Chaos Engineering Principles & Failure Injection

Answer key & explanations

Every question in this quiz, with the correct answer marked and an explanation of why it is right. Use it to revise before or after taking the quiz above.

1. 1.What is the primary definition of Chaos Engineering according to the Principles of Chaos Engineering?

   • ARandomly modifying production code to test developer alertness
   • BThe discipline of experimenting on a system in order to build confidence in its capability to withstand turbulent conditions in production✓ Correct
   • CShutting down the entire data center unannounced to test emergency electrical generators
   • DWriting unhandled exceptions directly into customer-facing web forms

   Correct answer: The discipline of experimenting on a system in order to build confidence in its capability to withstand turbulent conditions in production

   Chaos Engineering is the discipline of experimenting on a system to build confidence in its ability to withstand turbulent, unexpected conditions in production environments.

2. 2.What is the first step in designing a Chaos Engineering experiment?

   • AKilling the production database primary instance
   • BDefining 'steady state' as a measurable output of a system that indicates normal behavior✓ Correct
   • CSending an email warning all company executives that an outage is imminent
   • DRevoking all engineer access permissions to cloud dashboards

   Correct answer: Defining 'steady state' as a measurable output of a system that indicates normal behavior

   Before injecting failure, practitioners must define and measure the 'steady state' (normal business metrics like successful checkouts/min) to detect deviations during the experiment.

3. 3.What was the specific operational purpose of Netflix's original 'Chaos Monkey' tool?

   • AEncrypting movie files on AWS S3 to test video streaming quality
   • BRandomly terminating virtual machine instances in production during business hours to ensure systems survive instance loss✓ Correct
   • CInjecting fake movie reviews into the recommendation engine database
   • DGenerating artificial network traffic to overload customer service chat queues

   Correct answer: Randomly terminating virtual machine instances in production during business hours to ensure systems survive instance loss

   Chaos Monkey randomly terminated production EC2 instances during business hours, proving that stateless architectures and automated recovery could survive server loss without human intervention.

4. 4.What is 'Blast Radius' in the context of Chaos Engineering?

   • AThe physical explosion damage caused by an electrical failure in a data center
   • BThe scope and potential negative impact of a chaos experiment on real users and critical business operations✓ Correct
   • CThe number of lines of source code changed in a single pull request
   • DThe amount of network bandwidth consumed by automated testing scripts

   Correct answer: The scope and potential negative impact of a chaos experiment on real users and critical business operations

   Blast radius refers to the extent of disruption a chaos experiment can cause; best practices mandate starting experiments with a tightly contained blast radius (e.g., 1% of traffic, one canary container).

5. 5.What is a 'Stop Mechanism' (or Dead Man's Switch) in a chaos experiment platform?

   • AA physical emergency button mounted in the office kitchen
   • BAn automated safeguard that instantly halts the failure injection and rolls back the system if steady-state metrics breach safety bounds✓ Correct
   • CA cron job that reboots the server every twenty-four hours
   • DA software trigger that uninstalls the operating system kernel

   Correct answer: An automated safeguard that instantly halts the failure injection and rolls back the system if steady-state metrics breach safety bounds

   A stop mechanism monitors real-time business metrics; if an experiment causes severe collateral damage or exceeds defined thresholds, it instantly halts failure injection to protect users.

6. 6.How does Chaos Mesh or LitmusChaos inject network latency into a Kubernetes Pod?

   • ABy physically cutting the fiber optic cables connected to the server rack
   • BBy utilizing Linux kernel utilities like Traffic Control (tc) and NetEm inside the target Pod's network namespace✓ Correct
   • CBy modifying the application source code to add Thread.sleep() calls
   • DBy running an infinite while loop on the host CPU cores

   Correct answer: By utilizing Linux kernel utilities like Traffic Control (tc) and NetEm inside the target Pod's network namespace

   Kubernetes chaos tools use eBPF, iptables, or the Linux Traffic Control (tc) NetEm subsystem inside the target pod's network namespace to inject artificial latency, packet loss, or corruption.

7. 7.Why should Chaos Engineering experiments ideally be executed in production rather than purely in staging environments?

   • AStaging environments are completely free of charge to operate
   • BStaging environments rarely capture the authentic scale, traffic spikes, configuration drift, and third-party dependencies of real systems✓ Correct
   • CProduction systems have no monitoring tools installed
   • DCompany policies strictly forbid testing code before it reaches production

   Correct answer: Staging environments rarely capture the authentic scale, traffic spikes, configuration drift, and third-party dependencies of real systems

   Only production captures real-world conditions: realistic traffic patterns, third-party integrations, background churn, and actual scale. Staging environments rarely replicate these nuances accurately.

8. 8.What is a 'Cascading Failure' that chaos testing often seeks to uncover?

   • AA waterfall software development methodology project failure
   • BA failure in one component that triggers a domino effect of sequential outages across dependent components, collapsing the system✓ Correct
   • CA series of Git merge conflicts during a software deployment
   • DThe physical overheating of multiple server blades in an equipment rack

   Correct answer: A failure in one component that triggers a domino effect of sequential outages across dependent components, collapsing the system

   Cascading failure occurs when a localized fault (e.g., a slow cache) causes upstream services to exhaust threads or resources, propagating failures through the entire architecture.

9. 9.What failure mode is simulated by introducing DNS failure chaos into an environment?

