Practice for the Red Hat Certified Specialist in OpenShift AI (EX267) exam with 18 exam-style practice questions, instant answer reveals, and concise explanations of every correct answer. Topics include: What is the status condition of an InferenceService when the model file fails to download due to invalid S3 credentials?. 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
- What is the status condition of an InferenceService when the model file fails to download due to invalid S3 c…
- Which controller is responsible for monitoring and maintaining the health of the ModelMesh pods?
- Which configuration component represents an external storage repository configuration such as AWS S3 within O…
- When configuring a Retrieval-Augmented Generation (RAG) pipeline, what is the role of the vector database com…
- What custom resource does an engineer deploy to initiate a distributed compute cluster for model training via…
- What tool integrated within OpenShift AI allows data scientists to build visual pipeline workflows directly i…
Answers and explanations
Tap a question to expand the answer and the exam reasoning. Try to commit to your own pick first.
Q1. What is the status condition of an InferenceService when the model file fails to download due to invalid S3 credentials?
Answer: D. Ready: False, with a status message detailing storage download failure
An InferenceService reports a Ready: False state with an error describing the storage download failure when S3 credentials are invalid. Always check the status conditions to debug storage access issues during deployments.
Q2. Which controller is responsible for monitoring and maintaining the health of the ModelMesh pods?
Answer: C. modelmesh-controller pod running within the application namespace
The modelmesh-controller pod monitors and maintains the health of ModelMesh pods. Standard OpenShift controllers like the scheduler do not manage the internal lifecycle of these specialized AI runtimes.
Q3. Which configuration component represents an external storage repository configuration such as AWS S3 within OpenShift AI?
Answer: B. Data Connection
A Data Connection represents external storage repository configurations like AWS S3 within OpenShift AI. It securely stores endpoint and credential information as a Kubernetes Secret for workbenches and pipelines.
Q4. When configuring a Retrieval-Augmented Generation (RAG) pipeline, what is the role of the vector database component?
Answer: C. To store high-dimensional document embeddings for semantic search retrieval queries
A vector database stores high-dimensional document embeddings for semantic search retrieval queries. It does not compile code, balance traffic, or cache weights, but allows the language model to fetch relevant context based on mathematical vector similarity.
Q5. What custom resource does an engineer deploy to initiate a distributed compute cluster for model training via Ray?
Answer: C. RayCluster
The RayCluster custom resource provisions the head and worker pods required for distributed training. RayJob is for running a specific task to completion, whereas RayCluster sets up the persistent compute pool itself.
Q6. What tool integrated within OpenShift AI allows data scientists to build visual pipeline workflows directly inside JupyterLab?
Answer: B. Elyra Pipeline Editor
The Elyra Pipeline Editor extension enables data scientists to build visual pipelines natively within JupyterLab. OpenShift Console Topology displays deployed resources but does not author workflows inside the notebook interface.
Q7. When configuring a custom notebook image, what label format helps OpenShift AI identify it as an available workbench image?
Answer: D. opendatahub.io/notebook-image: 'true'
The opendatahub.io/notebook-image equals true label flags the image for the workbench deployment picker. Standard Kubernetes labels exist but lack the OpenShift AI specific metadata needed for discovery.
Q8. Where should a data scientist place private repository access credentials to keep them hidden from shared version control repositories?
Answer: D. In a secure Kubernetes Secret linked to the OpenShift project
Kubernetes Secrets securely isolate sensitive repository credentials away from version control. Committing keys to Git or leaving them in plain text inside a workbench risks unauthorized access.
Q9. Which visual layout view in the OpenShift console provides an interactive graphical representation of data science project assets?
Answer: D. Topology view
The OpenShift Console Topology view renders a graphical layout of project assets and their connections. Administrator YAML editors or monitoring dashboards do not offer this interactive visual mapping.
Q10. Which environment variable is automatically populated to help applications inside an OpenShift AI workbench locate the cluster's internal storage?
Answer: B. STORAGE_URI
The STORAGE_URI environment variable is injected to expose the internal cluster storage path. PVC_MOUNT_PATH indicates volume attachment, but STORAGE_URI specifically defines the endpoint location.
Q11. Which container image configuration file is modified when adding custom packages to a workbench environment permanently?
Answer: D. Containerfile or Dockerfile
A Containerfile or Dockerfile is modified to build a custom workbench image with permanent packages. Runtime configurations like bashrc scripts are lost when the pod restarts, so baking dependencies into the image ensures persistence.
Q12. Where do OpenShift AI pipelines store intermediate artifact records, execution metrics, and logs?
Answer: C. External S3-compatible object storage
OpenShift AI pipelines utilize external S3-compatible object storage to store artifacts, metrics, and logs. This ensures data persistence across pipeline runs, unlike ephemeral pod directories or cluster persistent volumes.
Q13. Which CRD configuration controls the enablement or disablement of individual OpenShift AI components like pipelines or dashboards?
Answer: C. DataScienceCluster
The DataScienceCluster custom resource definition manages component enablement across the cluster. Administrators toggle components like the dashboard or pipelines here, as standard logging or service mesh resources do not control them.
Q14. What specification in the workbench definition determines the maximum CPU and Memory resources a notebook pod can consume?
Answer: C. Resource Requests and Limits
Resource Requests and Limits within the workbench specification determine the maximum CPU and memory a pod can consume. Node selectors and storage classes control pod placement and disk allocation, not compute resources.
Q15. What happens if you delete a project/namespace that contains active OpenShift AI workbenches and data connections?
Answer: B. All associated workbenches, data connections, PVCs, and configurations within that namespace are deleted
Deleting a project namespace permanently removes all associated resources, including workbenches, data connections, and PVCs. OpenShift does not preserve state globally, so ensure any critical data is backed up before deletion.
Q16. What is the primary web console used to manage Red Hat OpenShift AI component lifecycles?
Answer: D. OpenShift Web Console
The OpenShift Web Console is the primary interface for managing the platform and its components. While OpenShift AI has its own dashboards, the core administration and operator lifecycle management still happen in the web console.
Q17. When configuring multi-model serving, what underlying architecture optimizes resource cost by packing models onto shared pods?
Answer: D. ModelMesh pods
ModelMesh is designed for multi-model serving and optimizes resources by packing multiple models into shared pods. Single-model runtimes, by contrast, deploy dedicated pods for each model, which is less efficient for large fleets.
Q18. Which CLI command allows you to view the current status of all pods in the Red Hat OpenShift AI applications namespace?
Answer: B. oc get pods -n redhat-ods-applications
The command uses the oc CLI to query the redhat-ods-applications namespace. Remember that core OpenShift AI components typically reside in this specific namespace during the exam.
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.