I B M   C E R T I F I C A T I O N

C9004400 IBM Certified Administrator – IBM Cloud Pak for Business Automation v21.0.3 Practice Exam

Exam Number: 4322 | Last updated April 17, 2026 | 438+ questions across 5 vendor-aligned objectives

Cluster admins who run Cloud Pak for Business Automation v21.0.3 on Red Hat OpenShift are the target audience for this exam. The credential validates daily administration tasks across Business Automation Workflow, Automation Decision Services, Content Services, Operational Decision Manager, and the Automation Navigator. Candidates should be comfortable with operator-based install patterns, cluster sizing, and the user- and role-management structures specific to business-automation workloads.

Accounting for 26% of the exam, Installation and Upgrade covers operator subscriptions, custom resource design, cluster sizing, and upgrade paths between v21 fix packs. User and Access Administration takes 22%, covering LDAP integration, service accounts, role assignments, and workspace tenancy. A further 20% targets Workload Configuration, covering application server tuning, database sizing, and container resource limits for each capability.

Completing the blueprint, Backup and Restore accounts for 18% and spans persistent-volume backups, application-level exports, and multi-site restore procedures. Monitoring and Troubleshooting represents 14% and spans log aggregation, metric endpoints, and support-bundle generation. Many scenario questions are layered — a single problem may require you to reason about LDAP, OpenShift operators, and capability-specific configuration together.

 Operator patterns differ enough across Cloud Pak capabilities that memorizing one does not translate to the next — study the Automation Decision Services and Content Services operators side by side so the differences stick. Database sizing questions are more quantitative than candidates expect; be ready to estimate storage and IOPS given a process volume.

Every answer links to the source. Each explanation below includes a hyperlink to the exact IBM documentation page the question was derived from. PowerKram is the only practice platform with source-verified explanations. Learn about our methodology →

749

practice exam users

94%

satisfied users

91%

passed the exam

4.5/5

quality rating

Test your C9004400 cloudpak business v21 admin knowledge

10 of 438+ questions

Question #1 - Installation and Upgrade

An admin at Ashford Freight is preparing to install Cloud Pak for Business Automation v21.0.3 on OpenShift for the first time.

Which installation pattern matches v21.0.3 guidance?

A) Download raw Helm charts and apply them by hand
B) Subscribe the IBM Cloud Pak Foundational Services and CP4BA operators through OperatorHub, then create the custom resource for the automation capabilities required
C) Install each capability product as a standalone container image
D) Skip operators and install each capability directly from source

 

Correct answers: B – Explanation:
Operator-based install through OperatorHub plus a CR for the automation capabilities is the documented v21.0.3 pattern. Raw Helm bypasses lifecycle management. Standalone images and source installs forfeit operator-provided upgrade paths. Source: Check Source

A CP4BA v21.0.3 upgrade at Pelham Insurance must move from the v21.0.2 fix pack to v21.0.3 without data loss.

Which upgrade practice is appropriate?

A) Upgrade only the web UI container and leave others at v21.0.2
B) Delete the existing deployment and reinstall from scratch
C) Skip backups because the upgrade is minor
D) Follow the documented fix-pack upgrade path, take backups first, and upgrade the operator subscriptions which drive in-place capability upgrades

 

Correct answers: D – Explanation:
Backup first, upgrade subscriptions along the documented path is the CP4BA v21.0.3 upgrade reference. Delete-and-reinstall loses data. Skipping backups is reckless. Partial upgrades violate capability version coherence. Source: Check Source

A cluster-sizing review at Holborn Pharma reveals that CP4BA Business Automation Workflow is under-provisioned at peak.

Which admin adjustment addresses the symptom at the right layer?

A) Edit the underlying Deployment objects directly
B) Adjust the replica count and resource requests/limits in the CP4BA custom resource for the BAW capability, then let the operator reconcile the deployment
C) Add an unrelated third-party autoscaler that ignores the operator
D) Move the pods to a different namespace manually

 

Correct answers: B – Explanation:
Changing the CR and letting the operator reconcile is the CP4BA-supported path. Direct Deployment edits drift and get reverted. Third-party autoscalers without operator awareness fight the controller. Manual namespace moves break the operator state. Source: Check Source

A CP4BA admin at Lindenhall Mutual must integrate the cluster with the company’s LDAP directory so existing groups grant capability access.

Which configuration satisfies the requirement?

A) Share a single admin service account
B) Create every user as a local account and re-key passwords monthly
C) Configure LDAP on the Cloud Pak foundational services identity provider and map LDAP groups to CP4BA roles for each capability
D) Disable authentication for capability UIs

 

Correct answers: C – Explanation:
IAM-fed LDAP plus group-to-role mapping is the CP4BA access-admin reference. Local accounts break lifecycle. Shared service accounts break attribution. Disabled auth is a control failure. Source: Check Source

A workspace-tenancy question at Easterbrook Bank asks how to keep two business units’ BAW workflows isolated on a shared CP4BA deployment.

Which feature implements the isolation?

A) Create separate BAW environments or tenants with per-tenant role assignments and workspace boundaries, mapped to distinct LDAP groups
B) Put both BUs in the same tenant and trust them to self-police
C) Run one BU’s workloads on developer laptops
D) Remove all role checks

 

Correct answers: A – Explanation:
Per-tenant environments with scoped role assignments is CP4BA’s workspace-tenancy answer. Shared tenants break isolation. Laptop deployment is not a platform. Removing role checks removes tenancy. Source: Check Source

An Operational Decision Manager capability at Trentham Retail exhausts its database connections during month-end processing.

