Error deploying AWS EC2 runner in Gitlab

Hey everyone, I am not being able to use the cml runner in my gitlab CI pipeline.

I forked the repo: https://gitlab.com/iterative.ai/cml-cloud-case for a quick test before setting the pipeline in my actual project.

I setted the CI/CD variables needed: I setted both repo_token and REPO_TOKEN because it says lower case in Gitlab’s repo README and upper case in CML Documentation.

And I am getting this error: job.log For me it looks like the job is not getting a “heartbeat” from the runner deployed.

I tried some different machines but right now my gitlab-ci.yml is like this:

stages:
  - deploy
  - train

deploy_job:
  stage: deploy
  when: always
  image: dvcorg/cml
  script:
    - cml-runner
      --cloud aws
      --cloud-region us-east
      --cloud-type t2.micro
      --cloud-hdd-size 64
      --labels=cml-runner-gpu

train_job:
  stage: train
  when: on_success
  image: dvcorg/cml-py3
  tags:
    - cml-runner-gpu

  script:
    - apt-get update -y
    - apt-get install python3-dev -y
    - pip install -r requirements.txt
    - cat gpu_info.txt >> report.md
    
    - cml-send-comment report.md 

One weird thing is that the machine is being created, but the job is falling.

If I could provide any other relevant information please let me know!

Thank you all in advance!

[dacbd]

You can target a specific vpc by creating a security group for it and then use the aws-security-group option. You could follow https://github.com/iterative/cml/pull/831 as it may be relevant to your setup.

[dacbd]

\u001b[32m 😭https://github.com/iterative/cml/pull/818 [^1]

[^1]: not yet released in image