“Create a top-down blue-green production deployment flowchart: an approved release candidate builds an immutable container image, passes integration tests and a critical-CVE scan, then deploys to green for smoke tests. After production approval, switch the load balancer to green and check whether error rate is within the SLO. Successful health drains blue and completes deployment; failed health switches traffic back to blue. Route any test or CVE failure to stopping the release, and a declined approval to holding green.”
Read the decisions behind it.
The same built image is scanned and deployed, avoiding a rebuild between validation and production.
A critical vulnerability stops the release before the green environment is changed.
Smoke tests run in green before the load balancer directs production traffic there.
Error-rate monitoring determines whether blue is drained or traffic is immediately restored to it.
Use this when two production-capable environments sit behind a load balancer. Add your own service-level indicators, approval policy, and database compatibility controls.
Other examples in this family.
Common questions
What makes this blue-green rather than canary deployment?
The flow switches the load balancer from blue to green after green validation, rather than gradually exposing percentages of traffic.
When is blue drained?
Only after green traffic meets the error-rate SLO.
What happens when smoke tests fail?
The release is stopped and the build is fixed; traffic is not switched to green.