Free EX288 Practice Questions

Question 1

Minutes after a Service manifest is changed, the orders Route begins returning HTTP 503. Its TLS handshake succeeds. All three orders Pods are Ready, declare containerPort: 8080, and answer HTTP requests at their Pod IPs on port 8080. The Route still targets the Service's port named http. The relevant Service fields are:

selector: {app: orders}
ports:
  - name: http
    port: 80
    targetPort: 8081

The Service's EndpointSlices contain the three expected Pod IPs. What is the smallest correction that restores the intended request path?

Accepted Answer

Change the Service's targetPort from 8081 to 8080.

Answer

Change the containers' declared containerPort from 8080 to 8081.

Answer

Change the Service's port from 80 to 8080, retaining targetPort 8081.

Answer

Roll back the application Deployment before making further networking changes.

Question 2

A support engineer changes LOG_LEVEL from info to debug in a mutable ConfigMap. One running container consumes that key in three ways: an environment variable, a file in an ordinary ConfigMap directory mount, and a separate file mounted with subPath. No container restarts, no controller changes the Pod template, and the ordinary directory projection has refreshed. The engineer checks the process environment and freshly reads both files. In that order, which values should appear?

Accepted Answer

info in the environment; debug in the directory file; info in the subPath file.

Answer

debug in the environment; debug in the directory file; debug in the subPath file.

Answer

info in the environment; info in the directory file; info in the subPath file.

Answer

info in the environment; debug in the directory file; debug in the subPath file.

Question 3

During a brief database failover, a checkout process remains responsive but cannot serve transactions. It reconnects automatically when the database returns. Its /live endpoint reports event-loop health; /ready reports whether transactions can be served. Both endpoints are implemented correctly. How should health monitoring handle this outage without turning a recoverable dependency failure into a restart storm?

Accepted Answer

Use /ready for readiness and /live for liveness; keep the process running while it reconnects.

Answer

Use /ready for both probes; restart the process whenever the database connection is unavailable.

Answer

Use /live for both probes; keep routing transactions while the process attempts to reconnect.

Answer

Use /ready for readiness; scale the Deployment to zero until an operator confirms database recovery.

Question 4

A Tekton Pipeline's clone task writes a repository to $(workspaces.source.path). The test task declares runAfter: [clone] and reads that same relative directory. Both tasks map their source workspace to the Pipeline workspace checkout, which the PipelineRun binds to emptyDir: {}. Clone succeeds, but test reports that the source directory is empty. A suitable, writable PVC is available. What change addresses the missing files while preserving the task separation?

Accepted Answer

Bind checkout to the same PVC for both tasks, retaining their dependency.

Answer

Add retries to the test task so the checkout has more time to become visible.

Answer

Schedule both TaskRun Pods on one node while retaining the emptyDir binding.

Answer

Give both tasks the same container image while retaining the emptyDir binding.

Question 5

An application image works locally as UID 1001 but fails on OpenShift when creating /opt/app/cache/index. OpenShift assigns an arbitrary non-root UID, and the process belongs to group 0. The image directory is owned by 1001:1001 with mode 0750; no volume covers it, and the image filesystem is writable. Security policy must remain unchanged, and users outside the owning group must not gain access. The repair step can run as root during image construction. Select the appropriate build-time permission change.

Accepted Answer

RUN chgrp -R 0 /opt/app/cache && chmod -R g=u /opt/app/cache

Answer

RUN chown -R 1001:1001 /opt/app/cache && chmod -R 0770 /opt/app/cache

Answer

RUN chgrp -R 0 /opt/app/cache && chmod -R 0750 /opt/app/cache

Answer

RUN chmod -R 0777 /opt/app/cache

Question 6

A Helm chart renders replicas from .Values.replicaCount and the container tag from .Values.image.tag; neither expression refers to .Chart.AppVersion. The chart's values.yaml specifies replicaCount: 1 and image.tag: "2.0". The file prod.yaml specifies replicaCount: 3 and image.tag: "2.4". A developer changes Chart.yaml's appVersion to "3.0" and runs:

helm template storefront ./storefront -f prod.yaml --set replicaCount=5

The rendered Deployment will request:

Accepted Answer

5 replicas using image tag 2.4.

