gitlab-org/gitlab-foss v17.8.0

on GitLab

Previously, when you suspended Flux reconciliation from the dashboard for Kubernetes, there was no clear indicator of the suspended state. We've added a new "Paused" status to the existing set of status indicators, making it clear when Flux reconciliation is suspended and providing better visibility into the state of your deployments.

On the dashboard for Kubernetes, finding specific pods in large deployments can be time-consuming. A new search bar lets you quickly filter pods by name. The search works across all available pods, and you can combine it with status filters to find exactly the pods you need to monitor or troubleshoot.

You can now keep track of multiple discussions and mentions within a single issue or merge request. With the new multiple to-do items feature, you'll receive separate to-do items for each mention or action, ensuring you don't miss important updates or requests for your attention. This enhancement helps you manage your work more effectively and respond to your team's needs more efficiently.

Project creation can be restricted to specific roles in a group using the Allowed to create projects setting. The Owner role is now available as an option, enabling you to restrict new project creation to users with the Owner role for the group. This role was previously unavailable in the selection options.

Thank you @yasuk for this community contribution!

While GitLab has long supported creating releases from Git tags and tracking deployments, this information previously lived in multiple separate places that were difficult to piece together. Now, you can see all deployments related to a release directly on the release page. Release managers can quickly verify where a release has been deployed and which environments are pending deployment. This complements the existing deployment page integration that shows release notes for tagged deployments.

We would like to express our gratitude to Anton Kalmykov for contributing both features to GitLab.

You can now set a primary domain in GitLab Pages to automatically redirect all requests from custom domains to your primary domain. This helps maintain SEO rankings and provides a consistent brand experience by directing visitors to your preferred domain, regardless of which URL they initially use to access your site.

Administrators can now control pipeline resource usage by setting CI/CD limits for their GitLab Community Edition installations. Previously, this feature was only available in GitLab Enterprise Edition.

We're thrilled to introduce support for protected PyPI packages, a new feature designed to enhance the security and stability of your GitLab package registry. In the fast-paced world of software development, accidental modification or deletion of packages can disrupt entire development processes. Protected packages address this issue by allowing you to safeguard your most important dependencies against unintended changes.

From GitLab 17.8, you can protect PyPI packages by creating protection rules. If a package is matched by a protection rule, only specified users can update or delete the package. With this feature, you can prevent accidental changes, improve compliance with regulatory requirements, and streamline your workflows by reducing the need for manual oversight.

We're thrilled to announce the rollout of protected container repositories, a new feature in GitLab's container registry that addresses security and control challenges in managing container images. Organizations often struggle with unauthorized access to sensitive container repositories, accidental modifications, lack of granular control, and difficulties in maintaining compliance. This solution provides enhanced security through strict access controls, granular permissions for push, pull, and management operations, and seamless integration with GitLab CI/CD pipelines.

Protected container repositories offers value to users by reducing the risk of security breaches and accidental changes to critical assets. This feature streamlines workflows by maintaining security without sacrificing development speed, improves overall governance of the container registry, and provides peace of mind knowing that important container assets are protected according to organizational needs.

This feature and the protected packages feature are both community contributions from gerardo-navarro and the Siemens crew. Thank you Gerardo and the rest of the crew from Siemens for their many contributions to GitLab! If you are interesting in learning more about how Gerardo and the Siemens crew contributed this change, check out this video in which Gerardo shares his learnings and best practices for contributing to GitLab based on his experience as an external contributor.

Data scientists and Machine Learning engineers primarily work in Python environments, but integrating their machine learning workflows with GitLab's MLOps features often requires context switching and understanding of GitLab's API structure. This can create friction in their development process and slow down their ability to track experiments, manage model artifacts, and collaborate with team members.

The new GitLab MLOps Python client provides a seamless, Pythonic interface to GitLab's MLOps features. Data scientists can now interact with GitLab's experiment tracking and model registry capabilities directly from their Python scripts and notebooks. The client includes:

• GitLab Experiment Tracking: Easily track machine learning experiments within GitLab.
  
• Model Registry Integration: Register and manage models in GitLab's model registry.
  
• Experiment Management: Create and manage experiments directly from the client.
  
• Run Tracking: Initiate and monitor training runs with ease.
  

This integration allows data scientists to focus on model development while automatically capturing their ML lifecycle metadata in GitLab. The Python client works seamlessly with existing ML workflows and requires minimal setup, making GitLab's MLOps features more accessible to the data science community.

We welcome the wider Python and data science community to contributions and share feedback directly in our project's repository

When creating machine learning models, data scientists often experiment with different parameters, configurations, and feature engineering to improve the performance of the model. Keeping track of all this metadata and the associated artifacts so that the data scientist can later replicate the experiment is not trivial. Machine learning experiment tracking enables them to log parameters, metrics, and artifacts directly into GitLab, giving easy access later on while also keeping all experimental data within your GitLab environment. This feature is now generally available with enhanced data displays, enhanced permissions, deeper integration with GitLab, and bug fixes.