Operations · Server Monitoring
Your server, always in sight. Installed with one command.
The provider.tools agent monitors CPU, RAM, disk, load and network every 60 seconds — installed with one line, updates itself, and removes itself when you are done. Measured and alerting from 3 datacenters in Germany.
60s
interval
3
datacenters
30 days
history
1 line
install
# Linux — one line
curl -sS https://provider.tools/api/v1/agent/install/TOKEN | sudo bash
# Windows (PowerShell)
irm "https://provider.tools/api/v1/agent/install/TOKEN?os=windows" | iex
✓ agent installed · reporting every 60s
CPU 8% · RAM 41% · Disk 27% · net ↑↓ 1.2 MB/s
Linux (root) — Windows: irm "…?os=windows" | iex
Multi-location monitoring
Measured from 3 datacenters in Germany
Frankfurt am Main, Falkenstein and Nürnberg measure independently. An outage only counts as down when a majority of locations sees it — network false positives never reach your inbox. And if our main system fails, the satellites keep alerting.
Frankfurt am Main
Checking status…
Falkenstein
Checking status…
Nuremberg
Checking status…
Features
Everything server monitoring needs
Live metrics
CPU, RAM, swap, disk usage per mount, load average and network I/O — every 60 seconds in your dashboard.
Per-mount disk monitoring
Every real partition individually — including a warning before the disk fills up.
Network I/O
Received and sent in KB/s — across all real interfaces, including OpenVZ and CloudLinux.
30-day history charts
Raw data for 48 hours, then aggregated hourly — 30 days of history stay available permanently.
Self-update & self-uninstall
The agent updates itself on new versions and removes itself when you delete the monitor.
Token auth, read-only
SHA-256 hashed tokens, no command execution from our side — the agent only collects, it controls nothing.
Set up in 3 minutes
How your server joins the monitoring
Create a monitor
Enter hostname and an optional label — Linux and Windows, nothing to install to get started.
Run one command
curl … | sudo bash on your server (Windows: one PowerShell one-liner). The agent runs as a systemd service or scheduled task.
Watch live & get alerted
Metrics appear instantly. Email alerts on outages or exhausted resources — with configurable thresholds.
Alerts
Get alerted before your customers notice
Instant emails on every incident — including recovery with outage duration. Email alerts from the Pro plan.
- Server offline: no agent report anymore (3× interval exceeded)
- Resource alert: CPU, RAM or disk above your threshold
- Recovery: resources back to normal — with duration
- Alerts come from 3 datacenters — even if provider.tools itself fails
alerts@provider.tools → du@firma.de
🔴 Server offline: webserver01
The server "webserver01" has not reported for 4 minutes. Possible causes: server powered off, network issue, or the agent service is not running.
Re-send suppressed · dashboard link included
FAQ
Häufige Fragen
How do I install the agent?
One command: curl -sS https://provider.tools/api/v1/agent/install/TOKEN | sudo bash — for Windows (PowerShell): irm "https://provider.tools/api/v1/agent/install/TOKEN?os=windows" | iex. The script elevates itself via UAC when needed.
Is my server safe?
The agent is read-only: it collects metrics and sends them to provider.tools. No command execution, no inbound access — connections always originate from your server.
Does the agent update itself?
Yes. On every report the agent checks the current version and updates itself automatically — on Linux and Windows.
How do I remove the agent?
Delete the monitor in the dashboard — the agent removes itself on its next report. Or manually: curl -sS https://provider.tools/api/v1/agent/uninstall | sudo bash
What happens if provider.tools itself goes down?
Your agent automatically fails over to our satellites in other datacenters and buffers there. After recovery all measurements are backfilled — no false alerts, no data gaps.
How long is data stored?
Raw data for 48 hours, then aggregated into hourly values — 30 days of history stay available permanently.
Your server monitors itself in 3 minutes.
Create a monitor, run one command — the rest happens on its own. Email alerts from Pro.