Learn the vocabulary for planning, running, and reporting chaos engineering game days: "we will simulate", "expected vs actual behaviour", "rollback procedure", and game day report writing.
0 / 5 completed
1 / 5
In a game day planning document, an engineer writes: 'We will simulate a regional failure by blocking all traffic to us-east-1.' What is the correct interpretation?
'We will simulate' is the core game day planning verb: it signals a deliberate, controlled fault injection designed to reproduce a specific failure scenario. The phrase distinguishes a chaos experiment from a real incident, from a tabletop exercise (discussion only), and from testing in a simulated environment. Simulation in chaos engineering means real fault injection in a real (or near-production) environment, not a theoretical discussion or staging-only test.
2 / 5
A game day report shows: 'Expected behaviour: circuit breaker opens within 5 failed requests. Actual behaviour: circuit breaker opened after 47 failed requests.' How should a facilitator frame this finding?
Expected vs. actual behaviour comparison is the core analytical step in game day reporting. A mismatch is a finding — not a failure of the experiment, but a discovery of a gap between design intent and actual system behavior. The good report names the mismatch precisely, identifies the probable root cause (misconfigured threshold), quantifies the impact (42 extra cascading errors), and produces a specific action item.
3 / 5
What must a rollback procedure include to be considered complete before a game day?
Rollback procedures are a game day safety prerequisite. Requirements: (1) specific, not abstract — actual commands or UI steps; (2) verified to work before the game day — never test the rollback for the first time during the actual experiment; (3) fast — the rollback should take seconds or low single-digit minutes; (4) known to multiple people — not just the experiment author. If the rollback procedure is unclear or untested, the game day should not proceed.
4 / 5
What is the facilitator's primary responsibility during a game day, and what should they NOT do?
The facilitator's separation from the responders is a key game day design principle: if the facilitator joins the debugging effort, no one is watching safety conditions, keeping the session on track, capturing observations, or ready to trigger the kill switch. The facilitator role is equivalent to an exercise controller in a military drill — they set up the scenario and observe, they do not fight the fire. Post-game, the facilitator also runs the retrospective.
5 / 5
Which of the following is a well-structured game day report summary?
A well-structured game day report includes all key fields: the scenario (specific fault injected), the hypothesis (measurable steady-state assertion), actual behavior (metrics observed), the outcome (confirmed/rejected), finding (specific gap discovered), and action items (specific change, named owner, deadline). The vague alternatives — 'the team did a good job,' 'some resilience issues,' 'discuss at sprint planning' — do not create accountability or organizational memory.