Instead of Jicofo signaling all the remote sources available in the call, the bridge now signals only a limited set of SSRCs and then rewrites the SSRC on the outgoing media streams. The SSRC mapping is done based on the sources requested by the clients through the receiver constraints. This limits the number of m-lines in the remote/local SDPs on the client and therefore results in better performance in large calls.
* Handle source remapping messages from bridge
* Added track_owner_changed events
* don't process an invalid rtx ssrc.
* keep track of remote ssrcs, only renegotiate on new ones.
* Change source name on remote track on ssrc remapping.
* Don't remove tracks on member leave.
* Remove (orphaned) tracks on session terminated.
* Use serial number (per media type) to create msid attribute.
* Update videoType on remapping.
Co-authored-by: James A <jqdrqgnq@users.noreply.github.com>
When the participant has joined after A/V moderation has been enabled
they are auto muted with the startmuted node in the session-initiate
packet. We were not marking those as muted by the focus and therefore on
unmute we were not sending the <muted>false</muted> IQ to jicofo.
The result was that the bridge wasn't forwarding the media to the
remote participants.
fix: Cleanups JitsiConference on reservation and max user failures. @damencho (#1986)
* Revert "fix: Cleans up rooms after failure to join."
This reverts commit f23372dd88.
* fix: Cleanups JitsiConference on reservation and max user failures.
They are companion rooms created in a separate MUC. The room relationship is
maintained by a Prosody plugin.
All signalling happens through the breakout rooms MUC component.
Co-authored-by: Saúl Ibarra Corretgé <saghul@jitsi.org>
* feat: Audio/Video moderation.
* squash: Fix docs.
* squash: Adds some warning logs when execution is rejected.
* squash: Changes a field name in the message for adding jid to whitelist.
* squash: Send to participants only message about approval.
Skips sending the whole list.
* squash: Fixes tests.
* squash: Adds more logs.
* feat: Separates enable/disable by media type.
Adds actor to the messages to inform who enabled it.
* squash: Fixes log line.
* squash: Fixes comments.
* squash: Fixes log messages.
* squash: Fixes comments.
feat(stats): Add a new bridge message "EndpointStats" for stats.
Use the new Colibri message "EndpointStats" for broadcasting the local stats. The bridge then will be able to filter the endpoint stats and send them only to the interested parties instead of broadcasting it to all the endpoints in the call.
feat(stats): Get audio levels for the top 5 speakers only.
Capture the audio levels only for the top 5 speakers as RTCRtpReceiver#getSynchronizationSources can be expensive when we have too many audio receivers in the call.
Also, capture the audio levels for track that are unmuted if RTCRtpReceiver#getSynchronizationSources is not supported.
Switch Safari to using getStats since its reporting errorneous values, i.e., 0.000001 as audio level for all remote audio tracks.
* Added video mute participant
* Trigger mute event
* Optimized mute type checks
* Fixed event name
* Fixed some linter issues
* Fixed more linter issues
* And even more linter issues fixed
* And more linter fixes
* Added media type to analytics event
fix(ice-restart): Force client reloads when call is migrated.
Force the client to reload when the bridge that is handling the media goes down.
This mitigates issues seen on the bridge because of a client re-joining the call with the same endpointId, BWE issues, etc.
This behavior is configurable through 'enableForcedReload' setting in config.js.
feat: Skips using disco-info for features. (#1450)
* fix: Drops unused parameter of join and sendPresence.
* feat: Skips using disco-info for features.
Uses presence to publish features added externally.
Recognizes jigasi participants using a specific presence extension used by jigasi.
* squash: Fixes tests.
* squash: Adds e2ee to the features in presence.
* squash: Fix function name and docs.
* squash: Drops detecting jigasi from initiator.
Using the newly added presence features.
* squash: Drops unused var.
* squash: Fix comments.
* squash: Adds a constant and for E2EEE feature.
fix: Maybe start p2p only when we have received all presences.
There are cases where we start p2p on the first participant received presence, where there are more presences coming. This starts the process of p2p resulting and few iq errors while terminating it quickly after starting it.
On a normal joining as third participant we see around 38 send or received iq messages, in the problem scenario we see 60.
* Initial impl of lobby rooms.
* Fixes tests to check the new fulljid added to MUC_MEMBER_JOINED.
* Updates few of the comments, renaming some functions.
* Renames disableLobby ChatRoom option to enableLobby.
* Fixes a comment.
* Moves setMembersOnly method to ChatRoom.
* Fixes counting members, to exclude jicofo.
* Moves setLobbyRoomJid earlier and renames a method.
Rename _maybeEnableDisable to maybeJoinLeaveLobbyRoom.
* Drops using custom roomconfig lobbypassword field and reuse room lock.
* Handles destroying the lobby room.
* Handles clear lobby room on destroy for moderators.
We do not try to leave the lobby room as it is server-side destroyed and we handle that. The only case of leaving a lobby room is when request to join room is being approved.
* Join main room if lobby is disabled while waiting.
* Adds MEMBERS_ONLY_CHANGED conference event.
* fix: Make sure we leave lobby if main room is joined.
* fix: Setting password when joining locked room.
* fix: Fixes case where we enable lobby for already locked room.
* fix: Fixes case where we enable lobby and then lock room.
* fix: Fixes lint.
* ref: Removes shared password for lobby.
This functionality is handled by the lock room password and handled there.
Removes duplication and unnecessary complicated API for lobby room.
* fix: Fixes comments.
* Updates kicked event to inform local and remote kicks and who it affects
* Parses actor from MuteIQ and adds participant info to mute track event.
* Fixes emitting both events.
* Fixes comments.
Fix: Old JitsiConference instance remains listening on incoming calls… (#858)
* Fix: Old JitsiConference instance remains listening on incoming calls after leave - rejoin to same room
* Remove all listeners added by setupXMPPListeners
Changes the behavior to actively open new WebRTC Data Channel instead
of waiting for the JVB to do it.
Adds ICE_RESTART_SUCCESS event used to re-initialize the data channel in
case of ICE restart where the conference could have been moved to
another bridge.
When the main domain advertises component with type 'speakerstats' every participant will send message to the component with the name of the room where this happen.
The code for handling device availability has been disabled for a long time,
plus it's ill named since it represents 2 abstractions: lack of permissions and
lack of devices.
Time for it to rest in the git graveyard.
This simple change helps to prevent `'The conference is has been already left'` error after kick.
`Conference.leave` will detach chatRoom instance itself.
damencho
Saúl Ibarra Corretgé
Jaya Allamsetty
Hristo Terezov
Gary Hunt
Gary Hunt
Andrei Bora
Werner Fleischer
robertpin
Jade Guiton
vp8x8
Jaya Allamsetty
Steffen Kolmer
Andrei Gavrilescu
paweldomas
theunafraid
Bettenbuk Zoltan
Gábor Takaró
Andrey Davydov