What?
Asimov is capable both of generating its own events, log messages, and errors, but is also responsible for monitoring third-party scripts for runtime events.
This merge request intends to implement a new interface for logging events both within the asimov code and within those external programs, so that they can be reported for each individual run and event being analysed.
Why?
In order to maintain flexibility, and so that asimov can run in both "multi-user" and single user modes, the back-end should be capable of using either a performant database engine, or a local database such as sqlite. For smaller projects a normal log file can also be used.
How?
The collector will run as a small server process which will receive log messages as POST requests containing the message, GW event, production name, pipeline and sub-pipeline details, time, and username. The server will allow events to be retrieved for each GW-event, filtered by pipeline and other columns.
Who?
This work was done by Daniel Williams (daniel.williams@ligo.org)
Checklist
-
New code -
Tests -
Documentation & examples -
Release target
Notes
Closes #14