Recherche de notes de version
documents correspondants
Aucun document correspondant n'a été trouvé
23.10 Media Server
Flussonic Media Server 23.10 release marks a significant milestone in optimizing video streaming services and enhancing user-centric experiences. This update introduces a wide range of improvements, from seamless integration with Prometheus for Kubernetes deployments, to expanded authorization methods for RTSP cameras, and the addition of Vivotek camera support through the ONVIF module. Users can now enjoy a more efficient and organized archive management system, bolstered by enhanced observability through OpenTelemetry trace export. The introduction of push metrics, improved API calls, and streamlined configurations further empower users to proactively manage their video streaming infrastructure. While with the Media Server documentation improvements we contribute to a more user-friendly, informative, and efficient experience for our customers. With these user-centric enhancements and robust security measures, Flussonic continues to elevate the standards of monitoring, scalability, and overall system performance.
Kubernetes:
With the current update we not only enhance operational efficiency and scalability, but also enable seamless customization, integration, and automation, allowing users to proactively manage their video streaming infrastructure for a more reliable and responsive service. The integration of Prometheus monitoring within Flussonic Media Server is a pivotal development for optimizing video streaming services in Kubernetes environments. It ensures the continuous and efficient operation of Flussonic Media Server, providing valuable insights, real-time performance data, and robust troubleshooting capabilities. By offering an API for stream and server metrics in the OpenMetrics format, Flussonic simplifies the process of collecting and analyzing critical data, facilitating precise resource allocation and performance optimization.
Users can gain deeper insights into the performance of their multiplexers and streams, facilitating proactive monitoring and troubleshooting. It enhances operational efficiency, scalability planning, and resource allocation, ensuring that Flussonic Media Server operates optimally within Kubernetes deployments. The inclusion of push metrics in the multiplexer pushes.stats and stream pushes.stats for Flussonic Media Server within Kubernetes environments adds significant value to the monitoring and performance optimization capabilities of the platform. This enhancement enables seamless integration with Prometheus, allowing users to collect and analyze essential push metrics data in real-time.
RTSP Ingest:
The system now tries multiple authorization methods to establish a connection with RTSP cameras, making it compatible with a wider range of cameras and devices that may have different or limited authorization capabilities, ensuring that video streams can be ingested and processed even if they do not support Basic authorization.
Supporting multiple camera brands:
To implement use cases with various camera brands and models to work together in the same surveillance infrastructure, the ONVIF module of Flussonic Media Server was updated to support Vivotek cameras. Those cameras can now be seamlessly incorporated in the surveillance setups on Flussonic Media Server, while the ONVIF module discovers, connects to, and manages Vivotek cameras, making it easier to create comprehensive and interoperable video surveillance solutions.
Archive clean-up:
To maintain a more organized and efficient storage system, particularly when dealing with a large volume of streams and their associated archives, not only old stream archives are properly cleaned up, but any directories that were left empty as a result of the clean-up process are also automatically removed.
OpenTelemetry:
To empower customers to gain deeper insights into server operations, allocate resources more efficiently, and ensure a smoother streaming experience, we added OpenTelemetry trace export in Flussonic Media Server, enhancing observability, troubleshooting capabilities, and performance optimization potential.
In 23.10 we improved observability into troubleshooting and debugging when issues or errors occur (to identify the source of problems and reduce downtime); and performance optimization potential (with access to detailed trace data, customers can pinpoint performance bottlenecks and areas of inefficiency within the live stream API). OpenTelemetry tracing was added to the live stream API.
Multicast streams:
We focus on delivering a reliable and high-quality streaming experience to end-users by addressing potential interruptions or disruptions in the multicast delivery process. With the current update we provide detailed instructions to assist users in establishing and managing an IPTV service that utilizes multicast streaming.
DVR:
By moving settings such as video quality and audio track to the ‘gear’ settings menu in the new Player UI, customers can enjoy a cleaner and more streamlined playback experience. This enhancement reduces clutter on the main interface, making it easier for users to access and adjust these settings when needed, ultimately enhancing their overall viewing experience.
