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: How can an administrator restrict OpenShift AI workbench creation to only a specific group of LDAP/OAuth users?. 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 an administrator restrict OpenShift AI workbench creation to only a specific group of LDAP/OAuth user…
- What component provides the storage management interface for pipelines in OpenShift AI?
- Which custom resource represents an end-to-end machine learning pipeline definition in OpenShift AI?
- Which parameter handles memory requests for a workbench instance to ensure it receives enough RAM upon startu…
- What is the maximum number of model server instances that can be run per namespace under standard ModelMesh c…
- What is the default deployment mode for Red Hat OpenShift AI on self-managed OpenShift clusters?
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 an administrator restrict OpenShift AI workbench creation to only a specific group of LDAP/OAuth users?
Answer: D. Configuring access groups within the OpenShift AI dashboard settings or cluster groups
Administrators restrict workbench creation by configuring specific dashboard user groups or OpenShift RBAC roles. Modifying network policies or deleting custom resources will not properly manage dashboard access permissions.
Q2. What component provides the storage management interface for pipelines in OpenShift AI?
Answer: A. MariaDB/MySQL database and S3-compatible storage managed by the pipeline operators
Data Science Pipelines rely on a database like MariaDB or MySQL for metadata and S3-compatible storage for artifacts. This combination ensures pipeline state and output persistence across operator-managed environments.
Q3. Which custom resource represents an end-to-end machine learning pipeline definition in OpenShift AI?
Answer: C. DataSciencePipeline
The DataSciencePipeline custom resource represents pipeline definitions in OpenShift AI. Note that PipelineRun represents execution instances, so a reasonable candidate might select it when focusing on the active runtime state rather than the core definition.
Q4. Which parameter handles memory requests for a workbench instance to ensure it receives enough RAM upon startup?
Answer: B. resources.requests.memory
The resources.requests.memory parameter guarantees the requested RAM for a container upon startup. While resources.limits.memory defines the absolute maximum memory allowed, the requests field is what Kubernetes uses for scheduling and initial allocation.
Q5. What is the maximum number of model server instances that can be run per namespace under standard ModelMesh constraints?
Answer: C. Determined by resource availability and configuration limits
ModelMesh does not have a strict numerical limit on model server instances. The actual capacity is determined entirely by the available cluster resources and your specific configuration limits, making it highly scalable within the namespace.
Q6. What is the default deployment mode for Red Hat OpenShift AI on self-managed OpenShift clusters?
Answer: D. Connected Deployment
Connected deployment is the standard default for self-managed clusters, relying on direct internet access to pull container images. Disconnected deployment is the strongest distractor, but it requires manual configuration for restricted network environments.
Q7. Which component manages user session tokens inside the Jupyter notebooks UI?
Answer: D. JupyterHub / Kubeflow Notebook Controller
The Kubeflow Notebook Controller correctly manages user session tokens and notebook lifecycle states within the dashboard. OpenShift Network Operator fails as a distractor because it strictly handles cluster networking, not identity or session management.
Q8. Which component in OpenShift AI allows data scientists to create isolated environments for data science workflows?
Answer: D. Projects
Data Science Projects provide isolated environments for workflows by mapping directly to OpenShift namespaces. Networks and channels do not provide the required access boundaries for users and administrators.
Q9. What is the format of the output artifact from a data science pipeline step that feeds into a subsequent step?
Answer: A. Managed file paths or URI pointers stored in object storage and passed via pipeline context
Pipelines pass data between steps using object storage paths. This ensures scalability across different compute nodes. Relying on stdout logs or local memory addresses prevents steps from executing reliably.
Q10. What container engine executes individual tasks inside an OpenShift Container Platform cluster node?
Answer: A. CRI-O
CRI-O is the default container engine used by OpenShift Container Platform to run pod containers on cluster nodes. While Docker and containerd are popular elsewhere, OpenShift specifically relies on CRI-O for its Kubernetes implementation.
Q11. What role must a user possess within an OpenShift project to deploy a machine learning model to a model server?
Answer: B. Admin or Edit role within that specific project namespace
Deploying a model requires write permissions within the specific Data Science Project namespace. The Admin or Edit role provides the necessary permissions to create and manage model server resources, whereas the View role only permits read operations.
Q12. Which object storage parameter specifies the exact storage division name when creating an S3 data connection?
Answer: B. AWS_STORAGE_BUCKET
The bucket parameter specifies the target S3 storage division for a data connection. While the endpoint defines the server address and region handles location routing, neither identifies the specific bucket name needed for access.
Q13. Which capability allows OpenShift AI to dynamically scale notebook pods down to zero when they are not in use?
Answer: B. Idling/Culling feature
The idling or culling feature dynamically scales notebook pods down to zero when inactive. Standard cluster autoscalers adjust hardware nodes, while culling specifically targets the workbench pods to free up cluster resources during idle periods.
Q14. What type of OpenShift object maps directly to an active Data Science Workbench backend deployment?
Answer: B. Deployment
Data Science Workbenches are backed by standard Kubernetes Deployment objects that manage the pod replicas. However, StatefulSet is also a highly defensible answer because workbenches use persistent volumes for state, introducing ambiguity to the wording.
Q15. Which environment variable dictates the default region parameter when binding an external S3 cloud bucket connection?
Answer: B. AWS_DEFAULT_REGION
The AWS_DEFAULT_REGION environment variable dictates the default region parameter when binding external S3 bucket connections. The other options are fabricated variables, making standard AWS environment variables your clear exam cue for connections.
Q16. Which custom resource coordinates dependencies, initialization steps, and cluster prerequisites before OpenShift AI initializes?
Answer: B. DSCInitialization
The DSCInitialization custom resource handles dependencies and cluster prerequisites before components deploy. Standard OpenShift objects like an OperatorGroup or CSV manage the operator lifecycle rather than OpenShift AI specific initialization.
Q17. What parameter determines the maximum execution runtime limits for individual pipeline steps inside OpenShift AI?
Answer: D. Pipeline timeout parameters within the Tekton configuration
Pipeline timeout parameters within Tekton configuration determine maximum execution runtimes for individual pipeline steps. OpenShift AI leverages Tekton under the hood, meaning step limits come from Tekton properties.
Q18. Which feature inside OpenShift AI pipelines enables the definition of conditional execution paths based on previous step outputs?
Answer: D. Elyra pipeline branch routing logic
The Elyra pipeline editor allows visual definition of conditional execution branches. Static build manifests or bucket triggers handle delivery, not logical routing based on the outputs of previous pipeline steps.
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.