I suggest to implement new Task Status: Finished With Warning. It works the same way as Crashed Tasks.
Possible goals:
- To notify that some rare case (logic for which you wanted to implement later probably due the lack of examples) occured.
- Third-party tool (e.g. sandbox, config extractor, etc.) succeeded to return the result but also returned a warning.
- Your ML said that this file is Benign/Malicious but with a low confidence -- probably requires some hand validation.
Way of implementation: as Task.warnings list:
...
result = generic_proceed_logic(...)
if result.warning:
self.current_task.warnings.append(result.warning)
...
In most cases, we can live without it just by raising the exception at the end of work, but having warnings instead of crashes can add more QoL in the process of improving services.
I suggest to implement new Task Status: Finished With Warning. It works the same way as Crashed Tasks.
Possible goals:
Way of implementation: as
Task.warningslist:In most cases, we can live without it just by raising the exception at the end of work, but having warnings instead of crashes can add more QoL in the process of improving services.