Here’s the relevant portion of my clearml.conf file located at ~/clearml.conf :

sdk {
    aws {
        s3 {
            key: "AWS_ACCESS_KEY"
            secret: "AWS_SECRET_KEY"
            region: "us-east-1"
            use_credentials_chain: true

            credentials: [
                {
                    bucket: "clearml-s3"
                    key: "AWS_ACCESS_KEY"
                    secret: "AWS_SECRET_KEY"
                    region: "us-east-1"
                    use_credentials_chain: true
                }
            ]
        }

        boto3 {
            pool_connections: 512
            max_multipart_concurrency: 16
        }
    }

    storage {
        default_output_uri: " [None](s3://clearml-s3/) "
    }
}

I'm using a t3a.large instance and running the default Docker Compose file on Ubuntu.

Despite following the documentation, I’m having trouble getting it to work. If anyone has experience setting up S3 with ClearML or can point me in the right direction, I would really appreciate your help!

Thanks in advance!

After copying the clearml.conf file to /opt/clearml/config/ , the error has disappeared, but the files still aren't being reflected on S3.

using below code to upload the file.

task = Task.init(project_name="Test Project", task_name="Test Task", output_uri="

")
# task = Task.init(project_name="Test Project", task_name="Test Task")
task.set_parameter(name='disable_caching', value=True, description='Disable caching for this task')
task.upload_artifact(name="/content/sample_data/README.md", artifact_object={"key": "value"})
# artifact = task.artifacts.get("/content/sample_data/README.md")  
# print(artifact.get_local_copy())

Hey @<1523701070390366208:profile|CostlyOstrich36> could you help me out? I’m having trouble figuring out why it’s not working. I’d really appreciate your help!