No jibris available issue which was fixed by meet instance reboot

Hi.
Had an issue that recording couldn’t start because of no free jibri’s.
Here is jicofo log:

Jicofo 2021-10-08 09:30:49.159 INFO: [7966] JibriSession.handleJibriStatusUpdate#650: Jibri failed, trying to fall back to another Jibri
Jicofo 2021-10-08 09:30:49.160 SEVERE: [7966] JibriSession.startInternal#305: Unable to find an available Jibri, can't start
Jicofo 2021-10-08 09:30:49.161 WARNING: [7966] JibriSession.handleJibriStatusUpdate#661: Failed to fall back to another Jibri, this session has now failed: org.jitsi.jicofo.jibri.JibriSession$StartExcepti
on$AllBusy: All jibri instances are busy
org.jitsi.jicofo.jibri.JibriSession$StartException$AllBusy: All jibri instances are busy
        at org.jitsi.jicofo.jibri.JibriSession.startInternal(JibriSession.java:308)
        at org.jitsi.jicofo.jibri.JibriSession.start(JibriSession.java:284)
        at org.jitsi.jicofo.jibri.JibriSession.retryRequestWithAnotherJibri(JibriSession.java:586)
        at org.jitsi.jicofo.jibri.JibriSession.handleJibriStatusUpdate(JibriSession.java:654)
        at org.jitsi.jicofo.jibri.JibriSession.processJibriIqFromJibri(JibriSession.java:441)
        at org.jitsi.jicofo.jibri.JibriSession.processJibriIqRequestFromJibri(JibriSession.java:425)
        at org.jitsi.jicofo.jibri.BaseJibri.doHandleIQRequest(BaseJibri.kt:154)
        at org.jitsi.jicofo.jibri.BaseJibri.access$doHandleIQRequest(BaseJibri.kt:42)
        at org.jitsi.jicofo.jibri.BaseJibri$incomingIqQueue$1.handlePacket(BaseJibri.kt:53)
        at org.jitsi.jicofo.jibri.BaseJibri$incomingIqQueue$1.handlePacket(BaseJibri.kt:42)
        at org.jitsi.utils.queue.PacketQueue$HandlerAdapter.handleItem(PacketQueue.java:381)
        at org.jitsi.utils.queue.AsyncQueueHandler$1.run(AsyncQueueHandler.java:133)
        at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
        at java.util.concurrent.FutureTask.run(FutureTask.java:266)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
        at java.lang.Thread.run(Thread.java:748)
Jicofo 2021-10-08 09:30:49.161 INFO: [7966] [room=test@conference.example.com] JibriRecorder.publishJibriRecordingStatus#161: Got jibri status off and failure error

However, we had about 10 free jibri’s available at that time. Restarting of jicofo/prosody on meet instance didn’t helped, same as jibri restart on jibri instance. Issue was solved by itself only with jitsi-meet instance reboot. Anyone have any ideas what could gone wrong? Maybe I can retrieve somehow the list of available jibri’s from meet instance?

Is nickname different for each jibri instance?

you mean this one, in /etc/jitsi/jibri/config.json?

      "control_muc": {
        "domain": "internal.auth.example.com",
        "room_name": "JibriBrewery",
        "nickname": "j2dfvfdvdfbDGBGERGvevevedfvdvR"
      },

Yea, seems config is right…

Here is a tip, when restarting Jibri service, you can see Jibri member joined in the room which is loged in the Jicofo (/var/log/jitsi/jicofo.log).
Check whether jibri succesfully joined room.

Yea, using custom module under host internal.auth.example.com i think so,

No, in fact, nicknames are identical in all my jibri’s. So seems that is the reason - Meet instance had only 1 instance all this time and when it failed - it couldn’t use any other jibri. Thank you for the help.

Nickname should different for each jibri instance, because all the Jibri are joined to a room named JibriBrewery ini Jitsi meet.