Self-hosted monitoring
Sit above it all.
Keep Grafana, Prometheus, and Loki for the deep dives. Perch reads them and folds live host, container, and uptime monitoring into the one dashboard you actually open every day. Additive, not rip-and-replace.
Perch 
Docker FIG.01 · overview console
Keep your stack. Ditch the tab mess.
You've spent a lot of time getting Grafana, Loki, Dozzle, Uptime Kuma, and more set up and configured, Perch doesn't ask you to tear any of it down, it sits on top (Like a Bird!), reads what you've already got, and folds in the live pieces you're still missing.
Plug in what you have
- Prometheus metrics
- Loki logs
- InfluxDB metrics
- Graphite metrics
Perch reads your existing sources. Your history, retention, and Grafana dashboards stay exactly where they were.
One pane
one login
Get the rest with Perch
- Live host metrics no node_exporter
- Docker stats + live logs no separate Dozzle
- HTTP health checks no separate Uptime Kuma
- Public status pages custom domains
- One alert ruleset Discord / Slack / ntfy
- SSO across all of it no proxy per app
One tiny agent per host. No exporters, no scrape configs, no new containers to babysit.
The result: the dashboard you open every morning instead of three, without throwing away the work you've already done.
The everyday view, out of the box
Host metrics, containers, logs, uptime, and alerts under one login and one set of rules, the operational layer you check constantly, with no dashboards to build and nothing new to scrape.
Server metrics
CPU, memory, disk, network, load average, and uptime across every host, sampled by a tiny agent and streamed live.
Read the docs →Live dashboard
Real-time WebSocket updates. No polling, no refresh button, the numbers move on their own.
Read the docs →Docker monitoring
Container stats, status, resource usage, and a live log viewer for every host.
Read the docs →Health checks
HTTP uptime monitoring with latency tracking and heartbeat history.
Read the docs →Alerting
Rules for health-check and container events, sent to Discord, Slack, or ntfy.
Read the docs →Status pages
Public-facing status pages with custom domains and full theming.
Read the docs →SSO / OAuth
GitHub, Google, Microsoft, GitLab, Discord, Okta, and any custom OIDC provider.
Read the docs →Data sources
Wire in Prometheus, Loki, InfluxDB, and Graphite.
Read the docs →Build your setup, copy, paste, done
Pick how you want to run Perch, whether it sits behind a reverse proxy, and if you'll host status pages on custom domains. The exact files and commands are generated below, live.
Deploy with
Reverse proxy
Status page custom domains
Routing so users' own domains (e.g. status.acme.com) reach the hub.
Separate tools, folded into one dashboard
Docker command to bring the whole stack up
Telemetry, trackers, or phone-home, ever
Hosts and status pages, no per-host pricing
A boring, dependable stack
8 components
Close two tabs today
Perch is free, open-source, and runs entirely on your own hardware. No account, no telemetry, no per-host pricing.