Yeah, I saw that. I didn’t see where to specify the bucket/prefix of the desired cache. It seems like this is for when the user (me) calls Storage.upload or whatever with a specific s3 Uri. I’m looking for s3 storage of all artifacts and return objects and everything.

Am I missing something?

I’m not sure yet . It’s my first day playing around with it .
As for the s3 storage, that would make the local cache only for locally run tasks, right? As in, the server stores the data on s3 and the agent will populate the local cache as needed?

Hi @<1715538373919117312:profile|FoolishToad2> ,
Storage is used by the clients, to set up S3 storage, see here .
Changes to the server usually require a restart of the docker container - what exact changes are you thinking of?

@<1715538373919117312:profile|FoolishToad2> , I think you're missing something. ClearML backend only holds references (links) to artifacts. Actual interaction with storage is done directly via the SDK, aka on the machine running the code