Which admin tuning best addresses the symptom?

A) Restart the pods on every DB error
B) Size the ODM database connection pool to match the observed peak, and tune the container resource requests in the CP4BA custom resource to match the load pattern
C) Reduce the database size to shrink the problem
D) Disable month-end rules

 

Correct answers: B – Explanation:
Connection-pool sizing plus CR-driven container tuning is the CP4BA workload-configuration approach. Restart loops hide the issue. Shrinking the DB is irrelevant. Disabling rules breaks the business need. Source: Check Source

A Content Services capability at Brindleford Energy fills its persistent volume faster than forecast.

Which admin action fits the CP4BA configuration model?

A) Move content to a developer’s laptop
B) Delete content at random to reclaim space
C) Expand the persistent volume size in the CP4BA CR for Content Services, ensuring the OpenShift storage class supports online expansion
D) Turn off Content Services

 

Correct answers: C – Explanation:
PVC expansion through the CR with a supporting storage class is the CP4BA pattern. Deleting content risks data loss. Laptop storage is not a platform. Turning off the capability removes the service. Source: Check Source

An auditor at Morningside Credit asks the CP4BA admin how the platform would recover from loss of a persistent volume holding process data.

Which recovery design fits the CP4BA model?

A) Schedule persistent-volume and application-level exports per the CP4BA backup guide, with test restores on a non-production cluster to verify RTO and RPO targets
B) Rely on nightly OpenShift node snapshots only
C) Assume another team handles backups
D) Never back up because the data can be re-entered

 

Correct answers: A – Explanation:
PV plus application-level exports with tested restores is CP4BA’s documented backup pattern. Node snapshots miss application-consistent state. Assumptions fail audit. No backups is no recovery. Source: Check Source

A CP4BA admin at Rosevale Insurance is asked to restore a specific workflow instance from a backup without rolling back the entire capability.

Which CP4BA restore practice fits the ask?

A) Delete all workflows and start again
B) Full-restore the entire PVC, losing other changes
C) Recreate the workflow manually from memory
D) Use application-level export/import tools for BAW, rather than a full volume restore, to bring a specific workflow state into the running deployment

 

Correct answers: D – Explanation:
Application-level export/import for targeted restores is CP4BA’s recommended pattern. Full PVC restores overwrite unrelated state. Memory-based recreation is not restore. Deleting and restarting is worse, not better. Source: Check Source

Support at Oldcastle Holdings asks the CP4BA admin for a single package containing logs and configuration to debug a workflow issue.

Which CP4BA capability produces that package?

A) The CP4BA support-bundle generation tool, which collects operator logs, capability logs, and configuration into a single archive for support
B) A screenshot of the terminal
C) A verbal summary over a phone call
D) The admin’s personal notes

 

Correct answers: A – Explanation:
Support bundles from the CP4BA tooling are IBM’s answer for this request. Screenshots, verbal summaries, and notes do not collect the needed depth. Source: Check Source

Get 438+ more questions with source-linked explanations

Every answer traces to the exact IBM documentation page — so you learn from the source, not just memorize answers.

Exam mode & learn mode · Score by objective · Updated April 17, 2026

Learn more...

What the C9004400 cloudpak business v21 admin exam measures

  • Install and upgrade operator subscriptions, custom resources, cluster sizing, and fix-pack paths to deliver reliable automation platforms that scale with business demand
  • Administer and delegate LDAP integration, service accounts, role assignments, and tenant workspaces to enforce least-privilege access across diverse business-automation teams
  • Configure and size application servers, databases, and container resource limits to keep response times within SLA for interactive and batch workloads alike
  • Back up and restore persistent volumes, application-level exports, and multi-site recovery plans to keep automation estates recoverable through planned upgrades and unplanned outages
  • Observe and diagnose log aggregation, metric endpoints, and support-bundle generation to shorten outages and give vendor support the evidence they need to help you quickly

  • Review the official exam guide to understand every objective and domain weight before you begin studying
  • Work through the relevant IBM Training learning path — ibm certified administrator ibm cloud pak for business automation v21 0 3 C9004400 — to cover vendor-authored material end-to-end
  • Get hands-on inside IBM TechZone or a comparable sandbox so you can practice the console tasks, CLI commands, and APIs the exam expects
  • Tackle a real-world project at your workplace, a volunteer role, or an open-source repository where the technology under test is actually in use
  • Drill one exam objective at a time, starting with the highest-weighted domain and only moving on once you can teach it to someone else
  • Study by objective in PowerKram learn mode, where every explanation links back to authoritative IBM documentation
  • Switch to PowerKram exam mode to rehearse under timed conditions and confirm you consistently score above the pass mark

Cloud Pak administrators with OpenShift skills are increasingly valued as business-automation platforms scale:

  • Cloud Pak Administrator — $100,000–$140,000 per year, operating IBM Cloud Pak deployments across OpenShift (Glassdoor salary data)
  • Business Automation Engineer — $110,000–$150,000 per year, supporting process and decision automation platforms (Indeed salary data)
  • OpenShift Platform Engineer — $115,000–$155,000 per year, running OpenShift clusters that host business-critical apps (Glassdoor salary data)

Work through the official IBM Training learning path for this certification, which bundles videos, labs, and skill tasks aligned to every objective. The official exam page lists the full objective breakdown, prerequisite knowledge, and scheduling details.

Related certifications to explore

Related reading from our Learning Hub