Deploying a Multi-Node Looking Glass Panel NetMirror with Docker

Recently, a customer asked us to run quad-network monitoring tests. After searching around for a Looking Glass panel, we realized that most of the existing projects on the market are old PHP-based solutions: outdated UIs, limited features, or abandoned for years — none of them really meet the needs of modern network operations or multi-line visibility.

So we decided to build our own and teamed up with a few friends to develop an all-new Looking Glass project — NetMirror. Our goal is to provide IDC operators, one-man ops engineers, and network enthusiasts with a more modern and user-friendly solution.

Special thanks to wikihost-opensource and Hybula, whose projects inspired many of our design and architecture decisions.

What is NetMirror?#

NetMirror is a feature-rich, visually appealing, modern Looking Glass server panel designed to display and test the status of multi-line network nodes. It supports common network diagnostic tools such as ping, traceroute, mtr, speedtest, etc., and can present nodes from multiple regions in a unified interface, making it easy for users to compare performance and reachability across different routes.

Official Demo#

Open-source repo: https://github.com/catcat-blog/NetMirror (stars are welcome!)

See it in action here!

✨ Features #

Modern UI: Responsive interface built with Vue 3 + TailwindCSS.
Network tools: Integrated suite including Ping, iPerf3, Speedtest, BGP endpoints, and more.
Real-time traffic: Live monitoring of network interface traffic.
Interactive shell: Simulated shell environment for basic diagnostics.
Easy deployment: Distributed as a single Docker container.
Customizable: Configure features and server details via environment variables.
Multi-node support: For providers with many regions, a module for master/secondary region switching is also included.
Security-focused design: Only the panel and related APIs are exposed; nodes communicate securely via gRPC.

🧪 Why NetMirror?#

If you want to provide network test services for your VPS, or showcase network quality across multiple regions, NetMirror is a lightweight yet complete option. You no longer need to maintain legacy PHP projects or wrestle with front-end/back-end integration — we’ve packaged everything for you.

📸 UI Preview#

Here you can define multiple data centers and quickly jump to the LG of other regions.

Deployment#

Docker Deployment#

Clone the repo#

1
git clone https://github.com/Yuri-NagaSaki/NetMirror.git cd NetMirror

Create the environment file#

Copy the example env file to create your own configuration.

1
cp .env.example .env

Note: If the .env.example file does not exist, you can create an empty .env file and add the variables you need from the table below.

Customize your configuration (optional)#

Edit the .env file to set your server location, public IP address, and other options.

Start the service:#

1
services:
2
  als:
3
    image: soyorins/netmirror:latest
4
    container_name: looking-glass-e
5
    restart: always
6
    network_mode: host
7
    user: root
8
    env_file:
9
      - .env
10
    volumes:
11
      - ./data:/data
12
      - ./.air.toml:/app/.air.toml
13
    healthcheck:
14
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:${HTTP_PORT:-80}/"]
15
      interval: 30s
16
      timeout: 10s
17
      retries: 3
18
    logging:
19
      driver: "json-file"
20
      options:
21
        max-size: "10m"
22
        max-file: "3"

1
docker compose up -d

The app will be available at http://<your-server-ip>. The default port is 3000 and can be changed via the HTTP_PORT environment variable.

ENV configuration explained#

1
# ALS - Another Looking-glass Server environment variables
2
# Copy this file to .env and fill in the values you need.
3

4
# --- Network settings ---
5
# IP address for the server to bind to. Leave empty to bind to all interfaces.
6
LISTEN_IP=0.0.0.0
7
# HTTP port to listen on.
8
HTTP_PORT=3000
9

10
# --- Server information ---
11
# String describing the server location.
12
LOCATION=BGP
13
# Public IPv4 address of the server. If empty, it will be auto-detected.
14
PUBLIC_IPV4=
15
# Public IPv6 address of the server. If empty, it will be auto-detected.
16
PUBLIC_IPV6=
17

18
# --- Site branding ---
19
# Logo settings, supports multiple formats:
20
# 1. Image URL: LOGO=https://example.com/logo.png
21
# 2. Base64 image: LOGO=data:image/png;base64,iVBORw0K...
22
# 3. Emoji: LOGO=🌐
23
# 4. Plain text: LOGO=NetMirror
24
# 5. SVG markup: LOGO=<svg viewBox="0 0 24 24">...</svg>
25

