100+ Free Harness Certified Chaos Engineering Practice Questions

Pass your Harness Certified Chaos Engineering (Developer / Administrator) exam on the first try — instant access, no signup required.

✓ No registration✓ No credit card✓ No hidden fees✓ Start practicing immediately

100+ Questions

100% Free

1 / 100

Question 1

Score: 0/0

A chaos engineer wants to validate that a Kubernetes Deployment has the expected number of replicas during and after a pod-delete fault. Which probe type is most appropriate?

HTTP Probe

Kubernetes Probe

Prometheus Probe

Command Probe

to track

2026 Statistics

Key Facts: Harness Certified Chaos Engineering Exam

Exam Fee

Harness University

Probe Types

Harness HCE Docs

CNCF Incubating

LitmusChaos Status

CNCF

2 Tiers

Developer & Administrator

Harness University

Probe Modes (SOT/EOT/Continuous/Edge)

Harness HCE Docs

100%

Resilience Score = all probes pass

Harness HCE Docs

The Harness Certified Chaos Engineering exams are free, online-proctored certifications covering chaos engineering principles, Harness HCE platform usage (ChaosHub, experiments, probes, Resilience Score, fault types), and CD pipeline integration. Developer is the entry-level tier; Administrator covers enterprise governance and infrastructure management. No formal prerequisites; Kubernetes and DevOps experience recommended. Preparation typically takes 20–40 hours.

Sample Harness Certified Chaos Engineering Practice Questions

Try these sample questions to test your Harness Certified Chaos Engineering exam readiness. Each question includes a detailed explanation. Start the interactive quiz above for the full 100+ question experience with AI tutoring.

1What is the primary goal of a steady-state hypothesis in chaos engineering?

A.To define the maximum fault injection rate for a system

B.To describe the measurable normal behavior of a system before and after an experiment

C.To identify the root cause of a previous production outage

D.To enumerate all possible failure modes in the infrastructure

Explanation: The steady-state hypothesis defines the measurable, observable behavior that indicates a system is functioning normally. Before running a chaos experiment, you validate the hypothesis; after the experiment, you validate it again to determine if the system maintained resilience. It typically references SLIs such as error rate or latency thresholds.

2Which Harness Chaos Engineering resource represents a reusable, declarative definition of how a specific fault is executed against a target?

A.ChaosEngine

B.ChaosExperiment

C.ChaosHub

D.ResilienceProbe

Explanation: A ChaosExperiment is the custom resource that declaratively defines the fault type, target selector, environment variables, and tunables for a specific chaos action. ChaosEngine is the per-run instantiation that references one or more ChaosExperiments and specifies the application under test. ChaosHub is the repository of fault templates, and ResilienceProbe validates steady state.

3In Harness Chaos Engineering, what does the Resilience Score of an experiment represent?

A.The percentage of infrastructure nodes that survived the fault injection

B.The ratio of successful resilience probes to total probes across all faults in the experiment

C.The mean time to recovery measured during the chaos run

D.The number of experiments that passed divided by the total experiments run in a GameDay

Explanation: The Resilience Score is calculated as the weighted percentage of successful resilience probes across all faults in an experiment. Each fault's probe success percentage (successful probes / total probes) is combined with a priority weight to produce the overall score, giving a quantitative measure of how well the system maintained its steady state during chaos.

4What is a blast radius in the context of chaos engineering?

A.The latency added to network traffic by a chaos fault

B.The scope of systems or components affected by a chaos experiment

C.The CPU overhead introduced by the chaos agent on target nodes

D.The number of probes that must pass for an experiment to be considered successful

Explanation: Blast radius refers to the scope of impact — how many services, pods, nodes, or users are potentially affected when a chaos fault is injected. A core chaos engineering principle is to start with the smallest viable blast radius (e.g., one pod, one zone) and expand only after multiple green runs to limit unintended harm.

5Which probe type in Harness Chaos Engineering executes a PromQL query and evaluates the result against a defined threshold?

A.HTTP Probe

B.Command Probe

C.Prometheus Probe

D.SLO Probe

Explanation: The Prometheus Probe allows users to run PromQL queries against a Prometheus endpoint and compare the result against an expected criterion during chaos. This makes it ideal for validating that metrics-based SLOs (e.g., error rate < 0.1%) remain within bounds throughout an experiment.

6A chaos engineer wants to validate that a Kubernetes Deployment has the expected number of replicas during and after a pod-delete fault. Which probe type is most appropriate?

A.HTTP Probe

B.Command Probe

C.Prometheus Probe

D.Kubernetes Probe

Explanation: The Kubernetes Probe performs CRUD operations against native and custom Kubernetes resources, making it ideal for checking the status of Deployments, replica counts, or other K8s objects. It can assert that a Deployment's available replicas match an expected value before and after fault injection.

