Datadog
Monitoring and observability: metrics, logs, traces and alerts for servers and applications. The Infrastructure Pro plan costs $15 per host per month billed annually ($18 on demand).
Prometheus and Grafana, both free, cover metrics and dashboards, and Uptime Kuma covers availability monitoring. For a few servers, that is half a day. What you lose: more than a thousand ready-made integrations, logs and traces correlated in one place, and not having to operate it yourself (storage, updates, alerts that do not go down with the monitored server).
What you lose
- More than 1,000 ready-made integrations
- Logs, metrics and traces correlated in a single interface
- A service monitored outside your own servers
- Built-in alerting and on-call
Existing alternatives to Datadog
- PrometheusOpen source
Metrics collection and alerting (Apache-2.0), the standard of self-hosted monitoring.
- GrafanaOpen source
Dashboards and visualization (AGPL), built on Prometheus.
- Uptime KumaOpen source
Website and service availability monitoring with notifications (MIT).
The prompt to build it yourself
Set up self-hosted monitoring to replace Datadog on my few servers. Do not build a tool: deploy Prometheus, Grafana and Uptime Kuma. To do: - A docker-compose.yml with Prometheus, Grafana and Alertmanager, persistent volumes, automatic restart. - node_exporter on each server to monitor, with restricted access (private network or firewall). - A Grafana dashboard for CPU, memory, disk and network, and alerts for nearly full disk and stopped service. - Uptime Kuma to monitor my sites from a different machine than the one being monitored, with notification by email or Telegram. - Caddy in front of Grafana with HTTPS and authentication, nothing else exposed. - A backup of the configuration and dashboards. Out of scope: application traces, log collection, anomaly detection. README: launch, adding a server, update, and a reminder that if this server goes down, monitoring goes down with it.
Frequently asked questions
- Can Datadog be replaced by a custom-built tool?
- casevibecode verdict: Partly replaceable. Prometheus and Grafana, both free, cover metrics and dashboards, and Uptime Kuma covers availability monitoring. For a few servers, that is half a day. What you lose: more than a thousand ready-made integrations, logs and traces correlated in one place, and not having to operate it yourself (storage, updates, alerts that do not go down with the monitored server).
- What is Datadog used for?
- Monitoring and observability: metrics, logs, traces and alerts for servers and applications. The Infrastructure Pro plan costs $15 per host per month billed annually ($18 on demand).
- What alternatives to Datadog already exist?
- Prometheus (open source), Grafana (open source), Uptime Kuma (open source). Options to compare before committing to a custom build.
Community feedback
Readers tell what they built instead of this tool, or what broke along the way. Every entry is reviewed before publication.
No feedback published yet. Have you tried replacing this tool? Tell us about it.