-
Notifications
You must be signed in to change notification settings - Fork 351
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Room remains open after destroy #905
Comments
I'm seeing jicofo staying in the room. room_census filter jicofo. |
The destroy should be like this for the moment:
Till we make the reason optional for jicofo. But still there is a the problem you describe. |
I'm looking into this. |
Yep, I'm already using also the reason argument inside the prosody plugin |
Discovered it during investigation of jitsi/jicofo#905
Discovered it during investigation of jitsi/jicofo#905
* fix(SDP) Always modify the streamId part of msid when source-name signaling enabled. Do this even if the browser provides a non '-' string for streamId since we rely on this to generate the source name. * fix(multi-stream) Update the <ssrc, source-name> map for p2p as well. * fix(conference) Disable p2p between eps running in different SDP modes. Disable p2p between endpoints that are not running in the same SDP mode. The source-add/source-remove handlers assume the mids to follow the same pattern on both self and peer but the mids generated in plan-b mode are different from unified plan mode. Fixes jitsi/jitsi-meet#11100. * fix(SDP) fix unit test for LocalSdpMunger. * chore(deps): bump minimist from 1.2.5 to 1.2.6 Bumps [minimist](https://github.com/substack/minimist) from 1.2.5 to 1.2.6. - [Release notes](https://github.com/substack/minimist/releases) - [Commits](https://github.com/substack/minimist/compare/1.2.5...1.2.6) --- updated-dependencies: - dependency-name: minimist dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com> * fix(multi-stream) fix p2p issues with secondary video sources. * fix(multi-stream) add new m-line when remote adds a second video source in p2p. A new m-line needs to be added in the remote description when a new secondary video source is signaled by the peer. * fix(TPC) Suppress lower layers only for screenshare tracks. With multi-stream support enabled, an extra check for videoType for local track is needed to dtermine if the lower layers need to be suppressed. * fix(TPC) Mark the direction inactive on inactive media connection. If a new secondary video source is added on the jvb connection while p2p is active, make sure the direction is set to inactive on the jvb connection, so that no media is being sent out on the jvb connection. * squash: Address review comments. * fix(BreakoutRooms) fix checking for isBreakoutRoom early * Plumbing for e2e pings. (#1973) * Plumbing for e2e pings. * linter fixes * Re-use existing event instead of introducing a new one. * fix(presence) Parse peer presence in legacy format correctly. When source-name signaling is enabled, get correct peermedia info when the peer sends presence in the legacy format. This fixes an issue where Jigasi users cannot be heard by endpoints that have source-name signaling enabled. * ref(face-expressions) refactor face landmarks namings (#1977) * fix(presence): send presence update when a local track is removed. When a screenshare track is removed in the legacy mode, presence needs to be sent with the muted state and videoType so the remote gets updated. * fix(presence) Only apply default videoType on video tracks * fix: p2p reject reason (#1840) * fix(BridgeChannel): re-send constraints and videoType message after every re-connect. This fixes an issue where the receiver constraints and video type messages are sent on the bridge channel only the first time the ws conn gets re-established. * fix(video-quality) Update frame heights in content-modify for p2p. (#1983) * fix(video-quality) Update frame heights in content-modify for p2p. When the user changes the preferred video quality settings from the UI, update the frame height values in content-modify for p2p connection when source-name signaling is enabled. * fix(multi-stream) Send presence for desktop track before signaling when it is the first track to be added to the conference. Also do not suppress renegotiation for desktop track on p2p since the browser doesn't fire negotiation needed event. * squash: fix build * squash: fix linter issues * squash: add comment. * fix: Cleanups JitsiConference on reservation and max user failures. @damencho (#1986) * Revert "fix: Cleans up rooms after failure to join." This reverts commit f23372d. * fix: Cleanups JitsiConference on reservation and max user failures. * fix(multi-stream) Send the initial session-accept and then