error-budget-tests

Per [Google SRE - Embracing Risk], "the difference between [SLO] and [actual uptime] is the 'budget' of how much 'unreliability' is remaining for the quarter." When the budget is consumed, releases freeze. Tests verify this contract is enforced.

When to use

• Adopting SLO-based reliability discipline.
• Auditing whether the error-budget process actually fires (often: defined but never triggers).
• After an incident: did the budget burn correctly? Did alerts fire? Did the freeze take effect?

Step 1 - Define the SLI + SLO