Practice & Reference · Exam Simulator

Exam Simulator

20 original practice questions, timed at 20 minutes, covering the real domains behind the three certifications on the Certifications page — AWS DevOps Engineer Professional, the Certified Kubernetes Administrator, and the Terraform Associate. These are not reproductions of any vendor's actual exam questions (that would be against every vendor's terms and wouldn't teach you anything real) — they're original questions testing the same underlying knowledge, so a good score here is a real signal, not a false one.

☺ Explain it like I'm 10

It's a practice fire drill, not the real fire. Nobody grades you and nothing is on your permanent record — you get a clock, twenty questions, and an honest score at the end so you know exactly what to study next.

🐢🦥Your hosts for this topic: Timmy the Turtle & Sol the Sloth — Timmy times you and checks every answer, Sol reminds you that rushing a DORA-metrics question is how good engineers get an easy one wrong.

Before you start

☺ Like you're 10: One question at a time, twenty minutes on the clock, and you can go back and change an earlier answer before you submit.

20 questions, single-select, one at a time — use Previous and Next to move around before submitting, your answers are kept as you go. The 20-minute timer starts the moment you click start and auto-submits when it hits zero, the same pressure a real timed exam applies. Your best score is saved on this device only — nothing is sent anywhere.

Q1. Your team practices Continuous Delivery, not Continuous Deployment. What does that mean in practice?

Every commit that passes the pipeline is automatically released to production with no human involved. Every commit is automatically built, tested, and verified as release-ready, but a human decides when it actually ships. Code is merged to main only once a week in a scheduled release train. There is no automated testing; a QA team manually verifies each release.

Q2. In a blue-green deployment, how does traffic move from the old version to the new one?

Gradually, a small percentage at a time, while watching error rates. All at once — the router/load balancer cuts over from the old environment to the fully-provisioned new one. One server at a time is drained, updated, and returned to the pool. Traffic is duplicated to both environments and only the new environment's response is logged.

Q3. Which of the four DORA metrics measures the percentage of deployments that cause a degradation in production requiring remediation?

Deployment frequency Lead time for changes Change failure rate Time to restore service

Q4. You run terraform plan and see a resource will be destroyed and recreated instead of updated in place. What is the most likely cause?

You changed an attribute that requires replacement (a "ForceNew" attribute for that resource type). Terraform always destroys and recreates resources on every apply. The state file is missing. You ran terraform apply without first running init.

Q5. On the CKA exam, which resource lets you read the official Kubernetes documentation during the test?

No outside resources are allowed at all — it is fully closed-book. A single open browser tab to kubernetes.io/docs is permitted. Any website is allowed, including Stack Overflow and blog posts. A printed reference sheet is provided instead of internet access.

Q6. What is the core difference between configuration management and infrastructure as code?

They are two names for the exact same activity. IaC provisions the machines/resources themselves; configuration management configures software on machines that already exist. Configuration management only applies to containers; IaC only applies to virtual machines. IaC is always pull-based; configuration management is always push-based.

Q7. A canary release is best described as:

Running two full environments and cutting all traffic over instantly. Releasing a new version to a small percentage of real traffic first, watching metrics, then gradually expanding. A business experiment comparing two designs to see which converts better. Deploying to every server in the fleet simultaneously.

Q8. Why does a saved Terraform plan file (terraform apply tfplan) matter for a safe approval workflow?

It applies exactly the reviewed diff with no re-prompt, so what was approved is what actually happens. It automatically emails the plan to the security team. It encrypts the state file. It is required before terraform init can run.

Q9. What does a Kubernetes readiness probe actually gate?

Whether the container image is scanned for vulnerabilities. Whether the container gets restarted if it hangs. Whether the pod receives traffic from a Service — an unready pod is removed from the routing endpoints. Whether the node is allowed to join the cluster.

Q10. Conway's Law, applied deliberately (the "reverse Conway maneuver"), suggests that to get a decoupled microservice architecture, you should:

Hire more senior architects. Restructure your teams into small, decoupled units first, since system architecture tends to mirror team communication structure. Buy a more expensive CI/CD tool. Write the architecture diagram before writing any code.

Q11. Which of these is the best example of "toil" as SRE/DevOps literature defines it?

Designing a new service's architecture. Manually restarting the same crashed service every few days with no permanent fix. Writing automated tests for a new feature. Reviewing a colleague's pull request.

Q12. In Terraform, what is the practical difference between count and for_each for creating multiple resource instances?

They are identical in every way. count indexes instances by position in a list (reordering causes churn); for_each keys instances by a stable string from a map/set. count only works with AWS providers. for_each cannot be used inside a module.

Q13. Which of the four golden signals measures how "full" a service's resources are (e.g. CPU, memory, queue depth near their limit)?

Latency Traffic Errors Saturation

Q14. Why do elite-performing DevOps teams (per the Accelerate/DORA research) manage to deploy far more often with a lower change failure rate, rather than trading speed for stability?

They hire only senior engineers. Small, frequent, well-tested changes are inherently lower-risk than large, infrequent, undertested ones — speed and stability reinforce each other rather than trading off. They skip code review to move faster. They use a single specific vendor's tools.

Q15. An "immutable artifact" / "build once, promote everywhere" approach means:

The same tested binary/image moves from staging to production unchanged, rather than being rebuilt for each environment. Artifacts are rebuilt fresh for every environment to pick up the latest dependencies. Artifacts are never versioned. Artifacts are stored only on the build server, never in a registry.

Q16. What does Terraform state locking prevent?

Unauthorized read access to the state file. Two concurrent terraform apply runs from corrupting the state by writing at the same time. Resources from being destroyed accidentally. Providers from being upgraded.

Q17. During a live incident, what is the primary job of the Incident Commander?

To personally write the code fix. To coordinate the response and make decisions — not necessarily to be the one debugging. To handle only external customer communications. To decide who gets blamed once the incident is resolved.

Q18. Which best describes trunk-based development?

Long-lived feature branches that merge to main only at the end of a release cycle. Short-lived branches merged to main frequently (often daily), often behind feature flags. Committing directly to production servers with no version control. A branching model with separate develop, release, and hotfix branches held indefinitely.

Q19. What is a sidecar container?

The main application container in a pod. A helper container running alongside the main one in the same pod, e.g. for logging or a service mesh proxy. A backup copy of the entire cluster. A container that only runs during cluster upgrades.

Q20. Why is "pipeline as code" (the CI/CD pipeline definition stored and versioned alongside the application) preferred over configuring the pipeline through a UI?

It's the only way to run tests at all. UI-configured pipelines are always faster. It gets version history, code review, and reproducibility for the pipeline itself, the same discipline applied to the app. UI configuration is deprecated by every CI vendor.

⚠ What this is and isn't

A good score here means you know the underlying concepts, not that you're guaranteed to pass a specific vendor's exam — CKA and similar performance-based exams also test hands-on speed at a keyboard, which no multiple-choice quiz can simulate. Pair this with real practice: spin up a free-tier cluster or a local one and actually run the commands from the toolchain and checklist pages.

Exam Simulator

Before you start

Your results