◈ MonitorMGR Demo

MonitorMGR Demo Documentation

Agent-based server monitoring: install a lightweight agent on each host and watch CPU, memory, disk, load and network stream into a live dashboard. The agent dials out to the panel over outbound HTTPS — there are no inbound ports to open on the machine you monitor.

Linux x86_64 Ubuntu 22.04+ · Debian 12+ Outbound HTTPS only

The Panel — this app — stores metrics and renders dashboards; it never reaches out to your servers. On each host you want to watch, a small agent posts a metrics snapshot to the panel over outbound HTTPS every few seconds, so there are no inbound ports to open on the monitored machine. Every agent authenticates with a per-host key that is issued once, at enrollment.

Because the flow is agent-to-panel, hosts behind NAT or a firewall work with no extra plumbing — if the box can reach the panel URL, it can be monitored.

Supported OS: Linux x86_64 (Ubuntu 22.04+, Debian 12+).

The big pictureBack to top ↑

Adding a host gives you a one-time enrollment token. On the server you want to monitor, run the installer as root with your panel URL and that token:

curl -fsSL https://monitor-demo.scriptgain.com/downloads/agent-install.sh | sudo bash -s -- \
  https://monitor-demo.scriptgain.com 

The installer downloads a small static agent to /opt/monitor-agent, trades the token for a permanent agent key, writes its config to /etc/monitor-agent/agent.json, and installs a monitor-agent systemd service that begins reporting immediately.

One line, per hostBack to top ↑

In the panel, open Hosts → Add Host and give it a name. You'll be shown a one-time enrollment token and the exact install command — copy that command and run it on the target server.

Within a few seconds the host flips from Pending to Online and its metrics start streaming in. The enrollment token is single-use and shown only once; if you need a new one, regenerate a fresh token from the host's page and re-run the installer.

From the panel to the boxBack to top ↑

Each snapshot the agent sends carries a full picture of the host's health:

CPU

Overall utilization plus a per-core breakdown.

Memory

Used vs. total RAM, plus swap usage.

Disk

Per-filesystem usage across every mounted volume.

Load

1, 5 and 15-minute load averages.

Uptime

Time since the last boot.

Network

Receive and transmit throughput (rx / tx bytes per second).

One snapshot, full healthBack to top ↑

Each host gets a live view with animated rings for CPU, memory and disk and sparklines for the recent time series. The page refreshes on its own every 5 seconds, so you're always looking at near-real-time numbers without reloading.

Offline detection. Every snapshot updates the host's last check-in time. If a host hasn't reported within the offline window (default 90 seconds, configurable), it's shown as Offline in the list and on its page — even if its last stored status said online — so a crashed agent or an unplugged server surfaces quickly.

Live & self-refreshingBack to top ↑

The agent runs as a systemd service, so it survives reboots and restarts on failure. Key paths:

  • Binary — /opt/monitor-agent/monitor-agent
  • Config — /etc/monitor-agent/agent.json (the panel URL and the per-host agent key)
  • Service — monitor-agent.service

Check its status, follow its logs, or restart it:

systemctl status monitor-agent
journalctl -u monitor-agent -f
systemctl restart monitor-agent

To retire a host, delete it in the panel and run systemctl disable --now monitor-agent on the server.

Runs as systemdBack to top ↑