Answer

3 replicas using image tag 2.4.

Answer

5 replicas using image tag 3.0.

Answer

1 replica using image tag 2.0.

Question 7

Pods in project storefront use ServiceAccount web and pull an image from the integrated registry's shared-images/catalog repository. The image exists, registry authentication works, and the pull is denied because web lacks access to the source project. Only this ServiceAccount should receive pull access. Which authorization should an administrator add?

Accepted Answer

Bind system:image-puller to system:serviceaccount:storefront:web in project shared-images.

Answer

Bind system:image-puller to system:serviceaccount:storefront:web in project storefront.

Answer

Bind system:image-puller to system:serviceaccount:shared-images:web in project shared-images.

Answer

Bind system:image-puller to the system:serviceaccounts:storefront group in project shared-images.

Question 8

A repository overrides an S2I builder's assemble script. The original assembler is /usr/libexec/s2i/assemble. With no shell errexit option enabled, the override contains:

#!/bin/sh
/usr/libexec/s2i/assemble
printf '%s
' 'Assembly finished'

The original assembler exits 2, but the override reports success because printf succeeds. A supplied /opt/check-artifacts program must run only after successful assembly, and either program's failure must fail the override. Which replacement body preserves that contract?

Accepted Answer

/usr/libexec/s2i/assemble && /opt/check-artifacts

Answer

/usr/libexec/s2i/assemble; /opt/check-artifacts

Answer

/usr/libexec/s2i/assemble || /opt/check-artifacts

Answer

/usr/libexec/s2i/assemble && /opt/check-artifacts; exit 0

Question 9

Before approving a rollout, a developer checks a Deployment with 7 desired replicas, maxSurge: 25%, and maxUnavailable: 25%. All existing replicas are available. The application must keep at least 6 replicas available, and the rollout policy must cap this Deployment at 8 non-terminating Pods. No other failures occur, and terminating Pods are excluded from the capacity check. Which assessment and adjustment meet both requirements?

Accepted Answer

The settings allow 9 Pods and 6 available replicas; set maxSurge to the integer 1.

Answer

The settings allow 8 Pods and 6 available replicas; retain both percentages.

Answer

The settings allow 9 Pods and 5 available replicas; set maxUnavailable to the integer 1.

Answer

The settings allow 9 Pods and 6 available replicas; set maxUnavailable to the integer 0.

Question 10

An OpenShift Docker-strategy build uses a successful postCommit hook to write /opt/app/release.txt and then run a verification program. The logs confirm that the file exists during verification. The build pushes successfully, yet a container started from the pushed digest has no release.txt. The file must be part of the published image, and verification must still block publication on failure. Where should the two operations run?

Accepted Answer

Create the file in a Dockerfile RUN instruction and retain verification in postCommit.

Answer

Create the file in a Dockerfile RUN instruction and move verification into a liveness probe.

Answer

Keep both operations in postCommit and use imagePullPolicy: Always for application containers.

Answer

Create the file in the application's entrypoint and retain verification in the postCommit hook.

Free EX288 Practice Questions

Domain 1: Work with Red Hat OpenShift Container Platform

Question 1

Question 2

Domain 2: Deploy multi-container applications

Question 3

Domain 3: Work with container images in OpenShift Container Platform

Question 4

Domain 4: Troubleshoot application builds and deployment issues

Question 5

Domain 5: Work with image streams

Question 6

Domain 6: Work with configuration maps and secrets

Question 7

Domain 7: Work with the source-to-image (S2I) framework

Question 8

Domain 8: Work with build hooks and build triggers

Question 9

Domain 10: Work with OpenShift Pipelines

Question 10

The rest of the EX288 blueprint

That's 10 of 1,030