System Monitoring and Troubleshooting Questions

Explain a methodical approach to monitoring, diagnosing, and resolving system issues. Topics include instrumentation strategy, choosing and interpreting key metrics, centralized log collection and analysis, alerting and noise reduction, investigative workflows for performance and resource bottlenecks, use of diagnostic tools and command line utilities, runbooks and escalation paths, rollback and mitigation techniques, and post incident analysis to prevent regressions. Provide examples of incidents you diagnosed, how you isolated root causes, and what operational changes resulted.