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: What component handles scheduled, recurring executions of data science pipelines in OpenShift AI?. 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 component handles scheduled, recurring executions of data science pipelines in OpenShift AI?
- What security context parameter is enforced by OpenShift on workbench pods for safety?
- Which component is responsible for orchestrating the build and lifecycle steps of pipeline workflows?
- Which storage technology ensures user data scales fluidly across nodes without tying workloads to single avai…
- What happens to active pipelines when the central OpenShift AI Operator undergoes a minor version patch upgra…
- What kind of OpenShift resource is automatically created when you launch a new Workbench?
Answers and explanations
Tap a question to expand the answer and the exam reasoning. Try to commit to your own pick first.
Q1. What component handles scheduled, recurring executions of data science pipelines in OpenShift AI?
Answer: D. Pipeline Scheduler/Trigger feature within Elyra or Kubeflow Pipelines backend
Recurring data science pipeline executions are handled by triggers configured through the pipelines backend. Avoid choosing OpenShift CronJobs, as the exam focuses on native pipeline scheduling via the dashboard.
Q2. What security context parameter is enforced by OpenShift on workbench pods for safety?
Answer: C. Restricted non-root user execution access privileges configuration
OpenShift enforces restricted security contexts, requiring workbench pods to run as non-root users. Knowing this is essential when building custom notebook images, as root-level privileges will be denied by default.
Q3. Which component is responsible for orchestrating the build and lifecycle steps of pipeline workflows?
Answer: D. Tekton pipelines controller deployment operator instance
OpenShift AI relies on the Tekton Pipelines controller to execute workflow tasks. Distractors about KServe or ModelMesh are wrong because they manage model inference, not pipeline orchestration. Look for keywords linking steps or runs to Tekton on the exam.
Q4. Which storage technology ensures user data scales fluidly across nodes without tying workloads to single availability zones?
Answer: A. Distributed Cloud Native Storage via ODF systems
Distributed storage like OpenShift Data Foundation allows data volumes to scale fluidly across nodes and availability zones. However, this question is somewhat ambiguous because generic cloud native storage could encompass other technologies, making it loosely worded.
Q5. What happens to active pipelines when the central OpenShift AI Operator undergoes a minor version patch upgrade sequence?
Answer: A. Running pipelines execute uninterrupted to completion
Pipelines operate within deployed pods and typically continue executing to completion during minor operator upgrades. This question is ambiguous because while running tasks finish, scheduling new steps might be delayed depending on the exact upgrade sequence.
Q6. What kind of OpenShift resource is automatically created when you launch a new Workbench?
Answer: A. StatefulSet
An OpenShift AI workbench creates a StatefulSet to maintain the notebook environment and persistent storage state. However, some background components might use Deployments, making the underlying orchestration slightly ambiguous.
Q7. What parameter determines the maximum processing duration allowed for an Elyra pipeline task before getting aborted?
Answer: A. Pipeline Step timeout configuration parameter
The Pipeline Step timeout configuration parameter sets the maximum duration for an Elyra pipeline task before it aborts. This prevents stuck tasks from running indefinitely, whereas LimitRange only restricts memory sizes.
Q8. What configuration element prevents normal application pods from landing on expensive dedicated GPU-enabled worker nodes?
Answer: C. Taints applied to GPU worker nodes
Taints are applied to nodes to repel pods that do not have a matching toleration. This prevents standard pods from scheduling on specialized GPU nodes, whereas node selectors merely restrict pods to specific nodes.
Q9. How are updates to the Red Hat OpenShift AI Operator delivered to the platform cluster?
Answer: C. Operator Lifecycle Manager via configured catalog sources channels
Updates for the OpenShift AI Operator are delivered through the Operator Lifecycle Manager using configured catalog sources. This standard mechanism ensures controlled updates via specified channels.
Q10. What parameters are specified in a secret configuration to connect to a secure private Git repository?
Answer: D. Git repository authentication credentials like tokens or SSH keys
To connect securely to a private Git repository, you must configure a Secret with Git credentials like tokens or SSH keys. Avoid mixing these with object storage or container registry authentication parameters.
Q11. What technology allows data scientists to write python scripts to generate pipelines instead of using visual canvas tools?
Answer: A. Kubeflow Pipelines SDK or Tekton Python tools
The Kubeflow Pipelines SDK allows data scientists to define pipeline steps using Python code. This provides a programmatic alternative to visual canvas tools for building Data Science Pipelines.
Q12. Which platform layer manages the lifecycle state transitions of notebooks and deployments within OpenShift AI clusters?
Answer: B. Red Hat OpenShift Operators platform
Red Hat OpenShift Operators manage the packaging, deployment, and lifecycle of OpenShift AI components like notebooks and models. Core Kubernetes controllers handle generic pods but lack the domain-specific reconciliation logic provided by the Operators.
Q13. Which configuration parameter defines the specific location of a machine learning model artifact within an S3 data connection metadata block?
Answer: D. AWS_S3_BUCKET and path properties
Defining the AWS_S3_BUCKET and path properties specifies exactly where the model artifact resides within the connected storage. The registry path maps to internal cataloging and does not point to the physical S3 storage location like the bucket properties.
Q14. What is the purpose of configuring a Data Connection within an OpenShift AI project?
Answer: A. To link a workbench directly to an object storage bucket
The correct answer is correct because a Data Connection creates a Kubernetes secret that securely links workbenches and pipelines to S3-compatible object storage. While generic secrets can hold Git tokens, the primary OpenShift AI Data Connection type is tailored for object storage buckets.
Q15. Where are custom environment variable key-value collections safely persisted across multiple workbench restarts?
Answer: C. ConfigMaps or Secrets bound to workloads
ConfigMaps and Secrets provide Kubernetes-native mechanisms to store non-confidential and sensitive data, which are then mounted as environment variables into the workbench pod. Storing them directly in a pod specification would not persist across restarts.
Q16. What dashboard view lets teams review logs of the central OpenShift AI operator to diagnose missing platform pages?
Answer: C. Installed Operators -> OpenShift AI -> Pod Logs
Administrators navigate to Installed Operators via the web console to locate the OpenShift AI operator and view its pod logs for troubleshooting. The Data Science Projects view does not expose backend operator logs needed to diagnose missing components.
Q17. Which container image format standard is supported for custom workbench images registered within Red Hat OpenShift AI?
Answer: B. OCI / Docker Image Spec v1
OpenShift AI supports standard OCI or Docker image specifications for custom workbench images. Singularity, Flatpak, and App Container formats are not recognized by Kubernetes for deploying standard pod workloads.
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.