7What is ChaosHub in Harness Chaos Engineering?

A.A Kubernetes namespace dedicated to running chaos infrastructure components

B.A git-backed repository of fault templates and experiment definitions that teams can import and share

C.A dashboard that displays the Resilience Score history across all experiments

D.The Harness module responsible for scheduling and triggering chaos pipelines

Explanation: ChaosHub is a git-backed catalog of chaos fault templates and experiment definitions. Teams can connect their own private ChaosHub repository or use the default Enterprise ChaosHub provided by Harness. This GitOps model allows versioning and sharing of chaos artifacts across teams.

8Which of the following best describes the pod-delete fault in Harness Chaos Engineering?

A.It kills the main container process inside a running pod without deleting the pod object

B.It forcibly deletes one or more Kubernetes pods to test the application's self-healing and rescheduling capabilities

C.It drains a Kubernetes node and evicts all pods to test node-level failover

D.It injects CPU stress into a pod to exhaust its resource limits

Explanation: The pod-delete fault forcibly deletes selected pods, triggering Kubernetes to reschedule them. This tests whether the application and cluster recover automatically within an acceptable time window. It is one of the most commonly used faults for validating Kubernetes workload resilience and restart policies.

9What is the key distinction between pod-delete and container-kill faults in Harness Chaos Engineering?

A.Pod-delete targets deployments; container-kill targets stateful sets only

B.Container-kill terminates the container process within the pod while the pod object remains, whereas pod-delete removes the entire pod object

C.Pod-delete removes all pods in a namespace; container-kill removes only one container per pod

D.Container-kill also removes the persistent volume claim attached to the pod

Explanation: Container-kill uses signals (e.g., SIGKILL) to terminate the container process while the pod object itself stays in Kubernetes; the container runtime restarts the container per the pod's restart policy. Pod-delete removes the pod object entirely, requiring Kubernetes to reschedule a new pod. This distinction tests different recovery paths.

10A GameDay in chaos engineering is best described as:

A.An automated nightly run of all chaos experiments in a CI/CD pipeline

B.A scheduled, collaborative exercise where teams run chaos experiments and practice incident response together

C.A feature flag system for gradually rolling out chaos faults to production

D.A Harness dashboard view that aggregates Resilience Scores over a calendar month

Explanation: A GameDay is a planned, cross-team event that simulates realistic failure scenarios. Teams run chaos experiments while simultaneously practicing detection, communication, and incident-response procedures. Unlike automated pipeline gates, GameDays prioritize organizational learning and preparedness, often involving on-call engineers and stakeholders.

About the Harness Certified Chaos Engineering Exam

The Harness Certified Chaos Engineering exam validates knowledge of chaos engineering principles and practical use of Harness Chaos Engineering (HCE), the enterprise platform built on the CNCF-incubating LitmusChaos project. The Developer tier covers core concepts — ChaosEngine, ChaosExperiment, ChaosHub, fault types (pod-delete, container-kill, resource hogs, network faults, HTTP chaos, cloud faults), resilience probes (HTTP/Command/Kubernetes/Prometheus/Datadog/SLO), and Resilience Score calculation. The Administrator tier adds infrastructure management, RBAC governance, Chaos Guard, CD pipeline integration, and multi-environment chaos at scale. Both exams are free at university-registration.harness.io.

Questions

20 scored questions

Time Limit

Not disclosed

Passing Score

Not disclosed

Exam Fee

Free (Harness)

Harness Certified Chaos Engineering Exam Content Outline

20%

Chaos Engineering Principles

Steady-state hypothesis, blast radius minimization, production experimentation, GameDays, cascading failure patterns, circuit breakers, fault tolerance, chaos maturity models

25%

Harness HCE Core Concepts

ChaosEngine, ChaosExperiment, ChaosHub (Enterprise and custom), chaos infrastructure/delegate (namespace vs cluster scope), LitmusChaos CRDs (ChaosRunner, ChaosResult, ChaosSchedule), chaos-as-code GitOps

20%

Resilience Probes

HTTP, Command, Kubernetes, Prometheus, Datadog, SLO, and Dynatrace probe types; SOT/EOT/Continuous/Edge probe modes; reusable probes; Resilience Score calculation with fault priority weighting

20%

Fault Types and Targets

Pod-delete, container-kill, CPU/memory/disk hog, I/O stress, network latency/packet-loss/DNS/HTTP chaos, node-drain/restart/taint, AWS (EC2/ECS/Lambda/RDS), Azure VM, GCP GKE node faults; fault tunables (TOTAL_CHAOS_DURATION, PODS_AFFECTED_PERC, SEQUENCE, RAMP_TIME)

