My autoscaled instance fails when running "git clone" on a private repo.
I do have the SSH key placed at /root/.ssh/id_rsa on the machine, and when I SSH into the machine and run sudo su; git clone <the repo> it succeeds.
Also, in the extra_vm_bash_script field: I added a whoami command which prints root , so it seems like the user being used to run the git clone during task execution is in fact root .
For context, here's the startup command that the autoscaler runs:
python -m clearml_agent --config-file /root/clearml.conf daemon --queue aws_4gpu_machines --docker python:3.9
Full log included...
Here's a screenshot if a session where I first try to clone as ssm-user , but it fails, then I change to root and it succeeds
I don't see it as an argument in Task.init or Task.execute_remotely
I'm not seeing a extra_docker_shell_script in my clearml.conf generated by clearml-agent init like in this guide
I have the same behavior whether or not I put task.execute_remotely(...) before or after the call to run_shell_script()