add the secondary video tracks when a session is re-created (#1987) * new changes from origin master * fix: Cleanups chat room on destroy. Discovered it during investigation of jitsi/jicofo#905 * chore(deps) bump async from 0.9.0 to 3.2.3 * fix(browser-support) Enable device selection in mobile Safari. With https://bugs.webkit.org/show_bug.cgi?id=179363 being fixed, we should now be able to switch between devices in call. * feat: Add an option to set bogus resolution constraints for getDisplayMedia. (#1991) * feat: Add an option to set bogus resolution constraints for getDisplayMedia. * fix(multi-stream) Inject multiple SIM groups on Firefox when multiple sources are added. * JitsiTrackEvents.TRACK_STREAMING_STATUS_CHANGED expose jitsiTrack as listener parameter (#1984) * provide jitsiTrack as parameter of the listeners on JitsiTrackEvents.TRACK_STREAMING_STATUS_CHANGED * reorder params * feat(ssrc-rewriting) Add a flag to signal support for ssrc re-writing to Jicofo. * feat: add AV1 mime type (#1990) With that, it's possible to set AV1 as preferred codec for p2p and try it on Chrome. * fix(BrowserCapabilities) Fallback to plan-b for Electron clients that have chromium ver < 96. * fix(RTC) Add AV1 to CodecMimeType unit test * fix(multi-stream) Set the source name of replaced track before configuring it. With just source-name enabled, set the source name of the replaced track before configuring the encodings, this fixes an issues where the sender constraints are not applied on the p2p connection because the source name is undefined. * fix(presence) Send a presence update on every track replace operation. Removing and adding a track back to the conference doesn't generate new source signaling all the time so presence needs to be updated every single time. Fixes https://community.jitsi.org/t/screen-share-doesnt-work-a-second-time-on-p2p/114024 and possibly jitsi/lib-jitsi-meet#1997. * fix(stats) Emit stats using source name as key (#1996) This allows us to display resolution and framerate stats on separate tiles when multi-stream mode is enabled. * feat: Adds a flag runInLiteMode useful for testing. Enables lite mode which disables any media decoding. * fix(rtc): fix local track removeTrack promise conference check (#1797) * fix(rtc): fix local track removeTrack promise conference check * ref: ignore if promise was successful or not * fix(device-selection) Support multiple audio inputs in Firefox 101 (#1988) * fix(multi-stream): Fix local SSRC cache to include multiple video streams. (#2006) * fix(multi-stream): Fix local SSRC cache to include multiple video streams. If multiple local video streams are found in the SDP, cache all of them instead of the first video SSRC. This fixes an issue where the resolution/fps stats for the screenshare track are not available. * squash: new track inherits the source name of old track if it exists. * fix(SDP): Reset sources to MSID map for plan-b clients. When source-name signaling is enabbled on plan-b clients, the sources to MSID map needs to be cleared after every transformation since browser in plan-b mode produces a new set of SSRCs and trackID when a track is removed and new track is added to the peerconnection. This fixes an issue where the track index in the source-name gets incremented after a track replace operation. * feat: Handles release number coming from backend. Adds it to deploymentInfo. * feat: Updates deployment info key for release and push to analytics. * fix(device-change): enable on iOS Safari < 15.4 * fix(RTC): Add function declaration to JitsiTrack. RN clients throw a function undefined error otherwise. * Revert "fix(RTC): Add function declaration to JitsiTrack." This reverts commit 7d3f13c. This change was not needed. * feat(RTC) drop no longer supported PC media constraints They have been deprecated and their valued can no longer be changed from the default after M103. Ref: https://groups.google.com/u/0/g/discuss-webrtc/c/2l7jWgreYpw?pli=1 * fix(deployment-info) Prevent error when setting shard and region (#2017) * fix(deployment-info) Prevent error when setting shard and region * fix * fix(docs) Added docs on how to define tokens (#2018) * added docs on how to define tokens * fixed typo Co-authored-by: James Green <james.green@affinity-digtal.com> * fix(types) remove setSuspendVideo declaration related to 5966576 * fix(JitsiConference) log a warning instead of an error is p2p fails In most cases it's not a real issue and it's still