# monitoring

Microservices give you deployment flexibility and team autonomy, but they'll absolutely destroy your ability to debug latency issues if you don't have the...

If you've spent any time with Prometheus, you've seen it. That red `DOWN` label in the Targets page, accompanied by the dreaded `connection refused` error....

Set up Grafana unified alerting with contact points, notification policies, silences, and on-call escalation that works in production.

Design Grafana dashboards that are actually useful in production — information hierarchy, panel types, variables, and layout patterns from the field.

Write Keepalived track_script health checks that trigger VIP failover based on application health, not just node availability.

Write Prometheus alerting rules that page on real problems, not noise — with practical PromQL, severity levels, and runbook patterns.

Configure Prometheus Kubernetes service discovery to automatically scrape pods, services, and nodes — no manual target management.

A practical PromQL reference covering request rates, latency percentiles, resource usage, and Kubernetes workload queries for real production dashboards.

Build a production Prometheus and Grafana monitoring stack from scratch — service discovery, recording rules, alerting, and dashboards.

PromQL cheat sheet with copy-paste query examples for rates, aggregations, histograms, label matching, recording rules, and alerting expressions.

Diagnose Linux performance bottlenecks with top, htop, vmstat, iostat, sar, iotop, and strace — a systematic approach to finding and fixing issues.

Use Prometheus recording rules to pre-compute expensive queries, speed up dashboards, and make SLO calculations reliable at scale.

How to configure Kubernetes HPA with Prometheus custom metrics so your workloads scale on what actually matters — not just CPU and memory.

Design Prometheus alerting rules that catch real incidents and ignore noise — practical patterns from years of on-call experience.

Build Grafana dashboards that surface real signals instead of decorating walls — a structured approach rooted in SRE principles.

A step-by-step guide to implementing SLOs and error budgets using Prometheus — from defining SLIs to building burn-rate alerts.