a moderator, it went away.

To summarize, in case someone else runs into this issue:

Cause:
In trying to access logs through sftp, I changed the ownership of “/var/log/jitsi/ *” I couldn’t access the directory otherwise. However, changing the ownership caused permission problems. JVB would not start and I was getting the following error:

Job for jitsi-videobridge2.service failed because the control process exited with error code.
See “systemctl status jitsi-videobridge2.service” and “journalctl -xe” for details

Fix
The fix was to re-assign ownership to the Group “jitsi” -

chgrp jitsi /var/log/jitsi/

I also explicitly added jvb and jicofo to the group (not sure this is necessary because it seems to be their primary group):

sudo chown -R jvb /var/log/jitsi/
sudo chown -R jicofo /var/log/jitsi/

However, jicofo.log still was not populating at all…

…even though I confirmed the service was running:

Screen Shot 2020-09-04 at 10.28.13 PM

The final bit for me was to grant jicofo ownership of the jicofo.log file (I’d assumed since it was a member of the ‘jitsi’ group, it automatically had ownership):

sudo chown -R jicofo /var/log/jitsi/jicofo.log

And bingo! That fixed the issue for me! :smiley:

Deepest thanks to @damencho for his patience and forbearance. Apparently I do a really good job of screwing things up with a simple move. :joy: Thanks also to @corby for chiming in. :+1:t5: