Flussonic Media Server 21.03
March 4, 2021
Flussonic Media Server 21.03 has been released.
Low-Latency HLS, streaming sessions API, configuration template for static streams, automatic allocation of transcoded streams among GPUs, SSM ingest, passing MPEG-TS teletext to SDI, and more in Flussonic 21.03
- #11436 (new) (Flussonic, Catena) With automatic allocation of transcoded streams among GPU cards you no longer have to check whether a GPU is overloaded and move streams to another card manually. Flussonic does it all based on the GPU load and memory consumption. Learn more
- #12604 (new) (Flussonic) Source-specific multicast (SSM) ingest was added to Flussonic. Learn more
- #14711 (new) (Flussonic) The schema ‘push udp://’ now has all the features of ‘push udp2://’, among them automatic constant bitrate on output and setting a custom MULTICAST_TTL. The syntax ‘push udp2://’ is no longer supported.
- #15086 (new) (Flussonic) Flussonic Media Server can pass teletext from MPEG-TS to analog streams. Learn more
- #15313 (new) (Flussonic) The support for Low-Latency HLS was added to Flussonic. Learn more
- #15386 (new) (Flussonic) A configuration template for static streams is now supported. Learn more
- #15517 (new) (Flussonic) In Flussonic 21.03, global configuration options will be moved to a template that will be applied to all streams by default. Learn more
- #15606 (new) (Сoder) Enhanced recovery of hanging Coder module.
- #9553 (new) (Сoder) Coder sends all data within a TCP session from those IP address and port that a client device has connected to initiate the session.
- #15309 (new) (Player) The DVR player can play two streams as a mosaic - use the URL like embed.html?dvr=true&streams=fake1,fake2
- #15772 (new) (Player) WebRTC voice chat demo is available, it is used for creating a chat for two people, one publishing a stream, the other playing it.
- #13612 (new) (Catena) DVR playback in Kodi is possible now.
- #14191 (new) (Catena) The UI allows you to edit the configuration settings, you can change all the settings that are contained in /etc/catena/config.json.
- #14795 (improved) (Catena) Catena installation improvements: example is available of how to configure Catena to run in Docker.
- #15492 (new) (Catena) A JSON API for managing service subscribers is available.
- #15539 (new) (Catena) Published UGC streams coming from ingesting servers can be redirected via code 302 to different transcoding servers in order to balance the server load. This redirection works only via m4s. Learn more
- #13010 (fixed) (Flussonic) Excessive logging in mpegts_encoder3 about bitrate was removed.
- #13648 (fixed) (Flussonic) Stream sources now properly issue the events source_opened and source_started.
- #13905 (fixed) (Player) If a camera had an audio track, then the video from this camera did not play on the mini-player on the map and in the camera settings in Watcher.
- #13944 (fixed) (Player) The DVR player no longer stops responding after receiving a chunk with incorrect media info.
- #14295 (fixed) (Flussonic) Filtering by the module that generated an event (module=) was fixed.
- #14349 (fixed) (Flussonic) Export to MP4 was fixed.
- #14504 (fixed) (Flussonic) Memory leaks during SRT publishing was fixed.
- #14524 (fixed) (Flussonic) The bitrate is no longer increased when transcoding on NVENC without FPS specified.
- #14806 (fixed) (Flussonic) The event audio_silence_end is generated correctly now.
- #14811 (fixed) (Flussonic) Changed: stream_started event to stream_opened and stats fields started_at to opened_at.
- #14884 (fixed) (Flussonic) Transcoding of RTMP published streams was fixed.
- #15164 (fixed) (Flussonic) Now the user cannot save an invalid source URL for a stream.
- #15264 (fixed) (Сoder) Encoding audio in mp2a was fixed.
- #15286 (fixed) (Flussonic) Subtitles burning together with MBR transcoding now works correctly.
- #15305 (fixed) (Flussonic) The option ‘aac=true’ for PCMU audio in RTSP streams now works.
- #15328 (fixed) (Flussonic) Playback of DVR video with subtitles via DASH was fixed.
- #15343 (fixed) (Flussonic) The Transcoder page no longer shows the default value for FPS selected.
- #15366 (fixed) (Flussonic UI) Protocol and duration of client sessions are shown in Clients again.
- #15382 (fixed) (Flussonic) If transcoding of audio track to mp2a or ac3 was chosen, than subtitles and logo burning didn’t work.
- #15396 (fixed) (Flussonic) The ‘reserved_future_use’ and ‘reserved’ fields are set to 1 in MPEG-TS EIT.
- #15409 (fixed) (Flussonic) Increasing number of requests to an authorization backend was fixed.
- #15463 (fixed) (Flussonic) The bitrate in MSS manifest for live streaming no longer varies and causes switching to another quality in the player.
- #15464 (fixed) (Flussonic) Publishing via WebRTC was fixed.
- #15509 (fixed) (Flussonic) Ingest from cards AJA 44 Crovid 12G was fixed for all channels.
- #15516 (fixed) (Player) Playback of published WebRTC streams in our MSE-LD player was fixed.
- #15431 (doc) Session-related events were added to Flussonic API for you to control streaming sessions.
- #13955 (doc) You can find how to get help in the beginning of our documentation.
- #15187 (doc) A number of screenshots were added to Quick Start guide and the screenshot of the start page was updated.
- #15524 (doc) The value ‘fast’ for the option ‘preset’ in the transcoder settings was removed from the documentation because it is no longer supported. The list of supported presets is described in Transcoder.
- #15605 (doc) The list that shows relations between the Deinterlace box in transcoder settings and the options in the configuration file was replaced with an easier-to-read table.