26
# Example: use a network icon SVG
27
LOGO=<svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2"><circle cx="12" cy="12" r="2"/><path d="m16.24 7.76-1.41 1.41M15.76 16.24l1.41 1.41M7.76 16.24l1.41-1.41M7.76 7.76l1.41 1.41M12 3v6M12 15v6M3 12h6M15 12h6"/></svg>
28

29
# Logo type (optional, leave empty for auto-detection)
30
# Supported values: auto, url, base64, emoji, text, svg
31
LOGO_TYPE=auto
32

33
# Other examples (uncomment to test):
34
# Use emoji: LOGO=🌐
35
# Use text: LOGO=NetMirror
36
# Use a simple SVG icon: LOGO=<svg viewBox="0 0 24 24" fill="currentColor"><path d="M12 2L2 7v10c0 5.55 3.84 9.95 9 11 5.16-1.05 9-5.45 9-11V7l-10-5z"/></svg>
37

38
# --- Feature toggles ---
39
# Set to 'true' or 'false' to enable/disable features.
40
# Display real-time NIC traffic charts.
41
DISPLAY_TRAFFIC=true
42
# Enable LibreSpeed (HTML5) speed test.
43
ENABLE_SPEEDTEST=true
44
# Enable Ping tool.
45
UTILITIES_PING=true
46
# Enable MTR tool.
47
UTILITIES_MTR=true
48
# Enable Traceroute tool.
49
UTILITIES_TRACEROUTE=true
50
# Enable Speedtest.net CLI integration.
51
UTILITIES_SPEEDTESTDOTNET=true
52
# Enable interactive fake shell.
53
UTILITIES_FAKESHELL=true
54
# Enable iPerf3 server tool.
55
UTILITIES_IPERF3=true
56

57
# --- Feature configuration ---
58
# File size list for download speed tests, space-separated.
59
# Example: "10MB 100MB 1GB"
60
SPEEDTEST_FILE_LIST=100MB 1GB 10GB
61
# iPerf3 server port range - start.
62
UTILITIES_IPERF3_PORT_MIN=30000
63
# iPerf3 server port range - end.
64
UTILITIES_IPERF3_PORT_MAX=31000
65

66
# --- Custom content ---
67
# Display sponsor information. Supports multiple formats:
68
# 1. Local file path:
69
#    - Markdown file: SPONSOR_MESSAGE=/data/sponsor.md
70
#    - HTML file: SPONSOR_MESSAGE=/data/sponsor.html
71
# 2. Remote URL:
72
#    - Markdown link: SPONSOR_MESSAGE=https://example.com/sponsor.md (content will be downloaded)
73
#    - HTML page: SPONSOR_MESSAGE=https://example.com/sponsor.html (embedded via iframe)
74
# 3. Direct content:
75
#    - Plain text: SPONSOR_MESSAGE="Thanks for using our service"
76
#    - HTML code: SPONSOR_MESSAGE="<div><h2>Sponsor Us</h2><p>Content...</p></div>"
77
#    - Markdown text: SPONSOR_MESSAGE="## Title\n**Bold** content"
78

79
# Test with the sample Markdown/HTML file under the data directory
80
SPONSOR_MESSAGE=data/sponsor-iframe-example.html
81

82
# Try other formats (uncomment to test):
83
# Use HTML iframe page
84
# SPONSOR_MESSAGE=/data/sponsor-iframe-example.html
85
# Use external website iframe
86
# SPONSOR_MESSAGE=https://example.com/sponsor
87
# Use direct HTML content
88
# SPONSOR_MESSAGE="<div style='text-align:center;'><h2 style='color:#0ea5e9;'>💝 Thanks for your support</h2><p>Your support keeps us going!</p><a href='https://github.com/sponsors/username' target='_blank' style='color:#0ea5e9;'>GitHub Sponsors</a></div>"
89

90
# --- Looking Glass node configuration ---
91
# Node list format: NAME1|LOCATION1|URL1;NAME2|LOCATION2|URL2
92
# Example: "London|London, UK|http://IP:Port;Singapore|Singapore, SG|http://IP:Port;Tokyo|Tokyo, JP|http://IP:Port"
93
LG_NODES=London|London, UK|http://IP:Port;Singapore|Singapore, SG|http://IP:Port;Tokyo|Tokyo, JP|http://IP:Port;Frankfurt|Frankfurt, DE|http://IP:Port;New York|New York, US|http://IP:Port;Los Angeles|Los Angeles, US|http://IP:Port

The image size is very small, only about 30 MB, so it can be deployed on virtually any machine.