If you're using Helm it would be at the service level in your values.yml , not pod level

As env variables

Hello, @<1523702000586330112:profile|FierceHamster54> !
Can you please clarify, where in the clearml-agent's values.yaml we can specify the credentials? We've found there only fileServerUrlReference

Or we have to go to some other file?
The only other place that I see, is the values.yaml of the main helm chart clearml (not clearml-agent). But there I think we can set the credentials for the default built-in Fileserver, not for some external bucket

Hello @<1523702000586330112:profile|FierceHamster54>
We specified credantials of s3 bucket in clearml.fileserverSecret and clearml.fileserverKey in values.yml of clearml and specified adress of s3 bucket in fileServerUrlReference in values.yml of clearml-agent.

When we run the task remotely we get an error:

clearml.storage - ERROR - Failed creating storage object s3:// Reason: Missing key and secret for S3 storage access ( s3://)
clearml.metrics - WARNING - Failed uploading to s3:// ('NoneType' object has no attribute 'upload')
clearml.metrics - ERROR - Not uploading 1/1 events because the data upload failed

upsy sorry didnt read the entire backtrace

I got some credentials issues to in some pipelines steps and I solved it using

task = Task.current_task()
task.setup_aws_upload(...)

It can allows you to explicitaly specify credentials

Thank you) we will try

@<1523702000586330112:profile|FierceHamster54> sorry, we got the same error ((

from clearml import Task

task = Task.init(project_name='example',  task_name='task template')

task = Task.current_task()
task.setup_aws_upload(bucket="bucket",
                      host="host:port",
                      key="key",
                      secret="secret",
                      multipart=True, secure=False,verify=True)

task.execute_remotely(queue_name='default')

task.upload_artifact("list", [1, 2, 3])

Thanks a lot! @<1523702000586330112:profile|FierceHamster54>

Hum, must be more arcane then, I guess the official support would be able to provide an answer, they usually answer within 24 hours

@<1523702000586330112:profile|FierceHamster54> Should we write to some other chat?

No that's the right one

If you feel you have a specific enough issue you can also post a github issue and link this thread to it