feat: add structured and configurable logging support to Kubeflow SDK

[MansiSingh17]

Summary

Fixes #85

Changes

• Added kubeflow/common/logging.py with a get_logger() utility function
• Supports configurable log levels via KUBEFLOW_LOG_LEVEL environment variable (defaults to INFO)
• Replaced print() calls with logger.info() in production code:
  • kubeflow/trainer/backends/kubernetes/backend.py
  • kubeflow/trainer/backends/container/backend.py

How to test

Set log level via environment variable:

import os
os.environ["KUBEFLOW_LOG_LEVEL"] = "DEBUG"
from kubeflow.common.logging import get_logger
logger = get_logger(__name__)
logger.info("test")  # 2025-02-21 - __main__ - INFO - test

Notes

• print() calls inside print_packages() were intentionally left as-is since they run inside the training container
• print() calls in docstring examples were also left unchanged

[google-oss-prow]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign kramaranya for approval. For more information see the Kubernetes Code Review Process.

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[github-actions]

🎉 Welcome to the Kubeflow SDK! 🎉

Thanks for opening your first PR! We're happy to have you as part of our community 🚀

Here's what happens next:

• If you haven't already, please check out our Contributing Guide for repo-specific guidelines and the Kubeflow Contributor Guide for general community standards
• Our team will review your PR soon! cc @kubeflow/kubeflow-sdk-team

Join the community:

• Slack: Join our #kubeflow-ml-experience and #kubeflow-trainer Slack channels
• Meetings: Attend the Kubeflow SDK and ML Experience bi-weekly meetings

Feel free to ask questions in the comments if you need any help or clarification!
Thanks again for contributing to Kubeflow! 🙏

[Copilot]

Pull request overview

This PR adds structured logging support to the Kubeflow SDK by introducing a get_logger() utility function in kubeflow/common/logging.py that supports configurable log levels via the KUBEFLOW_LOG_LEVEL environment variable. The PR also replaces print() calls with logger.info() calls in two backend files to improve debugging and observability.

Changes:

• Added kubeflow/common/logging.py with a get_logger() function that provides custom formatting and environment variable-based log level configuration
• Replaced print() with logger.info() in get_runtime_packages() methods of kubernetes and container backends

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 5 comments.

File	Description
kubeflow/common/logging.py	New logging utility with configurable log levels via KUBEFLOW_LOG_LEVEL environment variable
kubeflow/trainer/backends/kubernetes/backend.py	Changed print() to logger.info() for job logs output
kubeflow/trainer/backends/container/backend.py	Changed print() to logger.info() for runtime package logs output

[astefanutti]

/ok-to-test

[MansiSingh17]

It looks like the e2e test failure is related to a CI infrastructure issue with loading the deepspeed-runtime image into KinD, not related to the logging changes in this PR. Could a maintainer re-run the tests or confirm?

[astefanutti]

@MansiSingh17 you're right. It should have been fixed in trainer now.

/retest

[MansiSingh17]

@astefanutti, Fixed the ruff formatting and import ordering issues — all 14 checks are now passing. The only pending one is tide, which is waiting on lgtm/approved labels. Happy to address any other feedback!

[astefanutti]

@MansiSingh17 thanks.

/assign @szaher @kramaranya @andreyvelich