My bad, there is a mixture in terms.
"configuration object" is just a dictionary (or plain text) stored on the Task itself.
It has no file representation (well you could get it dumped to a file, but it is actually stored a s a blob of text on the Task itself, at the backend side)

checking it 😄

\o/

today I'm in the middle of sprint plannings for my team so I will not be probably fast to help if needed, but feel free to ping me just in case (I will try to do my best)

I mean this blob is then saved on the fs

It can if you do:
temp_file = task.connect_configuration('/path/to/config/file', name='configuration object is a config file')Then temp_file is actually a local copy of the text coming from the Task.
When running in manual mode the content of '/path/to/config/file' is stored on the Task When running remotely by the agent, the content from the Task is dumped into a temp file and the path to the file is returned in temp_file

?

I have a theory

👍

I finally found where the issue is, I opened an issue on gh so it's more manageable:
https://github.com/allegroai/clearml/issues/273

I also found a not so good (at least for me) behaviour:
https://github.com/allegroai/clearml/issues/272

then I enqueue it and it's created but obv empty

Good, so we narrowed it down. Now the question is how come it is empty ?

yaya confirmed

My pleasure

btw it's on the filesystem

maybe this can cause the issue?

If the manual execution (i.e. pycharm) was working it should have stored it on the Pipeline Task.

perfect, now the whole process is clear to me

got it

this configuration object is stored as a file in /root/.trains ?

and I will give you feedback here

Hi JuicyFox94
I think you are correct, this bug will explain the entire thing.
Basically what happens is that remote_execute stops the local run before the configuration is set on the Task. Then running remotely the code pull the configuration, sees that it is empty and does nothing.
Let me see if I can reproduce it...

I mean this blob is then saved on the fs

Sure thing 🙂

ok the issue must be there, After first creation nothing is there

what is?

maybe this can cause the issue?

Not likely.
In the original pipeline (the one executed from the Pycharm) do you see the "Pipeline" section under Configuration -> "Config objects" in the UI?

a big ty for now

I have multiple agents not sharing /root/.trains

No worries, just found it. Thanks!
I'll make sure to followup on the GitHub issue for better visibility 🙂