Deployment and Delivery
Deployment and Delivery covers how tested changes reach users safely and repeatedly. The exam tests continuous delivery and deployment, deployment pipelines, DevOps practices, release techniques and production feedback.
Core rules and concepts
- Continuous delivery keeps every change in a deployable state; continuous deployment automatically releases every change that passes the pipeline.
- A deployment pipeline stages work: commit build, unit tests, integration tests, acceptance tests, non-functional tests, staging, production. Each stage is a gate that gives fast feedback.
- DevOps is often summarised by CALMS: Culture, Automation, Lean, Measurement, Sharing.
- Release techniques limit blast radius: blue-green deployment switches between two environments; canary release sends a small percentage of traffic first; feature toggles separate deployment from release.
- Configuration management covers code, test data, environments and infrastructure as code, so builds are reproducible.
- Production is a test environment too: monitoring, synthetic transactions, smoke tests and error budgets provide feedback.
Worked example
A team releases a checkout change with a canary at 5% of traffic. They watch error rate and latency against the baseline. Reasoning: a canary limits the number of users affected if the change is faulty, and real traffic exposes issues that staging cannot. If the error budget is breached, they roll back or disable the feature toggle. The decision is driven by measured signals, not by opinion.
Traps
- Treating continuous delivery as continuous deployment.
- Adding manual approval gates to every stage, destroying fast feedback.
- Letting environments drift from production.
- Testing only before deployment and ignoring production signals.