Practice for the Red Hat Certified Specialist in OpenShift AI (EX267) exam with 17 exam-style practice questions, instant answer reveals, and concise explanations of every correct answer. Topics include: How can you ensure that an updated custom notebook image becomes available to users if the image tag hasn't changed?. Follow @CertPunch and visit certpunch.com for more certification practice exams and study content.
Prefer hands-on? Take this round as an interactive practice test — answer every question, get instant feedback, and see your score: Start the Red Hat Certified Specialist in OpenShift AI (EX267) practice test →
What you will practice
- How can you ensure that an updated custom notebook image becomes available to users if the image tag hasn't c…
- Which setting determines the specific packages and tools available inside an OpenShift AI workbench?
- What happens if multiple workbenches inside the same Data Science Project attempt to use the exact same Persi…
- When configuring a custom notebook image in OpenShift AI, which configuration resource allows cluster admins…
- Which validation pattern checks that your OpenShift AI cluster environment is properly configured to use oper…
- Which component in OpenShift AI provides the default user interface for launching Jupyter Notebooks and manag…
Answers and explanations
Tap a question to expand the answer and the exam reasoning. Try to commit to your own pick first.
Q1. How can you ensure that an updated custom notebook image becomes available to users if the image tag hasn't changed?
Answer: D. Trigger an image import or change the image pull policy to Always
Triggering an image import or changing the pull policy to Always forces the cluster to fetch the latest image changes. On the exam, if your custom notebook updates do not appear, always verify your image pull policy before attempting a restart.
Q2. Which setting determines the specific packages and tools available inside an OpenShift AI workbench?
Answer: B. The selected notebook container image
The selected notebook container image defines the underlying packages, libraries, and tools available. Remember that environment variables only configure existing software, while the base image dictates what is actually installed.
Q3. What happens if multiple workbenches inside the same Data Science Project attempt to use the exact same Persistent Volume Claim concurrently if the storage class is RWO?
Answer: C. The first workbench starts successfully, while the second fails to bind the volume
A ReadWriteOnce volume only allows one node to mount it concurrently, so the second workbench fails to bind. Expect this behavior on the exam when sharing storage across workbenches on different cluster nodes.
Q4. When configuring a custom notebook image in OpenShift AI, which configuration resource allows cluster admins to make it visible in the Jupyter spawner?
Answer: C. ImageStream
Cluster admins make custom notebook images visible in the Jupyter spawner using an OpenShift ImageStream tagged with specific labels. While other resources manage deployments or configuration, the spawner specifically queries ImageStreams.
Q5. Which validation pattern checks that your OpenShift AI cluster environment is properly configured to use operational graphics accelerators?
Answer: A. Verifying the presence of nvidia.com resources under node descriptions
Validating GPU configuration involves checking node descriptions for extended resources like nvidia.com. During the exam, routinely inspect the node capacity using standard command-line tools to confirm the Node Feature Discovery operator.
Q6. Which component in OpenShift AI provides the default user interface for launching Jupyter Notebooks and managing projects?
Answer: A. OpenShift AI Dashboard
The OpenShift AI Dashboard provides the primary user interface for managing projects and launching Jupyter Notebooks. Pipelines and GitOps handle deployment automation, while Advanced Cluster Management operates at a multi-cluster level.
Q7. What is the primary role of the odh-dashboard pod in an active Red Hat OpenShift AI environment?
Answer: C. To serve the graphical user interface to connected web clients
The dashboard pod serves the web-based graphical user interface for OpenShift AI. Exam tip: know that user-facing interactions happen here, not in machine learning processing or low-level network encryption pods.
Q8. What security context parameter ensures an OpenShift AI workbench does not run with root permissions?
Answer: A. runAsNonRoot: true
Setting runAsNonRoot to true forces the container to execute without root privileges. Remember that runAsUser zero explicitly assigns root, while privileged true grants full host access.
Q9. Which resource type represents an instance of an executed pipeline workflow in OpenShift AI?
Answer: C. PipelineRun
A PipelineRun represents an actual execution of a pipeline workflow. PipelineServer merely defines the execution environment, and DeploymentConfig handles generic application deployments, not pipeline runs.
Q10. When configuring a custom ServingRuntime, what placeholder variable is used by the system to dynamically inject the model file storage path?
Answer: B. ${STORAGE_URI}
The STORAGE_URI placeholder is used in a custom ServingRuntime template to dynamically inject the model storage path. Avoid confusing this with generic container variables; the exam specifically tests storage path placeholders for model deployment configurations.
Q11. To inject object storage credentials (like AWS S3) directly into an OpenShift AI Data Science Project, what feature should you configure in the dashboard?
Answer: C. Data Connection
A Data Connection is used to inject object storage credentials directly into a Data Science Project or workbench. Do not confuse this with a StorageClass, which provisions persistent volumes rather than linking external S3 credentials.
Q12. What mechanism allows an OpenShift AI notebook to interact directly with cluster resources securely?
Answer: D. Associating a specific ServiceAccount with the notebook pod
Associating a specific ServiceAccount with the notebook pod allows it to interact securely with cluster resources via RBAC. Hardcoding cluster-admin credentials or mounting host sockets violates basic OpenShift security practices.
Q13. Which component is responsible for orchestrating the execution of scheduled tasks in an OpenShift AI pipeline?
Answer: B. Tekton Pipeline Controller pods running in the application namespace
OpenShift AI data science pipelines are powered by Tekton, meaning the Tekton Pipeline Controller orchestrates the execution of pipeline tasks within the target namespace. Do not confuse pipeline execution with the Jupyter interface; the controller actively manages the pod lifecycle.
Q14. What tool acts as the artifact repository controller within the OpenShift AI pipeline architecture?
Answer: C. MinIO or an equivalent S3 provider
Pipelines require an S3-compatible object storage backend to store intermediate artifacts and track runs, which is typically MinIO in a default OpenShift AI installation. Quay is for container images, while Nexus handles generic build dependencies.
Q15. What happens if you delete a Data Science Project within the OpenShift AI dashboard interface?
Answer: D. The corresponding OpenShift namespace and all its encapsulated resources (workbenches, PVCs, servers) are removed
Deleting a Data Science Project directly removes the backing OpenShift namespace and all associated resources, including workbenches and persistent storage. Be exceptionally careful during the exam, as this destructive action cannot be undone.
Q16. What mechanism handles user authentication and role management inside OpenShift AI?
Answer: B. OpenShift OAuth
OpenShift AI integrates natively with the underlying cluster authentication, relying directly on OpenShift OAuth for user identity and role mapping. Standalone identity providers like Keycloak are only configured at the platform level.
Q17. What parameter within an AcceleratorProfile resource determines which specific runtime nodes can run the GPU workloads?
Answer: A. nodeSelector labels matching the hardware attributes of the nodes
An AcceleratorProfile uses a node selector to target specific cluster nodes equipped with the necessary hardware accelerators like GPUs. This scheduling mechanism ensures compute workloads physically land on the correct machines.
More Red Hat Certified Specialist in OpenShift AI (EX267) drills and other practice exams are on @CertPunch. New rounds drop every few days at certpunch.com.