To help future analysis software decide whether a feature was present at the time data was written, various components of the stack should insert their own software and hardware versions as metadata that will eventually make it to Nexus file.
For example:
kafka_dae_control should stream it's own version (maybe as part of metadata of RunStart)- When
kafka_dae_control connects to VXI boards and/or FPGAs it should attempt to pull back some kind of version identifier/build date and stream those as part of a runstart too - How do we get this from
event_udp_to_kafka and similar programs? - Filewriter should self-describe:
- e.g.
/raw_data_1/program_name = <name_of_filewriter> - Attach a version attribute to that dataset to describe release/revision/build date etc
- Maybe something in the forwarder / forwarder-configurer to describe that version number
- If possible also pull back IBEX server version.
To help future analysis software decide whether a feature was present at the time data was written, various components of the stack should insert their own software and hardware versions as metadata that will eventually make it to Nexus file.
For example:
kafka_dae_controlshould stream it's own version (maybe as part of metadata of RunStart)kafka_dae_controlconnects to VXI boards and/or FPGAs it should attempt to pull back some kind of version identifier/build date and stream those as part of a runstart tooevent_udp_to_kafkaand similar programs?/raw_data_1/program_name = <name_of_filewriter>