Tag: monitoring

Grafana Agent hit EOL in November 2025. Here is how to migrate all your monitoring nodes to Alloy without losing your mind, your metrics, or your dashboards.

Replace five scattered agents with one OpenTelemetry Collector pipeline, metrics, logs, and traces unified with real config you can drop into a home server.

Falco watches every syscall your containers make and screams when something sketchy happens. Like someone exec'ing a shell inside your nginx container at 3am.

ELK does everything and wants all your memory. Loki does logging the Prometheus way, label indexes, not content, and runs on a fraction of the resources. Here's the honest comparison.

Prometheus scrapes metrics. Grafana makes them pretty. Alertmanager wakes you up at 2 AM. Here's how to wire all three together into a monitoring stack that actually works.

Wazuh gives you SIEM, HIDS, FIM, and threat detection in one stack. Here's how to deploy it in your home lab with Docker and actually use it.

ctop and lazydocker beat docker stats for live container monitoring. Keybinds, install options, multi-host setups, and tracking a memory leak with each.

You get 50 alerts a day and ignore all of them. That's not monitoring, that's noise. Here's how to build alerts people actually care about.

Shorter intervals = more data. But also more storage, CPU load, and potential instability. Here's the tradeoff you're actually making.

Build one Grafana dashboard that adapts to any host or environment using variables, query, custom, constant, and multi-select types with real examples.

Uptime Kuma goes past HTTP checks: TCP, DNS, and Docker monitors, push monitors for cron jobs, public status pages, and the scripting API that actually exists.

Docker health checks for Dockerfiles and Compose: configure HEALTHCHECK for PostgreSQL, Redis, Nginx, and Node.js with intervals, retries, and depends_on.