- What "Passing Score" Means for EX436
- How a Hands-On Exam Is Graded
- The Reboot Rule: Where Most Points Quietly Disappear
- The Nine Objective Areas and How They Feed Your Result
- Results, Score Reports and Retake Mechanics
- A Scoring Strategy Built for Cluster Tasks
- Sequencing Your Prep Around the Objectives
- Fee, Delivery and Renewal Facts That Affect Your Attempt
- Frequently Asked Questions
- EX436 is one three-hour, hands-on section graded against objective criteria; do not expect a multiple-choice style score.
- Configurations must survive reboot, so a cluster that works only until restart earns nothing for that task.
- This article quotes no pass mark, because none is verified here; check Red Hat's current handbook and your exam confirmation.
- Results usually arrive within three U.S. business days per the exam page; the program guide allows up to five.
What "Passing Score" Means for EX436
Candidates searching for the passing score on the Red Hat Certified Specialist in High Availability Clustering exam (EX436) usually expect a single percentage to memorize. This article will not hand you one, and that is deliberate. The sources we reviewed on October 14, 2026 describe how the exam is delivered and graded, but they do not give a verifiable pass mark, so we leave out any cut score rather than repeat an unsourced number from a forum thread.
What you can rely on is the structure of the exam, and that structure matters more than any number would:
- It is one hands-on section lasting three hours, focused on administering the RHEL High Availability Add-On.
- You perform real administration tasks on live systems rather than selecting answers.
- Outcomes are graded against objective criteria, meaning the state of your systems is checked, not your explanation of what you meant to do.
- Configurations must survive reboot.
How a Hands-On Exam Is Graded
Outcome-based, not method-based
Red Hat's performance-based exams check results. If a task says a clustered service must be running, constrained to specific nodes, and recoverable after a failure, the grader looks at the resulting cluster state. It does not care whether you used pcs from the command line, edited a resource step by step, or worked from a cheat sheet you memorized. Any method that produces the correct, persistent end state is a valid method.
What "objective criteria" implies for you
- Partial work is fragile. A resource that is created but not enabled, or a fencing device that is defined but never tested, may fail the criteria for the whole task.
- Verification is part of the job. You cannot assume a configuration works; you confirm it with cluster status output, failover tests, and checks on the underlying storage and services.
- Documentation allowances are not specified on the exam page. Do not plan around what you may or may not have open. Verify the rules for your booked session and prepare so that you can work from the system's own built-in help, such as man pages and
pcsusage output.
What the score report does and does not tell you
The exam page describes total scores, while Red Hat's handbook also describes feedback by objective area. Neither provides task-by-task performance details. In practice, this means a failing result gives you direction (which objective areas were weak) but not a replay of which exact steps went wrong. That makes disciplined self-verification during the exam even more valuable, because you will not get a second look at your mistakes afterward.
The Reboot Rule: Where Most Points Quietly Disappear
The single most consequential grading fact for this exam is that configurations must survive reboot. For a high availability cluster, this has specific, testable meaning. Consider what has to be true after a node restarts:
- Cluster services are enabled so the node can rejoin the cluster without manual intervention.
- Cluster configuration (resources, constraints, fencing devices, properties) is stored in the cluster information base, not just applied transiently.
- Filesystems, logical volumes, and any shared storage definitions are brought up by the cluster or by persistent configuration, not by a command you typed once.
- Firewall rules and SELinux settings that allow cluster traffic and service access are permanent, not runtime-only.
Practicing this is cheap in a home lab and expensive to learn on exam day. Our High Availability Clustering study guide covers lab construction in more detail.
The Nine Objective Areas and How They Feed Your Result
The published outline for EX436 lists nine first-level objective entries. These are unweighted: Red Hat does not publish percentage weights for them, and nine is the count of lines we preserved from the outline, not an officially declared "domain count." Because there are no published weights, you should treat every area as capable of costing you the exam. For a fuller walkthrough, read the complete guide to the exam domains.
Domains 1 and 2: Configure a high-availability cluster; Configure cluster fencing
These are foundational. If the cluster itself is unstable or fencing is missing, nearly everything built on top of it becomes unreliable.
- Node membership, authentication between nodes, and cluster creation and startup.
- Fence devices defined correctly and actually capable of fencing a node.
- Understanding why fencing is not optional for shared-storage clusters.
Domains 3 and 4: Configure cluster logging and monitoring; Configure cluster monitoring
The outline lists logging/monitoring and monitoring as separate entries, so prepare for both.
- Locating and interpreting cluster log output when diagnosing failures.
- Configuring how cluster events and resource health are observed.
- Using status and history tooling to confirm what the cluster actually did.
Domains 5 and 6: Configure a clustered fail-over service; Configure cluster service behavior
This is where resource design lives, and where careless constraints cause the most failures.
- Resource and resource-group creation, ordering, and colocation.
- Location preferences, stickiness, and failure-handling behavior.
- Testing failover deliberately rather than assuming it works.
Domains 7, 8 and 9: Configure storage; Configure GFS2 filesystems; Grow a GFS2 filesystem
The storage block is three separate entries, with GFS2 growth listed on its own.
- Shared and clustered storage preparation that multiple nodes can use safely.
- Creating and mounting GFS2 with the correct locking and journal settings for the number of nodes.
- Extending a GFS2 filesystem while it is in service, including the underlying volume.
Because the exam is one integrated hands-on block, these areas are not isolated. A GFS2 task can depend on your fencing being correct, and a failover service task can depend on your storage being correct. A weakness in an early area can cascade, which is why your score is better thought of as the cumulative result of a working cluster rather than nine independent quizzes.
