CAPA course and practice exam: Certified Argo Project Associate
This CAPA course has 22 video lessons and 54 graded labs, and its CAPA practice exam follows the official exam objectives: a 20-question sampler free with an account, the full question bank and timed mock exams with Pro.
Argo Workflows, Argo CD, Argo Rollouts and Argo Events
Prepares you for CAPA: Argo Workflows, Argo CD, Argo Rollouts and Argo Events on Kubernetes.
Intermediate · 2 hours · 22 lessons · 54 labs
CAPA exam guide
The CAPA exam (Certified Argo Project Associate, by CNCF / The Linux Foundation): 90 minutes; pass mark: 75%; cost: USD 250 (includes one retake).
- Format: Online, proctored, multiple-choice
- Duration: 90 minutes
- Questions: not published
- Pass mark: 75%
- Cost: USD 250 (includes one retake)
- Valid for: 2 years
- Delivery: Online, proctored (PSI)
- Prerequisites: None
Exam domains and weights
- Argo Workflows: 36%
- Argo CD: 34%
- Argo Rollouts: 18%
- Argo Events: 12%
A 20-question practice sampler is free with an account; Pro adds the full question bank and timed mock exams.
10 CAPA sample questions with answers: free to read, no account needed.
Common questions
How long is the CAPA exam?
The CAPA exam lasts 90 minutes.
What is the CAPA passing score?
The pass mark is 75%.
How much does the CAPA exam cost?
USD 250 (includes one retake).
Is there a free CAPA practice test?
Yes. A 20-question practice sampler is free with an account; Pro adds the full question bank and timed mock exams.
Source: CAPA exam page (Linux Foundation Training), domains, weights and competencies (CNCF / The Linux Foundation), checked 2026-10-03.
Who it is for
Intermediate level.
For platform and DevOps engineers preparing for the CAPA exam, who run pipelines, deployments and releases with Argo Workflows, Argo CD, Argo Rollouts and Argo Events.
What you don't need yet:
- No previous Argo experience
- No real cluster: labs check your manifests and simulate Argo
- No cloud account needed
What helps:
- Kubernetes basics: pods, Deployments and kubectl
- Reading and editing YAML manifests
No formal prerequisites. The courses under "Take first" cover what helps.
Take first:
- Certified GitOps Associate (CGOA) Exam Prep — Beginner
- Certified Kubernetes Application Developer (CKAD) Exam Prep — Intermediate
- Certified Kubernetes Administrator (CKA) Exam Prep — Intermediate
Learning paths:
Nobody needs to know everything before they start. Each course builds on the one before, so a learning path takes you there step by step.
What this exam proves, and what it doesn't
CAPA is an online, proctored, multiple-choice exam: passing shows you know the Argo projects and how they fit together. It doesn't test hands-on skill; our graded labs check your manifests and simulate Argo, not a real cluster.
Exam objectives this course teaches
Certified Argo Project Associate (CAPA)
- Argo Workflows (36% of the exam): Understand Argo Workflow Fundamentals; Generating and Consuming Artifacts; Understand Argo Workflow Templates; Understand the Argo Workflow Spec; Work with DAG (Directed-Acyclic Graphs); Run Data Processing Jobs with Argo Workflows
- Argo CD (34% of the exam): Understand Argo CD Fundamentals; Synchronize Applications Using Argo CD; Use Argo CD Application; Configure Argo CD with Helm and Kustomize; Identify Common Reconciliation Patterns
- Argo Rollouts (18% of the exam): Understand Argo Rollouts Fundamentals; Use Common Progressive Rollout Strategies; Describe Analysis Template and AnalysisRun
- Argo Events (12% of the exam): Understand Argo Events Fundamentals; Understand Argo Event Components and Architecture
Labs
54 labs, 54 of them graded: you write the code, it runs, and graded labs check the result. The outline lists each lesson's labs.
What is free
Start free: First module free with an account. Pro: Every course, lab, handbook, practice exam and certificate.
- Pro monthly — $29 a month. Every course, lab, handbook, practice exam and certificate
- Pro yearly — $249 a year. Every course, lab, handbook, practice exam and certificate
Course outline
Start Here
- One Commit, Four Argo Projects (1 lab): You can say which Argo project handles an event, a pipeline of jobs, a Git-to-cluster sync and a progressive release, and validate each project's resources against its own schema with kubeconform.
- How The Exam Works: Weights, Pacing And A Plan (1 lab): You know the CAPA exam's published facts, how its four domains are weighted, and how to turn a diagnostic score into a study plan weighted by domain and gap.
Argo Workflows
- What Argo Workflows Runs: Controller, Server And One Pod A Step (2 labs): You can describe the Workflow resource, the workflow controller, the Argo Server and the executor containers, predict the pods a steps workflow creates, and choose the right argo command to stop, retry or resubmit a run.
