deployment-pipeline-design

Structure the pipeline as build, test, staging deploy, approval gate, then production deploy. In GitHub Actions, use environment protection rules with required reviewers; in GitLab CI, use manual or delayed jobs; in Azure Pipelines, use ManualValidation tasks.

Canary suits high-traffic services needing metric-driven gradual rollout with tools like Argo Rollouts. Blue-green suits high-risk deployments or database migrations where instant switchover and rollback matter, at the cost of temporarily doubled infrastructure.

Argo Rollouts stalls when the AnalysisTemplate query returns no data, leaving the analysis inconclusive. Set inconclusiveLimit so the rollout fails fast instead of hanging, and verify the Prometheus metric names still exist.

Use the expand/contract pattern: add nullable columns in one release, backfill in the next, and drop old columns only after old code is retired. Keep versioned undo scripts alongside forward migrations and avoid destructive changes within a single cycle.

The pipeline is hitting a shallow endpoint that returns 200 even when dependencies are down. Use a deep readiness endpoint that verifies database, cache, and queue connections, and gate promotion on that check.