jitsi-meet: Too many open files, jvb becomes unavailable after some time

My problem is that I can enter a room, but when arrives the new participant, we get the timeout of ~24 seconds error.

So I first started thinking it was prosody #1386

that’s why I tried

service prosody restart
service jitsi-videobridge restart
service jicofo restart

after that, we could meet

next attempt, reduce one restart

service jitsi-videobridge restart
service jicofo restart

it works too after this operation

service jicofo restart

same…

conclusion: there is a problem with jicofo: after some time, it looses something

this is the final of a meeting according to /var/log/jitsi/jicofo.log, and after that you can see Health check failed

Note: I masked private information as:

  • user1: wherever appeared a nick (1 time)
  • meet.example.com: custom domain for the service (several times)
  • myroom: custom room inside jitsi (several times)
Jicofo 2017-03-09 17:42:00.567 INFO: [36] org.jitsi.jicofo.ChatRoomRoleAndPresence.log() Owner has left the room !
Jicofo 2017-03-09 17:42:00.567 INFO: [36] org.jitsi.jicofo.JitsiMeetConferenceImpl.log() Member myroom@conference.meet.example.com/64891030 is leaving
Jicofo 2017-03-09 17:42:00.568 WARNING: [36] org.jitsi.jicofo.JitsiMeetConferenceImpl.log() Participant not found for myroom@conference.meet.example.com/64891030 terminated already or never started ?
Jicofo 2017-03-09 17:42:00.568 INFO: [36] org.jitsi.jicofo.FocusManager.log() Disposed conference for room: myroom@conference.meet.example.com conference count: 0
Jicofo 2017-03-09 21:38:10.053 WARNING: [15] org.jitsi.jicofo.JvbDoctor.log() Health check failed on: jitsi-videobridge.meet.example.com error: <error code="500" type="WAIT"><internal-server-error xmlns="urn:ietf:params:xml:ns:xmpp-stanzas"/><text xml:lang="en" xmlns="urn:ietf:params:xml:ns:xmpp-stanzas">Failed to acquire DatagramSocket specific to STUN communication</text></error>
Jicofo 2017-03-09 21:38:10.053 INFO: [39] org.jitsi.jicofo.BridgeSelector.removeJvbAddress().192 Removing JVB: jitsi-videobridge.meet.example.com
Jicofo 2017-03-09 21:38:10.054 INFO: [39] org.jitsi.jicofo.JvbDoctor.log() Stopping health-check task for: jitsi-videobridge.meet.example.com

Here I put what happens next day when we try to join. It fails, until I restart jicofo

https://gist.github.com/pedro-nonfree/b4f46a1980e25ca45bc65599b32b001b

If you need other logs, tell me

Thanks

About this issue

  • Original URL
  • State: closed
  • Created 7 years ago
  • Comments: 50 (26 by maintainers)

Commits related to this issue

Most upvoted comments

Well first happy new year everyone! Since I manually added org.ice4j.ipv6.DISABLED=true to sip-communicator.properties the calls seem to be consistent after weeks of the last jicofo restart.

I’ll keep monitoring, but this seems to have solved my issue. Cheers!

This one solved my problem but if there is a common issue like this, why it’s not documented at somewhere?

Stable was updated a few hours ago 😃 Cheers 🍺