Suche in den Versionshinweisen
Keine passenden Dokumente gefunden
In this release, we are excited to introduce a range of significant enhancements that bolster the performance and efficiency of Central, aligning with our commitment to delivering value-driven solutions.
Firstly, to enhance the efficiency of surveillance systems, empower users with precise configurations and improved performance for more accurate and responsive analytics results, we merged Central and video analytics for a seamless synergy between data management and analysis.
With the latest update Central comes equipped with a script that carries out the automated task of retrieving information about streamers directly from a Kubernetes environment.
Moreover, we significantly worked on a quicker access to video content, real-time monitoring and smoother video service operations through optimized response time for managing a large number of streams by introducing cursors and performing updates in batches.
In the latest update, Central guarantees uninterrupted retrieval of recorded content by eliminating timeouts when accessing DVR episodes. Additionally, improvements to the episode function enhance usability and reliability, delivering meaningful responses even when the ’limit’ parameter is omitted.
In this release, we diligently addressed various bugs and issues, improving the stability and performance of our system to provide a smoother and more reliable user experience.
To streamline the process of gathering crucial data about streamers, ensure a more efficient and simplified workflow for users and administrators, Central now comes equipped with a script that carries out the automated task of retrieving information about those streamers directly from a Kubernetes environment.
Merging Central and Video Analytics
To manage information about persons, new methods are introduced in Central API. Refer to the Central API documentation for more information.
For a simpler configuration management process with a guarantee that video analytics functions are finely tuned to deliver accurate results, we added config external mechanism support for video analytics servers. This mechanism allows Central to efficiently deliver precise stream configurations to video analytics servers, ensuring that they operate with optimal settings.
Adding a filter to retrieve episodes without matched persons is crucial as it streamlines episode management by helping users quickly identify and review relevant footage, optimizing resource utilization and enhancing security.
For centralized data management, improved security and efficient incident investigation, Central is enriched with fetching DVR episodes from video analytics servers. Additionally, it promotes scalability and integration with analytics
It is essential to keep track of individuals detected by the video analytics algorithms and associate them with specific episodes, allowing for easier retrieval and analysis of person-related data within the surveillance system. Thus, Central was enriched with a feature that maintains a record of newly identified or matched persons when it receives a list of episodes from video analytics servers.
In 23.09 to manage and enrich episodes with identification information we added the identification role for a streamer. With this enhancement streamers now have all the necessary permissions to retrieve and add relevant identification data to video episodes. This integration enhances the security, compliance, and efficiency of the identification and enrichment processes within the surveillance or video analytics system. Learn more in Central documentation
Our customers are provided with better scalability and data integrity, data backups for fault tolerance, and analytical systems to efficiently handle increased workloads. All this is possible through a feature of preserving state and data of existing analytics nodes when adding new servers.
To enhance the reliability and availability of the video analytics server by actively monitoring its health and promptly addressing any issues that may arise, reducing downtime and optimizing resource utilization we also included a liveness probe.
Large number of streams:
For a quicker access to video content, real-time monitoring and smoother video service operation response time for a large number of streams was optimized in the latest Central update.
We added cursors to improve loading a huge number of agents.
To reduce the load on the database, maintain database performance and responsiveness when handling a larget number of streams, the stream list update is now performed in batches.
With this fresh update, Central ensures uninterrupted access to the recorded content and facilitates more efficient archive cleanup with returning the DVR episodes list without any timeout.
We made DVR always available to improve simplicity and maintainability by removing the “UI_DVR_ENABLED” variable from the Central configuration file.
To enhance security and customization options, tailoring the streaming experience to the specific needs and requirements of our customers, auth_backends was added to Central for a more flexible management of users permissions to streaming content.
We improved reliability, data integrity, control, and operational efficiency for customers through an optimized interaction of Central and the PostgreSQL database for the controlling server to run correctly.
External database: We fixed an issue and now Central does not fail to start with the external database to perform necessary actions with an initial database.
Central update: We fixed an issue when after an update from master to the actual version Central failed to start
The issue with Central failing to start after an update was resolved, after reviewing the code where this was taking place we made sure that it either points to a valid string or handles NULL values appropriately.
Database migration: Fixed the issue with Central failing database migration when downgrading through fixing migration version check at the startup.
DVR: Since this update DVR cleans up according to the settings, with fixing an issue of episode handling in terms of cursor operations (their retrieving, navigating or managing).
The fix for Flussonic Central issue with handling PUT requests for DVR episode configurations, which previously expected the optional episode_type field, enhances the software reliability and reduces the risk of configuration errors.
Config data: Optimizing the delivery of configuration data to video analytics servers in Flussonic improves system performance, reduces resource usage, and enhances the quality and timeliness of video analytics results.
Preview: We added a “preview” field to the PUT and GET requests to provide more flexibility and functionality when working with episodes. Preview data can now be included and retrieved conveniently for content managing, quality control, or providing preview options to end-users.
Streamer: The problem related to the transfer of the streams with video analytics enabled from one streamer to another with the ‘vision’ module installed was fixed. The fix ensures the correct transfer of video analytics-enabled streams during reprovisioning, promoting system stability, data continuity, and overall operational efficiency.
We fixed an issue with effective_streamer_hostname not updating as it should be, to ensure accurate resource routing and system reliability.
Central updates streamer configuration without restart no matter if there are any changes, that improves adding auth_backends to the configuration.
Episodes: In this update we improved the function when an empty list of episodes was returned if the
limit isn’t specified. The current improvement provides a meaningful response even when a ‘limit’ parameter is not provided, thus enhancing usability and reliability of the platform.