15%

CD Pipeline Integration and Governance

Chaos quality gates in Harness CD pipelines, Resilience Score thresholds for gate pass/fail, RBAC and workspaces, Chaos Guard (maintenance windows, error budget conditions), cron-based scheduling, continuous chaos automation

How to Pass the Harness Certified Chaos Engineering Exam

What You Need to Know

Passing score: Not disclosed
Exam length: 20 questions
Time limit: Not disclosed
Exam fee: Free

Keys to Passing

Complete 500+ practice questions
Score 80%+ consistently before scheduling
Focus on highest-weighted sections
Use our AI tutor for tough concepts

Harness Certified Chaos Engineering Study Tips from Top Performers

1Understand the Resilience Score formula thoroughly — it appears in multiple question forms involving fault priority and probe success percentage

2Memorize all 7 probe types (HTTP/Command/Kubernetes/Prometheus/Datadog/SLO/Dynatrace) and their specific use cases

3Know the 4 probe modes (SOT/EOT/Continuous/Edge) and when each is appropriate

4Distinguish between pod-level faults (pod-delete, container-kill) and node-level faults (node-drain, node-restart, node-taint, node-cpu-hog)

5Understand blast radius concepts: PODS_AFFECTED_PERC, SEQUENCE (serial/parallel), and TOTAL_CHAOS_DURATION are the primary blast radius controls

6Practice the chaos experiment lifecycle: pre-chaos ramp → SOT probe → fault injection → continuous probes → EOT probe → cleanup → post-chaos ramp

7Know the LitmusChaos CRD hierarchy: ChaosEngine orchestrates → ChaosRunner executes → ChaosResult records; ChaosHub is the template catalog

8Review real HCE documentation at developer.harness.io/docs/chaos-engineering/ to understand key concepts with practical examples

Frequently Asked Questions

Is the Harness Certified Chaos Engineering exam free?

Yes. Both the Developer and Administrator tiers of the Harness Certified Chaos Engineering exam are completely free. You can register and take the exam at university-registration.harness.io without any exam fee. Retakes are also free.

What is the difference between the Developer and Administrator chaos engineering certifications?

The Developer certification validates foundational knowledge of chaos engineering principles, HCE core concepts (ChaosEngine, ChaosExperiment, ChaosHub), fault types, and resilience probes. The Administrator certification adds enterprise-level topics: deploying and managing chaos infrastructure at scale, RBAC governance, Chaos Guard, ChaosHub administration, and advanced CD pipeline integration.

What is LitmusChaos and how does it relate to Harness HCE?

LitmusChaos is an open-source CNCF Incubating project for Kubernetes chaos engineering. Harness Chaos Engineering (HCE) is built on top of LitmusChaos, adding enterprise features including RBAC, Workspaces, a hosted control plane, Harness CD pipeline integration, SLO Probe, Chaos Guard, and the Enterprise ChaosHub. Harness contributes to the open-source LitmusChaos project while offering HCE as the enterprise product.

What is a Resilience Score in Harness Chaos Engineering?

The Resilience Score is a weighted percentage that quantifies how well a system maintained its steady state during a chaos experiment. Each fault in the experiment has a priority weight and a probe success percentage (successful probes / total probes). The score is the priority-weighted average of probe success percentages across all faults. A score of 100% means all probes passed; 0% means all probes failed.

Which fault types are most important to know for the Harness Chaos Engineering exam?

Focus on Kubernetes faults: pod-delete, container-kill, CPU/memory hog, disk-fill, network latency, network packet loss, DNS chaos, HTTP chaos, node-drain, node-restart, and node-taint. Also know cloud faults (AWS EC2 stop, ECS task stop, Lambda faults, GCP GKE node, Azure VM stop) and the difference between pod-level and node-level faults.

How many probe types does Harness HCE support?

Harness HCE supports 7 probe types: HTTP Probe (health endpoint checks), Command Probe (shell commands), Kubernetes Probe (K8s API CRUD operations), Prometheus Probe (PromQL queries), Datadog Probe (Synthetic test results), SLO Probe (error budget validation via Harness SRM), and Dynatrace Probe (Dynatrace health evaluation). Probes can be configured in SOT, EOT, Continuous, or Edge modes.

How do I integrate Harness Chaos Engineering into a CI/CD pipeline?

Add a Chaos Engineering step to your Harness CD pipeline after the deployment stage. Configure the step with the experiment to run and a Resilience Score threshold. If the experiment's score is below the threshold, the pipeline step fails and blocks promotion to the next environment. This creates an automated resilience quality gate that runs with every deployment.