or have I got this wrong, and it's the clearml-agent that needs to read/write to S3?

Hi @<1687643893996195840:profile|RoundCat60>

I think the best way will be to configure a default output_uri to be used by all tasks:  None , under default_output_uri just write your bucket path ( None ).
When using S3/google storage/azure, you will also need to add your credentials in this section - None (s3 in this link),
and install clearml with storage requirements:
pip install clearml[s3] for S3
pip install clearml[azure] for azure
pip install clearml[gs] for google storage

thank you

@<1687643893996195840:profile|RoundCat60> Looks like the docs have not caught up yet with recent structure change in the repo which renamed the 'server' folder to 'apiserver'.
So... the correct link would be None

Same credentials configuration for the ClearML-Agent.
Notice that when a task is created, in the UI, under EXECUTION tab, you can find (and change if you like) the output destination.

yep still referring to the S3 credentials, somewhat familiar with boto and IAM roles

Was asking about using iam roles without keys

As we can’t create keys in our AWS due to infosec requirements

Hmmm

Is it possible to use an IAM role rather than user credentials in the clearml.conf file?

None

@<1523701205467926528:profile|AgitatedDove14> - any thoughts on this. Would like to use profile / iam roles as well.

I suggest a bump in the GitHub issue

Still not supported 😞

None
So this is the only place we need to change to support it, do you feel like messing around with it and adding IAM roles ?

As we can’t create keys in our AWS due to infosec requirements

is there any documentation for connecting to an S3 bucket?

Let me check

@<1687643893996195840:profile|RoundCat60> I'm assuming we are still talking about the S3 credentials, sadly no 😞
Are you familiar with boto and IAM roles ?