Which sequence describes a blue-green deployment?

Which sequence describes a blue-green deployment?

• A. Build in place, restart services, do not switch traffic.
• B. Prepare new/green environment, deploy, run tests, switch user traffic to green, monitor, rollback by switching back if needed.
• C. Copy data, run tests, delete old environment, switch DNS.
• D. Deploy to production, inform users, rollback automatically after 24 hours.

Answer: (B)

Blue-green deployment is about having two production-like environments—one serving live traffic and the other prepared to take over. The sequence in question matches this approach: you create and prepare a new, green environment, deploy there, and run tests in that separate setup. When everything looks good, you switch all user traffic to the green environment, then monitor it. If issues appear, you can roll back quickly by routing traffic back to the previous environment. This minimizes downtime and provides a fast, safe rollback path. The benefit of this pattern is clear: you never deploy into the live environment without an already validated copy ready to take over, and switching traffic is the primary lever to go live or revert with minimal disruption. In practice, the switch is often done through a load balancer or DNS change, and monitoring is essential to catch any problems after the switch. The other options don’t describe blue-green deployment. Building in place and not switching traffic means there’s no separate idle environment to take over, so rollback isn’t as seamless. Deleting the old environment and using DNS alone can create risks and service disruption, and deploying directly to production with a time-bound rollback doesn’t establish the two-identity, switch-over workflow that defines blue-green.

Blue-green deployment is about having two production-like environments—one serving live traffic and the other prepared to take over. The sequence in question matches this approach: you create and prepare a new, green environment, deploy there, and run tests in that separate setup. When everything looks good, you switch all user traffic to the green environment, then monitor it. If issues appear, you can roll back quickly by routing traffic back to the previous environment. This minimizes downtime and provides a fast, safe rollback path.

The benefit of this pattern is clear: you never deploy into the live environment without an already validated copy ready to take over, and switching traffic is the primary lever to go live or revert with minimal disruption. In practice, the switch is often done through a load balancer or DNS change, and monitoring is essential to catch any problems after the switch.

The other options don’t describe blue-green deployment. Building in place and not switching traffic means there’s no separate idle environment to take over, so rollback isn’t as seamless. Deleting the old environment and using DNS alone can create risks and service disruption, and deploying directly to production with a time-bound rollback doesn’t establish the two-identity, switch-over workflow that defines blue-green.