INFO

This English version is translated by Gemini 3 Flash.

Linux Deployment

[Docker users can ignore this] Special configuration instructions for running WebAI2API on Linux servers.

1. Install Necessary Dependencies

Essential dependencies for Linux command-line mode that allow you to run graphical applications in a Linux environment without a desktop environment.

Ubuntu/Debian

sudo apt-get update
sudo apt-get install xvfb x11vnc

CentOS/RHEL

sudo yum install xorg-x11-server-Xvfb x11vnc

Arch Linux

sudo pacman -S xorg-server-xvfb x11vnc

2. Run the Program

Run the program using a virtual display and view it remotely via VNC. (The program will handle all the setup for you.)

npm start -- -xvfb -vnc

This will automatically:

• Start the Xvfb virtual display.
• Start the x11vnc server.
• Allow you to view the VNC screen directly through the WebUI.

3. Connecting to the Program

Via WebUI (Recommended)

Once the service is started, visit the "VNC Display" page in the WebUI to view it directly.

Via SSH Tunnel

Tip

The port might not always be 5900; the program will automatically search for an available VNC port in the range 5900-5999.

# In your local terminal
ssh -L 5900:127.0.0.1:5900 root@Server_IP

Then use a VNC client to connect to 127.0.0.1:5900.

Alternative Method: Terminal X11 Forwarding

This method is not recommended unless you prefer to configure your own environment.

1. Install an X Server locally (e.g., VcXsrv, Xming).
2. Use a terminal that supports X11 forwarding (e.g., WindTerm).
3. Enable X11 forwarding in your SSH session.

ssh -X user@server

FAQ

Port Already Occupied

If port 5900 is already in use, the VNC server will automatically look for an available port in the 5901-5999 range.

Display Number Conflict

Xvfb will automatically search for an available display number starting from 50 to avoid conflicts with existing X servers.

Unable to Connect to VNC

Please check if the dependencies were installed successfully.