mlcommons / mlcommons/mlcube

mlcube configure with local docker image

Open
#342 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
160
Forks
31
PR merge metrics
No merged PRs in 30d

Description

Running mlcube configure with an MLCube that uses a local docker image that hasn't been pushed yet to the remote registry is throwing an error (and exiting with status 1). Here is the output error:

2023-08-19 01:02:07 hasan-HP-ZBook-15-G3 mlcube.config[42674] WARNING Default runner config contains parameters that are not present in the effective runner config (params=['--mount_opts']). This probably means that a new version of a runner was installed that introduced new parameters.
Error response from daemon: manifest for mlcommons/mock-model-brats:0.0.0 not found: manifest unknown: manifest unknown
2023-08-19 01:02:08 hasan-HP-ZBook-15-G3 mlcube.shell[42674] ERROR Shell.run command='docker pull mlcommons/mock-model-brats:0.0.0' status=256 exit_status=exited exit_code=1 on_error=raise
Failed to configure MLCube with error code 1.
2023-08-19 01:02:08 hasan-HP-ZBook-15-G3 mlcube.__main__[42674] ERROR ERROR:
        message: DockerRun runner failed to configure MLCube.
        description: Error occurred while pulling docker image (docker=docker, image=mlcommons/mock-model-brats:0.0.0).
        context: {'status': 'exited', 'code': 1, 'cmd': 'docker pull mlcommons/mock-model-brats:0.0.0'}
Traceback (most recent call last):
  File "/home/hasan/work/medperf_ws/mlcube/runners/mlcube_docker/mlcube_docker/docker_run.py", line 169, in configure
    Shell.run([docker, "pull", image])
  File "/home/hasan/work/medperf_ws/mlcube/mlcube/mlcube/shell.py", line 104, in run
    raise ExecutionError(
mlcube.errors.ExecutionError: Failed to execute shell command.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/hasan/work/medperf_ws/mlcube/mlcube/mlcube/__main__.py", line 258, in configure
    runner.configure()
  File "/home/hasan/work/medperf_ws/mlcube/runners/mlcube_docker/mlcube_docker/docker_run.py", line 177, in configure
    raise ExecutionError.mlcube_configure_error(
mlcube.errors.ExecutionError: DockerRun runner failed to configure MLCube. Error occurred while pulling docker image (docker=docker, image=mlcommons/mock-model-brats:0.0.0).

The reason is that MLCube tries to pull the image by default (build strategy is pull/auto).

Question: should this command instead fallback to check if the local image exists and only throw a warning instead? Perhaps when build_strategy=auto.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start in runners/mlcube_docker/mlcube_docker/docker_run.py at configure(), then trace the configure entry point in mlcube/mlcube/main.py. Check how build_strategy=auto handles a failed docker pull and local image availability; done means a local-only image can configure without exiting with status 1, with the intended warning or fallback behavior covered.

Written by the indexing model from the issue text.

Assessment

Tech stack
docker, python
Domain
devops
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.