agency-sre-site-reliability-engineer

Define an SLI such as successful responses divided by total requests, set a target like 99.95% over a 30-day window, and attach burn-rate alerts. The error budget is the allowed failure margin, and it drives whether teams ship features or prioritize reliability work.

The four golden signals are latency, traffic, errors, and saturation. Latency measures request duration, traffic measures request volume, errors track failure rates by type, and saturation tracks resource usage like CPU, memory, and queue depth.

Burn-rate alerting measures how fast the error budget is consumed across short and long windows. A critical alert might use a 5-minute and 1-hour window with a 14.4x factor, while a warning uses 30-minute and 6-hour windows at 6x.

When the error budget is exhausted or burning too fast, reliability work takes priority over feature releases. If budget remains, teams can ship features, making the SLO the objective decision-maker rather than gut feeling.

MTTR measures how quickly a system recovers from failure, while MTBF measures time between failures. This Skill emphasizes tracking MTTR and building automated runbooks, since fast recovery matters more than preventing every possible failure.