Hi @<1541954607595393024:profile|BattyCrocodile47>

I

do

have the SSH key placed at

/root/.ssh/id_rsa

on the machine,

Notice that the .ssh folder is mounted from the host (EC2 / GCP) into the container,

'-v', '/tmp/clearml_agent.ssh.cbvchse1:/.ssh'

This is odd, why is it mounting it to /.ssh and not /root/.ssh ?

I can't think of any changes we might have made on our side to cause that 🤔

DM me the entire log, I would assume this is something with the configuration

Let's see. The task log? I think this is it.

Or the log of the init script?

cc: @<1565509803839590400:profile|MoodyBear54>

I

do

have the SSH key placed at

/root/.ssh/id_rsa

on the machine,

@<1541954607595393024:profile|BattyCrocodile47> is the SSH key part of the containers? or are you saying it is on the EC2 instance ?

So, we've been able to run sudo su and then git clone with our private repos a few times now

That's with the key at /root/.ssh/id_rsa

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

The key seems to be placed in the expected location

That's with the key at

/root/.ssh/id_rsa

You mean inside the container that the autoscaler spinned ?
Notice that the agent by defult would mount the Host .ssh over the existing .ssh inside the container, if you do not want this behavior you need to set: agent.disable_ssh_mount: true in clearml.conf

Let's see. The screenshots above are me running on the host, not attaching to a running container. So I believe I do want the keys to be mounted into the running containers.

I do agree with your earlier observation that the target of that mount seems wrong. I would think that the volume mount should be -v /root/.ssh:/root/.ssh but instead it's -v /root.ssh:/.ssh

Actually that's wrong: really this is the current volume mount

'-v', '/tmp/clearml_agent.ssh.cbvchse1:/.ssh',

Could changing these values to /root/.ssh work? Do you know what use within the docker image ClearML is using?

'-v', '/tmp/clearml_agent.ssh.cbvchse1:/.ssh',

It's my bad, after that inside the container it does cp -Rf /.ssh ~/.ssh
The reason is that you cannot know the user home folder before spinning the container
Anyhow the point is, are you sure that you have ~/.ssh on the Host machine configured?
And if you do, are you saying this is part of your AMI? if not how did you put it there?

So here's a snippet from my aws_autoscaler.yaml file

configurations:
  extra_clearml_conf: ""
  extra_trains_conf: ""
  extra_vm_bash_script: |
    aws ssm get-parameter --region us-west-2 --name /clearml/github_ssh_private_key --with-decryption --query Parameter.Value --output text > ~/.ssh/id_rsa && chmod 600 ~/.ssh/id_rsa
    source /clearml_agent_venv/bin/activate

hyper_params:
  iam_arn: arn:aws:iam::<my account id>:instance-profile/clearml-2-AutoscaledInstanceProfileAutoScaledEC2InstanceProfile56A5348F-90fmf6H5OUBx

It's an Amazon Linux AMI with the AWS CLI pre-installed on it. It uses the AWS CLI to fetch the key from AWS SSM Parameter Store. It's granted read access to that SSM Parameter via the instance role.

oh that makes sense.
I would add to your Task's docker startup script the following:

ls -la /.ssh
ls -la ~/.ssh
cat ~/.ssh/id_rsa

Let's see what you get

On it

Actually, dumb question: how do I set the setup script for a task?

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

Here we go. Trying with this

It doesn't seem to want to show me stdout

Trying as a python subprocess...

So I get output with this one, but the console only shows me the output from my machine. For example, the SSH key is present, and whoami results in ericriddoch

I have the same behavior whether or not I put task.execute_remotely(...) before or after the call to run_shell_script()

Actually, dumb question: how do I set the setup script for a task?

When you clone/edit the Task in the UI, under Execution / Container you should have it
After you edit it, just push it into the execution with the autoscaler and wait 🙂