EX294 Coordinate rolling updates Practice Question
An administrator needs to update a web application that runs as a Kubernetes Deployment with 5 replicas. The application is stateless, but the update must not cause any downtime. Which TWO strategies ensure zero-downtime rolling updates?
⚠ Common exam trap
Candidates often confuse `maxUnavailable` and `maxSurge` values, mistakenly thinking that allowing one unavailable pod (maxUnavailable=1) is acceptable for zero-downtime, when in fact it can cause a temporary capacity deficit if the readiness probe is not fast enough.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Set strategy type to RollingUpdate with maxUnavailable=0 and maxSurge=1.
Option B is correct because a RollingUpdate strategy with maxUnavailable=0 and maxSurge=1 guarantees that no existing pod is terminated until a new pod is fully available, so the Deployment always keeps all 5 replicas serving traffic during the update. Option E is correct because a readiness probe tied to the application's health endpoint ensures Kubernetes only routes traffic to new pods once they are actually ready, preventing requests from hitting uninitialized instances during the rollout. Together, maxUnavailable=0/maxSurge=1 plus a readiness probe are the standard mechanism for zero-downtime updates of a stateless Deployment. Option A is wrong because omitting a liveness probe does not prevent downtime and removes Kubernetes' ability to restart unhealthy containers. Option C is wrong because maxUnavailable=1 allows a pod to be taken out of service before its replacement is ready, reducing capacity and risking dropped requests. Option D is wrong because the Recreate strategy terminates all existing pods before creating new ones, causing guaranteed downtime.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Omit the liveness probe from the pod spec.
Why it's wrong here
Removing the liveness probe does not control rollout pacing; it only stops kubelet restarting unhealthy containers, so a bad pod can keep serving traffic during the update. Liveness probes are genuinely relevant to availability, and omitting one would be correct when a probe falsely kills healthy pods.
- ✓
Set strategy type to RollingUpdate with maxUnavailable=0 and maxSurge=1.
Why this is correct
maxUnavailable=0 guarantees all five existing pods stay ready during the rollout, while maxSurge=1 allows one extra pod to be created first. New pods must pass readiness before old ones terminate, preserving continuous availability throughout the update.
- ✗
Set maxUnavailable=1 and maxSurge=0.
Why it's wrong here
With maxSurge=0 no extra pods are created, so each new pod must replace a terminated one, dropping capacity to four replicas and risking downtime for in-flight requests. It is tempting because it conserves cluster resources, and it would be correct when capacity is fixed and brief reduced availability is acceptable.
- ✗
Use the Recreate strategy.
Why it's wrong here
Recreate terminates all five pods before starting replacements, so the Service has no endpoints and requests fail during the gap. It is tempting because it guarantees a clean, uniform state and suits stateful workloads with incompatible schema or storage changes, where a brief outage is acceptable.
- ✓
Configure a readiness probe that checks the application's health endpoint.
Why this is correct
A readiness probe gates traffic: Kubernetes only routes requests to a pod once the probe succeeds, and removes it from Service endpoints when it fails. During rolling updates this prevents sending traffic to pods still starting, which is essential for zero downtime.
Go deeper
Related to this question
About these practice questions
Courseiva writes every EX294 question from scratch — 392 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This EX294 practice question is part of Courseiva's free Red Hat certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the EX294 exam.