The addition of the “Export to mp4” option in the Player allows users to conveniently save content for offline viewing or sharing. This feature offers greater control over content consumption, facilitating the ability to store and utilize video content in a format that suits their needs, whether it’s for archiving, sharing, or offline playback.
Enhancing visibility of DVR operations, and ensuring security and compliance in video recording processes, ultimately resulting in improved reliability and performance of Flussonic Media Server, we added more tracing spans to monitor and capture specific events within the DVR writing process.
Flussonic Media Server can better allocate its resources to serving active playback requests and efficiently deliver video content without being hampered by irrelevant or outdated locks. This improvement ensures that when episodes are accessible, any outdated DVR locks, which might have previously caused unnecessary resource consumption or delays in handling playback requests, are disregarded.
Flussonic API:
The removal of deprecated parameters, such as vision.gpu, benefits customers by streamlining configuration, reducing the risk of errors, improving documentation, enhancing compatibility and performance, and simplifying troubleshooting. It contributes to a more efficient and secure Flussonic deployment, ultimately leading to a better user experience and system reliability.
For better user experience, scalability, cost-effectiveness, significantly faster access and efficient management of stream-related data, the streams_list API call was optimized to run faster.
In the current update we enhanced user experience, reduced latency, improved resource utilization, through the optimization of the config_get API call to run significantly faster.
With the dramatic speed increase retrieving configuration settings or performing actions related to configuration can be accomplished much more efficiently and quickly, that reduces server load and resource consumption.
Configuration:
With synchronized config saving and applying, users can now see the real-time save progress, even though API calls may become slightly slower. This feature provides a more intuitive and reassuring experience, as users can confirm when their configurations have been successfully saved.
Admin UI:
Enhanced clarity and usability are achieved through the fix to the name of the ‘Active task’ chart in the Pulse tab of Flussonic Admin UI, ensuring that users can readily interpret the data.
Flussonic Coder:
Improved monitoring, resource management, as well as better performance and efficiency in managing Flussonic Coder instances is possible by replacing the ‘Usage’ counter with the ‘Memory Throughput’ metric in the Hardware Modules Monitor. With that, customers gain more insightful data about memory bus load to make more informed decisions regarding resource allocation and system optimization.
Bugs fixed:
DVR:
To prevent excessive network traffic and resource usage, we fixed the issue with a loop when requesting JPEG thumbnails, even when the remote_dvr=nochain parameter is used.
Users can now rely on the calendar to correctly indicate the presence of video archives for all days of the month, making it easier to navigate and access recorded content.
The “Failed to clean blob” error was fixed. It is related to the ongoing DVR clean-up process, which starts another clean-up process while the previous one is already running. With this fixed, troubleshooting and managing the DVR cleaning processes are run more effectively.
Users can now successfully play back recorded video content using the DVR Player, even when DVR recording is configured for the stream. This is essential for surveillance and monitoring applications, as it ensures that recorded footage can be easily accessed and reviewed within the user interface.
The Player component uses PUT requests to interact with sessions or related functionality without encountering server errors. It improves the reliability and functionality of the software, making it possible to update or manage sessions as intended without disruptions caused by HTTP 500 errors.
The embed.html now respects the protocol specified in the query string, ensuring that HLS streams are played correctly when explicitly specified, enhancing the reliability and accuracy of video playback configurations.
Changing the WebRTC playback URL from the deprecated /whap to the proper /whep ensures that the system uses the correct and up-to-date method, reducing potential issues related to deprecated features and improving overall system stability and performance.
Having fixed a performance bug with massive DVR loss, we improved the reliability and functionality of the DVR system to continuously record and store video content without disruptions or data loss issues at specific times, such as the beginning of each hour.
The DVR clean-up process is optimized to run faster and more efficiently, even when episodes and old locks coexist in the archive. This improvement ensures quicker and smoother clean-up operations.
Streaming multimedia content:
A wider range of RTMP streams, including those with float-type resolutions, can be successfully published and processed by Flussonic Media Server. It improves the compatibility and reliability of the streaming process, reducing potential errors and ensuring smoother operation.
