investigate

You can investigate production incidents by following a structured 5-phase workflow: scope the problem, gather evidence from metrics and logs, form hypotheses, test those hypotheses, and conclude with documented remediation actions.

Finding the root cause of a service degradation requires an evidence-driven approach. You capture metrics, logs, and events to build a robust timeline, then form and test hypotheses against this evidence to pinpoint the exact failure.

Yes, this structured root-cause analysis applies to latency spikes, errors, and outages across both cloud and on-prem environments. The methodology enforces a formal workflow that adapts to various production incident types.

Evidence gathering during an SRE incident response works by collecting metrics, logs, and events to construct a robust incident timeline. This formalized evidence phase ensures hypotheses are tested against factual data rather than assumptions.

You do not need specific tools to start an automated incident investigation as there are no dependencies required. The skill provides the structured 5-phase methodology itself, which you apply directly to your existing production environment.