CKA Pass Rate 2026: Why No Official Rate Exists

There is no official CKA pass rate. The Certified Kubernetes Administrator exam is administered by the Cloud Native Computing Foundation together with the Linux Foundation, and neither organization publishes how many candidates pass — not on the CNCF exam page, not on the Linux Foundation exam detail page, and not in the candidate FAQ or the certification handbook. The only threshold either body publishes is the passing score, and the Linux Foundation's official CKA FAQ states it plainly: "For the CKA Exam, a score of 66% or above must be earned to pass." Every percentage circulating in 2026 as "the CKA pass rate" comes from somewhere other than the organizations that score the exam. This page is not going to add another one.

How the CKA is actually scored

The CKA is a performance-based exam. CNCF describes it as "an online, proctored, performance-based test that requires solving multiple issues from a command line" — you are dropped into live Kubernetes clusters and graded on whether the cluster ends up in the state the task asked for. There are no answer choices, no distractors, and nothing to guess at. You either made the pod schedule, the service resolve, or the node rejoin, or you did not.

The official parameters are short and unambiguous:

• Passing score: 66% or above (Linux Foundation CKA/CKAD/CKS FAQ)
• Time limit: 2 hours to complete the tasks
• Format: online, proctored, command line, live cluster
• Cost: $445 for the exam on its own
• Validity: 2 years from the date earned
• Exam environment: based on Kubernetes v1.35, with quarterly updates planned to track Kubernetes releases

Domain weightings, as published by both CNCF and the Linux Foundation, are Troubleshooting 30%, Cluster Architecture, Installation and Configuration 25%, Services and Networking 20%, Workloads and Scheduling 15%, and Storage 10%. Scores are emailed within 24 hours of finishing, and the achieved score also appears in the Results section of the exam preparation checklist in your portal.

Notice what is missing from that list: a fixed number of questions. Neither body publishes a task count, and because the environment is refreshed quarterly against new Kubernetes releases, the exam a candidate sits in January is not identical to the one sat in October. An aggregate pass rate across that would blend cohorts taking materially different exams — which is one plausible reason the publishers stay quiet, though they have never given a reason at all.

Why the pass rates you see online disagree

Search for the CKA pass rate and the first page of results will hand you numbers that cannot all be true at once. The spread runs from the low forties to the high seventies, and the pages quoting them share one trait: not one cites CNCF or the Linux Foundation, because there is nothing to cite. The numbers are back-formed from forum threads, from a single training provider's own student cohort, or from nothing identifiable at all.

The incentive behind the low numbers is worth naming. A course seller who tells you the CKA has a punishing failure rate has just explained why you need the course. A bootcamp that reports its own students passing at a high rate has just explained why you need that bootcamp. Both figures may be honestly measured within their own population and still tell you nothing about your odds, because neither population is a random sample of CKA candidates.

There is a blunter signal of how thin this topic is. As of August 2026, a page-one slot on this query is held by a service selling help cheating the exam. Search engines surface that kind of result when nothing authoritative exists to rank above it. The absence of a real number is not a gap in your research — it is the actual state of the world, and it is the single most useful thing to know before you plan your preparation.

What separates passing from failing candidates

Time is the binding constraint, not knowledge. Two hours against multiple hands-on tasks means the exam rewards fluency, not recall. Candidates who know exactly what to do but type it slowly, or who build manifests from scratch instead of generating them with kubectl, run out of clock. The Linux Foundation notes that exam hosts come with kubectl pre-configured with a k alias and Bash autocompletion; candidates who have never practiced with those conveniences leave time on the table.

Troubleshooting is the heaviest domain by design. At 30%, it outweighs every other area, and combined with Cluster Architecture, Installation and Configuration at 25%, more than half the exam is about diagnosing and repairing clusters rather than deploying into healthy ones. Study plans built around creating resources — the comfortable half of Kubernetes work — are misaligned with where the marks actually sit.

Open documentation is narrower than it sounds. The Linux Foundation permits access during the exam to the Kubernetes documentation and blog, the Helm documentation, and the Gateway API documentation, plus any links given in the task's own quick reference. Using the search function on kubernetes.io/docs is allowed, but the rules state you "must not open external search results." So the exam is open-book only if you already know which page the book is on. Candidates who plan to look things up as a substitute for practice discover that navigating docs under a clock is its own skill.

The 66% bar makes triage a legitimate strategy. You can miss roughly a third of the exam and still certify. That changes how a stuck task should be handled: abandoning one and coming back is usually better arithmetic than spending twenty minutes proving a point.

The retake changes the calculus more than any pass rate would. CNCF states that the exam "includes one free retake," and the Linux Foundation's exam listing bundles two exam attempts along with two exam simulation attempts. The certification FAQ adds that there is "no enforced wait period for retakes currently," and that a retake unlocks once the first attempt has been graded a No Pass — scoring happens within 24 hours of completion. You also have 12 months from purchase to sit the exam. A first attempt is therefore a diagnostic you have already paid for, not a single shot. Whatever the true first-attempt pass rate is, the rate that matters to you is the two-attempt one, and no one publishes that either.

The bottom line

Nobody knows the CKA pass rate, including the people writing confident numbers about it. What is official and verifiable is that you need 66%, you get two hours, the exam is hands-on in a live cluster running Kubernetes v1.35, Troubleshooting carries the most weight at 30%, the fee is $445, and enrollment includes a retake with no waiting period. Those facts are enough to build a study plan; a pass rate would not change a single decision inside it. The candidates who fail overwhelmingly fail on speed and on troubleshooting, not on breadth of Kubernetes trivia. Practice against a timer, in a terminal, fixing broken things.

The most useful next step is to find out where your own gaps are while a retake is still theoretical. Work through a free CKA practice test to see which domains you actually lose time in before you spend the $445.