Flussonic Media Server becomes more flexible when dealing with RTSP streams. It can now adapt to situations where the source of an RTSP stream redirects the connection to a different location or server. This is valuable for scenarios where redundancy, failover, or load balancing mechanisms are in place, ensuring that Flussonic Media Server can reliably ingest RTSP streams even when the source URL changes temporarily or dynamically due to network conditions or other factors.
Flussonic Media Server can effectively and reliably handle video streams from the specified Bosch camera model (NDP-5512-Z30)l. Users can now expect that the camera RTSP stream will be properly ingested and made available for further processing, monitoring, or distribution within the Flussonic Media Server environment.
Teletext data:
With the teletext data pushed via UDP, the software is capable of dynamically adjusting the bitrate to an optimal level. This prevents the loss of teletext information due to insufficient bitrate, ensuring that all the teletext data is successfully transmitted without trimming to ensure a better viewing experience for users who rely on teletext information alongside television broadcasts.
WebRTC:
The WebRTC Player provides more accurate and timely information to JavaScript-based applications or scripts. When a WebRTC stream publication starts or ends, the ‘onWebsocketClose’ event will now be triggered, allowing developers to respond to these events appropriately.
The intended functionality of the PUT /sessions request from WebRTC Player was restored by allowing it to work without requiring authorization via the auth backend. As a result, Flussonic Media Server can now receive statistics of play sessions accurately.
With enhanced video playback performance to customers using WebRTC WHEP we ensure a smoother and more efficient playback experience, reducing buffering, improving video quality, and providing a better overall viewing experience for users.
Let’s Encrypt SSL certificate:
Let’s Encrypt SSL certificates are automatically updated, even in scenarios where streamer.crt backups exist. By resolving this issue, customers can trust that their SSL certificates remain up to date, reducing the risk of security vulnerabilities and certificate-related disruptions to their services.
Backup running parameter:
Flussonic Media Server now accurately tracks the state of streams that temporarily switch to backup VOD inputs and then return to a live state. This ensures that the ‘backup_running’ parameter is correctly updated, preventing any confusion or misreporting of a stream status.
Video descriptor:
Flussonic Media Server now ensures that the descriptor for video within the PMT is correctly generated when performing HEVC to AVC transcoding. This correction ensures proper stream configuration and alignment, resulting in smoother and more reliable stream output when using this transcoding process.
Transcoder:
Flussonic Media Server no longer requires Flussonic Erlang for the use of the transcoder in generic OEM versions. This change enhances flexibility and simplifies a server installation process, allowing users to utilize the transcoder without additional dependencies.
With 23.10 documentation improvements we contribute to a more user-friendly, informative, and efficient experience for Flussonic Media Server customers, enabling them to make the most of their media streaming solutions and resources.
Prometheus Integration: The addition of an article on using the Prometheus toolkit for collecting metrics from Flussonic Media Server enhances monitoring and analytics capabilities.
MPTS Stream Track Selection: The improved list of tracks available for choosing in an MPTS stream simplifies the process of configuring and managing multiple tracks within media streams.
RTMP Protocol Documentation: The comprehensive summary article about the RTMP protocol and its use with Flussonic, along with grouped menu sections, provides customers with a clear and organized resource for understanding and working with RTMP.
Cloud Archive Storage: Enhanced documentation on storing the archive in a cloud offers customers more detailed insights into cloud-based archival solutions.
Grafana Compatibility: The updated Flussonic Exporter dashboard compatibility with the latest Flussonic API allows customers to seamlessly integrate Grafana for advanced monitoring and visualization of media server metrics.
MP4 Archive Export: The explanation of the archive export to MP4 mechanism, including insights into download delays for long periods, helps users manage and optimize the export process. This transparency ensures a smoother experience when exporting archived content.
Ad Insertion Markers: The updated article on ad insertion markers, including information about marker conversion in Flussonic and support for various transport protocols, empowers customers to efficiently implement ad insertion strategies.