Six Days In

Your service runs a 99.9% monthly availability SLO, giving roughly 43 minutes of error budget. A bad deploy on day 6 caused a 38 minute outage. Under a standard error budget policy, what should happen next?

Reset the budget once the postmortem ships and the bad deploy is rolled back, since the root cause is identified and cannot recur

Treat the remaining few minutes as nearly spent: pause risky launches and spend the rest of the month on reliability work until the budget resets

Continue shipping as planned, because the contractual SLA was not breached, no customer credits are owed, and the budget resets in 24 days anyway

Relax the objective to 99.5% for this month so the 38 minute outage fits inside the budget and the freeze policy is never triggered