Late joiners to conference could not join initially but reconnected OK?


We had a Jitsi environment set up (nginx to Openfire to Jicofo/Videobridge) and had been running continuously for few weeks. Then we noticed lately some late joiners to a conference call were not able to join initially but after couple reconnection attempts they were connected successfully. We captured the web client side error and JS stack trace showed that
[JitsiMeetJS.js] <Object.getGlobalOnErrorHandler>: UnhandledError: null Script: Logger.js:125… Error: Focus error, retry after 4000
at i._allocateConferenceFocusError (moderator.js:447)
at moderator.js:375
at s.Handler.handler (strophe.js:3388)
at (strophe.js:2556)
at s.strophe.js:3822
at Object.forEachChild (strophe.js:1522)
at s.Connection._dataRecv (strophe.js:3810)
at e.Bosh._onRequestStateChange (strophe.js:5559)

[conference.js] <e.value>: CONFERENCE FAILED: conference.focusDisconnected Logger.js:125

Has anyone seen similar problems before? Can someone have a suggestion how to pinpoint the root issue? For example, any good debugging mechanism/log?


This mean that the focus user(jicofo) got disconnected from the call.


@damencho Typically when the “focus” user was really disconnected from the call, every participant in the call conference would be notified. However, in our scenario, only some users got this error when joining. All existing participants were still engaging in the call. Can it be some timing issue in the client side?


None, if there is no focus, there is no conference. Focus leaving is just a regular xmpp event and all will receive it and show the reload screen.