Misconfiguring Alertmanager is the norm. Sensible routing, silencing, grouping, and on-call rotation patterns to prevent alert fatigue.
Read moreAlertmanager: Routing That Doesn’t Wake Your Team at 3am
Tag: prometheus
Container Monitoring: Beyond cAdvisor
cAdvisor was the historical default but today isn’t enough. How to combine eBPF, Kubernetes metrics, and APM for real container observability.
Read moreObservability and SLOs: Error Budgets That Get Met
SLOs only work if error budget is genuinely managed. How to define without ceremony and how to use them to balance speed and reliability.
Read morePrometheus: Writing Alerts That Won’t Get Ignored
A practical guide to writing Prometheus alert rules that reflect real problems rather than noise: symptoms vs. causes, SLOs, and the weight of the watchdog.
Read more