Skip to content

Platform Improvements and Changes

Jump to bottom
Yvette Nartey edited this page Feb 4, 2025 · 2 revisions

Overview

This document outlines the technical improvements and changes made to the platform(mainly backstage) after taking over its development. This aimed to improve the reliability, scalability and usability.

📌 Updated on 04.02.2025

🔷 Added

  • Backstage Backend Migration: Successfully migrated the Backstage backend to a new architecture for better maintainability and performance.

  • Addressed security Vulnerabilties: Some dependencies posed security threats the help of dependabot, upgraded versions of dependencies.

  • Google and GitHub Authentication: With the migration of the Backstage Backend, this led to refractoring the custom OAuth2 authentication via Google and GitHub, allowing users to log in securely using their existing credentials.

  • Documentation Updates: With the aim of usuability in mind, I updated mainly the User Guide for the Developer. This can be found here -> User Guide

  • Observabilty:

    • Logging: The backstage backend supplies a central logging service.
    • Health Checks: The new backend provides a RootHealthService which implements /.backstage/health/v1/readiness and /.backstage/health/v1/liveness endpoints to provide health checks for the entire backend instance.

🔶 Changed

Changed

  • Improved End-to-End (E2E) Tests: Enhanced test coverage to ensure a more reliable deployment pipeline, improving test execution speed and accuracy.

  • Optimized Docker Images:

    • Reduced image size from 5.1GB → 1.4GB, significantly improving deployment efficiency. Multi-stage builds were useful to optimize the Dockerfile.
    • Continued usage of Google Artifact Registry for streamlined container management.

  • Refined Continuous Deployment (CD) Workflow:

    • Improved build caching for faster deployments.
    • The smaller size in docker also played a vital role in reducing the docker build times and pipelines workflow times from over 10 mins to under 5 mins.

Next Steps

  • You can integrate a plugin tool like Prometheus and Grafana to improve system observabilty.
  • Setup Datadog to measure DORA Metrics for the system.
  • Conduct an extensive round of user testing post-backend and post-authentication changes.

Appendices

Please contact Yvette Nartey for questions with regards to the new Backstage setup.

References

Clone this wiki locally