Service MOT

All, Maintain, Active Development

Runtime

• Make sure either the latest medium-term support (MTS) or long term support (LTS) is in use.
• Redeploy to get the latest patches (make sure base image is up to date)
• Runtime development runtime is the same as production (with the use of a pinned version - i.e., .npmrc / .sdkmanrc file)
• Dockerfile is reviewed

Development

• Run the app locally making sure it is easy to do following the readme
• Fix code quality issues (Sonar, FindBugs)
• Upgrading build tool (e.g. Gradle) - Patch only (unless out of support)
• Remove old feature flags
• Make sure dependencies are up to date - Patch only (unless out of support)

Observability

• Review Alerts
• Review Dashboards
• Review Logging (no Debug/Info levels in Prod)

CI

• Make sure the deployment still works
• Make sure the build still works
• Releases create a REL ticket
• Check bgcloud.gradle
• Releases send a Slack notification to #warp-releases
• Check bgcloud.gradle
• The build process is clear and understandable
• Check yarn/npm scripts

Testing

• Tests are run in the pipeline
• No skipped tests
• Tests abide by the testing pyramid

Documentation

• Review Runbook
• Review README.md
• Review CONTRIBUTING.md
• Review directory.transport.dev
• API endpoints provide Swagger documentation

Resiliency

• Timeouts
• Circuit Breakers
• Bulkheads

Version Control

• Remove any dangling branches
• Check the latest changes have been deployed

Upgrades

This is an ordered list of ways to improve a service.

1. Major upgrade of framework
2. Major upgrade of build tool
3. Service Mesh
4. Major upgrade of libraries