Linux server monitoring

Linux server monitoring from CPU to disk I/O

Monitor Linux servers, ARM machines and Raspberry Pi from one dashboard. The statsd agent runs as a background service and collects the system metrics you need to investigate resource pressure.

Start free →Set up monitoring ↓
Linux machine overviewIllustrative data · not a live reading
MetricExample readingWhat to inspect
CPU usage42%Processor load
Memory usage6.8 / 16 GBMemory and swap
Disk usage85%Capacity and I/O

What you can monitor

CPU and memory

Inspect processor usage, load and frequency together with memory usage, available memory and swap. Use these readings to distinguish CPU load from memory pressure.

Disks and network

Track storage capacity and disk I/O, plus interface addresses and transfer rates. A filesystem filling up and a busy disk are different problems; both views help explain the machine state.

Hardware and workloads

See hardware inventory and supported GPU readings. On hosts with Docker or Proxmox, the optional readers add container or guest metrics alongside system monitoring.

Start monitoring a Linux server

  1. 1. Install and authorize statsd

    Create a System Stats account, then run the installer on your server. Open the printed login link in a browser and approve the machine; a headless server can be authorized from another device.

    Terminal
    curl -fsSL https://linux-stats.com/install.sh | sh
sudo statsd login
sudo statsd status

  2. 2. Confirm metrics arrive

    The installer starts a systemd service and enables it at boot. Open the dashboard to confirm delivery. If metrics are missing, inspect service logs with the same permissions and configuration directory.

    Terminal
    sudo statsd logs

  3. 3. Adjust collection intervals

    Core CPU, RAM, disk and network modules default to one-second collection. Saved interval changes apply live unless a command-line flag or environment variable overrides the module.

    Terminal
    sudo statsd interval
sudo statsd interval cpu 2s

Linux support and practical limits

Requires systemd and root or sudo access. The installer starts the service and enables it at boot. Supported architectures are x86_64, arm64 and 32-bit ARM (armv6/armv7).

If systemd is unavailable, use the documented foreground run or container installation. A container sees only the filesystems, interfaces and devices exposed to it; native installation gives direct access to the host.

Available metrics depend on hardware, drivers and service permissions. NVIDIA GPU readings require nvidia-smi. Use statsd doctor gpu to inspect missing GPU data.

Collection interval and stored history resolution are separate. Free includes 5 machines and 14 days of history at one-minute resolution; paid plans provide longer history and one-second resolution. Check the pricing page for current limits.

Explore monitoring for your systems

Docker monitoring →

See which containers are running, how much CPU and memory they use, and whether their health checks are passing. System Stats brings Docker and machine metrics into the same dashboard.

Proxmox monitoring →

Keep Proxmox guest state and resource usage beside the machine that runs them. Install statsd on the node to discover virtual machines and LXC containers through the local Proxmox tools.

HTTP uptime monitoring →

Check whether an endpoint responds as expected, how long it takes, and when it becomes unavailable. Keep uptime and response-time history beside the machines and containers behind your service.

Install the agent and see it all

Free up to 5 machines forever. No credit card. Two minutes from landing here to live metrics streaming in.

Start free → Talk to us