Results, Score Reports and Retake Mechanics
| Topic | What the sources say |
|---|---|
| Results timeframe | The exam page says usually three U.S. business days; the program guide allows up to five business days. |
| Score detail | The exam page describes total scores; the handbook also describes objective-area feedback. No task-by-task detail is provided. |
| Free retake | Eligible unsuccessful first individual attempts receive one free retake, subject to purchase or subscription eligibility periods and scheduling restrictions. |
| Delivery | Remote, testing-center and classroom delivery, subject to booking availability. |
The two results timeframes differ because they come from different documents. Plan on the longer window when making downstream commitments, such as telling an employer when you expect to be certified.
Key Takeaway
Do not treat the free retake as a plan. It is conditional on eligibility and scheduling, and your report will not tell you exactly which steps failed. Prepare to pass the first time, using the free retake only as a safety net you confirm in writing for your own booking.
For more on how attempts and outcomes tend to play out, see our discussion of the High Availability Clustering pass rate, which explains why we do not quote a statistic here.
A Scoring Strategy Built for Cluster Tasks
Order your work by dependency, not by the order listed
Read the entire task list first. Identify what everything else depends on: a functioning cluster, working fencing, and available shared storage. Complete and verify those foundations before layering services and filesystems on top. Spending your first stretch of the three hours on a stable base is usually better than racing through easy-looking items on a shaky cluster.
Verify like the grader will
- Check cluster and resource status after every meaningful change.
- Simulate a failure of a service or node (as the task permits) and watch where resources land.
- Confirm storage is mounted and accessible on every node that should have it.
- Reread the task wording against your final state, line by line.
Manage the clock without a known cut score
Because no pass mark is quoted here, the safe goal is to satisfy every requirement on every task you attempt. If a task stalls you, note where you left it, move on, and return with remaining time. A task that is 90 percent configured but fails a persistence check may earn nothing, so finish fewer tasks completely rather than many tasks loosely.
Sequencing Your Prep Around the Objectives
Preparation is a fixed-scope problem: nine objective areas, one three-hour practical. The supporting curriculum, the four-day RH436 course (documented against RHEL 8.3), is recommended, but class attendance is optional and equivalent experience is acceptable. Note that the exam page follows the most recent available product version without naming a release, so do not assume the exam is "RHEL 8.3" just because the course is documented that way. Practice on the current release you expect to see and confirm the version details when you book.
Foundation: cluster build and fencing
- Build a multi-node lab and form a cluster repeatedly until it is routine.
- Configure and test fencing early, since later tasks lean on it.
- Practice reboot recovery after every build.
Services and behavior
- Create resources and groups; apply ordering, colocation and location rules.
- Practice logging and monitoring tasks so you can diagnose failures quickly.
Storage, GFS2 and integration
- Prepare shared storage, create GFS2, then grow it while in service.
- Run timed, end-to-end labs of your own design and verify persistence.
Write your own practice scenarios rather than hunting for leaked or reconstructed exam content; original labs build the diagnostic skill the exam rewards. Our one-page cheat sheet is a good companion for quick recall, and you can pressure-test your knowledge with the practice questions on the main practice test site.
Fee, Delivery and Renewal Facts That Affect Your Attempt
- Fee: Red Hat's program-wide standard exam list price is USD 500. The regional booking-cart price controls, so treat that figure as a reference rather than an exam-specific checkout quote. Full breakdown in our certification cost guide.
- Prerequisites: RHCSA or RHCE is recommended, but neither is mandatory. See the requirements guide.
- Remote delivery: Remote candidates need LiveUSB, a successful compatibility test, and the required identity and hardware checks. Do the compatibility test well before exam day.
- Validity: Certifications are current for three years. Renewal is possible through the original exam or qualifying same-level or higher-level credentials under the current framework.
- Renewal caution: A legacy Red Hat renewal page still uses older, narrower wording. For pre-framework credentials, the FAQ says updated renewal and mapping rules apply to the active transcript after you pass a new exam following May 11, 2026. Consult the current handbook, the dated framework, and your active transcript rather than relying on the legacy page alone.
Scheduling windows are covered in our exam dates article.
Frequently Asked Questions
This article does not state a pass mark because one is not verified in the sources reviewed. The exam is one three-hour hands-on section graded against objective criteria. Check Red Hat's current handbook and your exam confirmation for any official scoring detail.
Without a published cut score, the safest approach is to satisfy every requirement on every task you attempt. Since results are graded on objective criteria and configurations must survive reboot, fully correct and persistent work matters more than touching many tasks loosely.
The exam page says usually three U.S. business days, while the program guide allows up to five business days. Plan around the longer window. You receive total scores, and the handbook describes objective-area feedback, but not task-by-task details.
Eligible unsuccessful first individual attempts receive one free retake under Red Hat's training policies, subject to purchase or subscription eligibility periods and scheduling restrictions. Confirm your eligibility for your specific booking before relying on it.
Yes. Configurations must persist across reboot, so enable services, store cluster configuration properly, make firewall and SELinux changes permanent, and ensure storage comes up without manual steps. Test by restarting a node in your practice lab.
- High Availability Clustering Study Guide 2026: How to Pass on Your First Attempt
- High Availability Clustering Pass Rate 2026: What the Data Shows
- High Availability Clustering Exam Dates 2026: Testing Windows, Deadlines & Scheduling
- How Hard Is the High Availability Clustering Exam? Complete Difficulty Guide 2026