   • AModifying the operating system's root passwords
   • BPreventing services from resolving hostnames to IP addresses, verifying fallback logic and connection pooling resilience✓ Correct
   • CDeleting all HTML templates from the web server directory
   • DEncrypting database transaction logs using AES keys

   Correct answer: Preventing services from resolving hostnames to IP addresses, verifying fallback logic and connection pooling resilience

   DNS chaos breaks domain name resolution, testing whether microservices cache IP mappings safely, use working fallbacks, or crash completely when upstream name resolution falters.

10. 10.What is the role of the 'Scribe' or 'Observer' during a Chaos GameDay?

    • AExecuting the destructive shell scripts on production terminals
    • BDocumenting timelines, observed behaviors, metric anomalies, and action items during the drill without executing injections✓ Correct
    • CMonitoring the building's fire alarm system
    • DOrdering refreshments for the participating engineers

    Correct answer: Documenting timelines, observed behaviors, metric anomalies, and action items during the drill without executing injections

    The Scribe focuses entirely on capturing notes, tracking telemetry observations, recording timeline milestones, and noting follow-up engineering action items as the experiment plays out.

11. 11.What is Chaos Kong in Netflix's Simian Army suite?

    • AA tool that randomly drops relational database tables
    • BA tool that simulates the complete outage of an entire AWS Availability Zone or Region to verify cross-region evacuation✓ Correct
    • CA tool that floods microservices with millions of fake user logins
    • DA tool that deletes Git repositories containing uncommitted code

    Correct answer: A tool that simulates the complete outage of an entire AWS Availability Zone or Region to verify cross-region evacuation

    Chaos Kong tests large-scale infrastructure resilience by simulating the loss of an entire cloud region, verifying that traffic can be evacuated and served from alternative regions without incident.

12. 12.How does I/O Error Injection (e.g., via Chaos Mesh) test database and storage resilience?

    • AIt physically removes SSD drives from server motherboards
    • BIt uses FUSE or kernel hooks to return read/write I/O errors to applications, testing corrupt sector and disk failure handling✓ Correct
    • CIt fills up the root partition with temporary log files
    • DIt re-formats the host hard drive into FAT32 format

    Correct answer: It uses FUSE or kernel hooks to return read/write I/O errors to applications, testing corrupt sector and disk failure handling

    I/O injection uses FUSE file systems or kernel tracing to return synthetic EIO or delay codes on system calls (read/write), validating how databases handle disk stalls and corruption.

13. 13.What is the primary danger of practicing 'Chaos Testing' without automated observability?

    • AThe testing tools will refuse to run on modern Linux kernels
    • BEngineers cannot distinguish normal system behavior from failure-induced degradation, invalidating the experiment✓ Correct
    • CThe server's network bandwidth is halved permanently
    • DIt triggers automated Git commits to the main branch

    Correct answer: Engineers cannot distinguish normal system behavior from failure-induced degradation, invalidating the experiment

    Without comprehensive observability (metrics, logs, traces), it is impossible to verify whether steady state was preserved or understand how failure propagated through the system.

14. 14.In chaos experimentation, what is the 'Hypothesis'?

    • AA wild guess made by an engineer without looking at the architecture
    • BA precise, testable prediction that the system will maintain its defined steady state despite the injected failure✓ Correct
    • CA mathematical equation calculating the financial cost of downtime
    • DA written apology sent to customers after an incident

    Correct answer: A precise, testable prediction that the system will maintain its defined steady state despite the injected failure

    A chaos hypothesis explicitly predicts how the system should behave: 'Even if we terminate the primary database instance, the checkout service will continue to succeed for >99.9% of requests using replicas.'

15. 15.What does Clock Skew (or Time Chaos) test when injected into distributed environments?

    • AWhether developers arrive at meetings on time
    • BHow the application handles desynchronized system clocks, testing token expirations, cache leases, and distributed lock validity✓ Correct
    • CThe battery life of physical motherboard CMOS batteries
    • DThe rendering speed of CSS animations in the browser

    Correct answer: How the application handles desynchronized system clocks, testing token expirations, cache leases, and distributed lock validity

    Clock skew experiments manipulate physical time inside target containers, revealing bugs in time-sensitive operations like JWT validation, cache leases, and consensus timeouts.

16. 16.What is the difference between Disaster Recovery (DR) testing and Chaos Engineering?

    • ADR testing verifies procedural recovery after catastrophic events; Chaos Engineering continuously tests systemic hypotheses to prevent outages proactively✓ Correct
    • BDR testing only applies to physical tape backups; Chaos Engineering only applies to web applications
    • CDR testing is fully automated; Chaos Engineering is entirely manual
    • DThere is no difference; they are identical processes with different corporate branding

    Correct answer: DR testing verifies procedural recovery after catastrophic events; Chaos Engineering continuously tests systemic hypotheses to prevent outages proactively

    Disaster recovery testing is typically an infrequent, scheduled validation of backup and restore procedures; Chaos Engineering is an ongoing experimental practice to discover unknown architectural weaknesses proactively.

More free quizzes

• ITDistributed Systems: Consensus & Fault ToleranceDeep dive into distributed systems engineering, covering consensus protocols (Raft/Paxos), the CAP theorem, and vector clocks.16 questions
• ITLLM Engineering: Fine-Tuning, Alignment & ServingTest your knowledge of transformer architectures, PEFT techniques (LoRA), RLHF, vLLM, and production deployment strategies.16 questions
• ITApplied AI: Systems, Embeddings & RAGAssess your understanding of practical machine learning architectures, embedding spaces, and Retrieval-Augmented Generation systems.16 questions