part of the logs, throwing an unhandled exception at the global handler just prints an ugly trackeback in the JS console which looks much worse than it really is. * fix(docs) improve documentation that jwt works on websocket see https://community.jitsi.org/t/lib-jitsi-meet-jwt-support-with-websockets/79806 * fix(presense) don't lose prev presenceChanged value * fix(TPC) Stop munging remote desc for imploding the SIM groups. Jicofo by default now strips the simulcast SSRCs and sends only the primary SSRC and RTX to all the receiving endpoints in the call. Therefore remote SDP munging is not required on the clients anymore. * fix(SignalingLayer): Skip checking for last presence if the client doesn't join muc. * fix(SignalingLayer): Skip checking for last presence if the client doesn't join muc. This fixes the broken wireless screensharing issue on spot clients. * squash: change log message to a warning. * fix(e2ee) keeps the initial key when decryption error * fix(e2ee) adds promise to map when sending key onParticipantPropertyChanged * ref(e2ee) change jsdoc return to correct promise the editor will no longer complain about superfluous await when calling these functions * ref(e2ee) make tests compatible with new karma/jasmine versions it removed support for functions being simultaneous async and have done parameter in callback * fix(ts) rebuild to fix types after jsdoc changes * ref(e2ee) mark _decryptFrame function with correct jsdoc the editor will no longer complain about superfluous await when calling * fix(breakout-rooms) Change callstats confID for breakout rooms * Add comment * fix(SS): missing remote screenshare Reusing(removing it and adding it again) an SSRC in the remote SDP is leading to the issue where remote video is received but not rendered. That's why we shouldn't remove the receive only SSRCs from the SDP. This way the sender won't send any source-remove and source-add for the same SSRCs. * fix(SS):Don't send source-remove/add for FF * sync with master * fix(SDP): don't separate fmtp parameter with spaces (#2034) * fix(SDP): don't separate fmtp parameter with spaces * adjusted tests accordingly to not expect spaces anymore * default to source name signaling * default to source name signaling Co-authored-by: Jaya Allamsetty <54324652+jallamsetty1@users.noreply.github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: Saúl Ibarra Corretgé <saghul@jitsi.org> Co-authored-by: George Politis <gp@jitsi.org> Co-authored-by: Avram Tudor <tudor.potecaru@8x8.com> Co-authored-by: Jorge Oliveira <neweracracker@gmail.com> Co-authored-by: Paul Tiedtke <PaulTiedtke@web.de> Co-authored-by: Дамян Минков <damencho@jitsi.org> Co-authored-by: brajendrak00068 <brajendra@Brajendras-MacBook-Air-M1.local> Co-authored-by: bgrozev <boris@jitsi.org> Co-authored-by: pangrr <rpang27@bloomberg.net> Co-authored-by: Paweł Domas <pawel.domas@jitsi.org> Co-authored-by: Daniel McAssey <DanielMcAssey@users.noreply.github.com> Co-authored-by: Nicolas Thumann <me@n-thumann.de> Co-authored-by: Hristo Terezov <hristo@jitsi.org> Co-authored-by: Horatiu Muresan <39557534+horymury@users.noreply.github.com> Co-authored-by: James Green <91469943+james-green-affinity@users.noreply.github.com> Co-authored-by: James Green <james.green@affinity-digtal.com> Co-authored-by: Tobias Schneider <Tobias.Schneider@Tobsch.org> Co-authored-by: cinques <cinquestyle@gmail.com> Co-authored-by: titus.moldovan <titus.moldovan@8x8.com> Co-authored-by: robertpin <robert.pin9@gmail.com> Co-authored-by: Nils Ohlmeier <nils-ohlmeier@users.noreply.github.com>
Description:
We have a custom prosody module which triggers a room destroy under certain conditions.
With latest Prosody versions, when the room is re-opened after a destroy, it remains open also if every participant leaves.
Steps to reproduce:
User A
opens a roomroom-name
User B
joins the roomroom-name
prosodyctl shell
) we trigger a room destroy using the commandUser A
andUser B
join again the same roomUser A
joins again the same roomExpected behavior:
Actual behavior:
The room timer starts from the last value. If we close again the room, the room timer does not reset.
Furthermore, if we enable the
muc_census
plugin on Prosody, we can see that when every participants quit, the room is still there!Server information:
Client information:
Additional information:
Same issue with stable/jitsi-meet_7001 and latest prosody 0.11
The text was updated successfully, but these errors were encountered: