Error: REPO_TOKEN does not have enough permissions to access workflow API

[quyetvk]

Hi!

I am trying to create self-hosted runner according to https://dvc.org/blog/cml-self-hosted-runners-on-demand-with-gpus
But I've got the error

{"level":"error","status":"terminated"}
Error: REPO_TOKEN does not have enough permissions to access workflow API
    at CML.repo_token_check (/cml/src/cml.js:236:13)
    at processTicksAndRejections (internal/process/task_queues.js:97:5)
        Destroy scheduled: 30 seconds remaining.
        No TF resource found

My command are:
docker run --name gpurunner --gpus all -e RUNNER_IDLE_TIMEOUT=1000 -e RUNNER_LABELS=cml,gpu -e RUNNER_REPO="https://<my_domain>/gitlab/<my_project_path_without_git_at_end>" -e repo_token="<my_personal_token>" dvcorg/cml-py3

"my_personal_token" was generated by go to "Edit Profile" > "Access Tokens" > create new token (permission is "Api" and I also try to enable all permission)

I executed above command in my AI training server

Self-hosted gitlab server (public) <--------> AI Training Server

Could you please tell me what wrong I did?
Thank you in advanced!

P/S: I also referred to #562

[0x2b3bfa0]

👋🏼 Welcome, @quyetvk! Can you please run again your command using iterativeai/cml:0-dvc2-base1-gpu instead of dvcorg/cml-py3 as the container image? The latter is deprecated.

[0x2b3bfa0]

Closing after a week of inactivity. Feel free to reopen the issue if you keep seeing this error after migrating to the new container.