Skip to content

LUNA PLATFORM v.5.76.4#

LP changes

  • The Docker Compose specification version included in the distribution package has been updated from 3.6 to 3.9.

    Docker Compose version 1.25.0 or later is required to use the 3.9 specification.

  • A Standalone-lambda example has been added to the Lambda developer guide, which implements the functionality of comparing faces from video with faces from a list.

  • Identical errors for each stream will now be sent as feedback to the Video Manager service no more than once every 5 seconds.

    This change simplifies the logs, reducing the frequency of repeated errors and improving their readability.

LP fixed errors

  • Fixed an issue where the influx2_cli.py script, which includes the collection of aggregated statistics in InfluxDB, did not ignore the existence of objects by default.

    This caused an error when the script was rerun during an update if statistics collection had already been enabled during the initial setup of LUNA PLATFORM (i.e., the luna-admin bucket had been created in InfluxDB). The script now includes an --ignore-integrity flag to ignore the existence of objects, which is enabled by default.

  • Fixed the "ContentLengthError: Not enough data to satisfy content length header" error that occurred when making a cross-matching request using the Python Matcher Proxy service with a large number of descriptors.

  • Fixed error handling during decoding.

    Previously, the Remote SDK service would return an Internal server error when making a "videosdk" request if a video decoding error occurred.

    Now, in this case, the service returns an error with code 43004, description "Error occurred while decoding video," and status code 400.

  • Fixed an issue where incorrect timestamps were received when processing some videos without specifying the "pts" parameter in the "videosdk" request.