- The Workflow Spec: Entrypoint, Arguments And Inputs (3 labs): You can read and write a Workflow spec: entrypoint, workflow arguments, template inputs and step arguments, and the run-wide settings for service account, deadline, pod clean-up and time to live, and explain why a schema-valid workflow can still fail validation.
- Template Types: Nine Ways To Do A Step (2 labs): You can name the nine template types and their two categories, choose between container, script, resource, suspend, HTTP, plugin and container set templates for a task, and predict which of them get a pod of their own.
- WorkflowTemplates, ClusterWorkflowTemplates And templateRef (3 labs): You can build a library of WorkflowTemplates and ClusterWorkflowTemplates, call their templates with templateRef (and clusterScope), submit a whole template with workflowTemplateRef or argo submit --from, and predict how arguments merge.
- DAGs: depends, Results And Who Decides The Phase (3 labs): You can write a DAG template with dependencies or depends expressions, predict when each task starts, which tasks are Omitted, what fail fast stops, and the DAG's final phase from its target tasks.
- Artifacts: Files That Outlive The Pod (3 labs): You can declare input and output artifacts, wire them between steps with from, pick an artifact repository and archive strategy, predict each artifact's key, and say what happens when an output file is missing.
- Retries, Exit Handlers, Conditions And Deadlines (3 labs): You can configure retryStrategy (limit, retryPolicy, backoff), write an exit handler that branches on workflow.status, use when and continueOn, set workflow and template deadlines, and predict the attempts, timings and final phase of a failing run.
- Data Processing Jobs And CronWorkflows (3 labs): You can build a list, fan-out, fan-in data job with withParam and parallelism, schedule it as a CronWorkflow with a timezone and concurrency policy, predict its run name and duration, and choose Allow, Forbid or Replace for an overlapping schedule.
Argo CD
- Argo CD: Components, Applications And Projects (2 labs): You can explain the GitOps pull model, name what the API server, repo server and application controller each do, write an Application with source, destination and sync options, and predict whether an AppProject admits it.
- Sync: Manual Or Automated, Prune And Self-Heal (3 labs): You can read an application's drift resource by resource, choose manual or automated sync with prune and selfHeal, predict what automated sync does on a new commit, a live edit or a failed sync, and pick the sync options a team needs.
- Sync Phases, Waves And Hooks (3 labs): You can order a sync with PreSync, Sync, PostSync and SyncFail hooks and sync waves, predict the exact order Argo CD applies resources in and where a sync waits or fails, and choose a hook delete policy.
- Helm, Kustomize And Applications With Several Sources (3 labs): You can say how Argo CD detects and renders Kustomize, Helm and plain directories, override images and names from the Application, rank Helm value sources by precedence, and build a multi-source Application that takes a chart from one repository and values from another.
- ApplicationSets And The App Of Apps (3 labs): You can choose a generator for an ApplicationSet, predict the Applications a git directory and list matrix produces with Go templates, set its sync policy and know when the app of apps pattern fits better.
- Health, Reconciliation And Fields Argo CD Should Ignore (2 labs): You can tell sync status from health, predict each resource's health and the app's from the built-in checks, add a custom check, explain how often Argo CD reconciles and what a refresh does, and write a narrow ignoreDifferences rule.
Argo Rollouts
- The Rollout: A Deployment That Can Stop Halfway (2 labs): You can explain what a Rollout keeps from a Deployment and what it adds, name the parts of Argo Rollouts, migrate a Deployment two ways, predict pod counts for a canary step without a router, and drive a Rollout with promote, abort, retry and undo.
- Canary And Blue-Green: Weights, Routers And One Switch (3 labs): You can predict canary pods and traffic with and without a router, explain why setCanaryScale needs one, follow a blue-green update through its preview and active Services, and choose the settings that decide when it switches.
- AnalysisTemplate And AnalysisRun: Letting Metrics Decide (3 labs): You can write an AnalysisTemplate with arguments, predict each measurement's verdict and a run's phase from its conditions and limits, plug analysis into a canary or blue-green Rollout, and say what a Failed or Inconclusive run does to the update.
Argo Events
- EventSource, EventBus, Sensor And Trigger (2 labs): You can name what each Argo Events component does, trace one event from an EventSource over the EventBus to a Sensor's trigger, and find why a Sensor never fires: the bus, the EventSource name or the event name.
- Filters, Parameters And A Workflow From A Push (2 labs): You can filter a Sensor's events by payload, context and time, pass fields from an event into the Workflow a trigger submits, predict which events fire a trigger, and use conditions to wait for more than one event.
Mock And Review
- Final Mock, Review And A Trap Sprint (2 labs): You can take a timed mock under exam rules, turn its results into a revision order by domain, review each miss down to the rule and the lesson behind it, run a trap sprint, and walk into the exam with a plan for both attempts.