👋 hi @<1523701070390366208:profile|CostlyOstrich36> I am running a training pipeline of my model on a remote machine.

The way I do it, is through a cron job that triggers my API and then run a subprocess. The CLI of this subprocess launches the pipeline.

With something like

if __name__ == '__main__':
    args = parse_args(sys.argv)

    if args.is_local:
        PipelineDecorator.run_locally()

    execute_pipeline(...)

where execute_pipeline is decorated with `Pipeli...

I didn't have the time yet. But here is what I found: I have a process that runs a first function in which I init a ClearML Task. Then, later in my process I run a sub process that triggers a ClearML Pipeline (execute_pipeline in my example is triggered). And then the problem happens.
If I don't run the first function that inits a ClearML Task, I don't have the issue. I think it is related to the cache of clearml ?

Hi John, thanks for your answer. Indeed, this is what I did. It forces me to develop some custom archiving process but it sounds acceptable for the moment 👍