Skip to content

Latest commit

 

History

History
133 lines (117 loc) · 4.21 KB

README.md

File metadata and controls

133 lines (117 loc) · 4.21 KB
Raw

GUI remote rendering with VirtualGl + VNC on Ubuntu Server (debian-based server)

Example requisites for installation:

Server example IP: 145.15.42.45
Server hostname: REMOTESRV
Username on server: testuser
Password for user: testpass
Password for VNC access: vncpass

Configuring server:

Install prerequisites:

sudo -i
# Openg GL necessary for Virtual GL:
apt-get install freeglut3-dev mesa-utils
# Install xserver (or reinstall):
apt-get --reinstall install xserver-xorg-core
# Extra packages for xserver:
 apt install x11-xserver-utils libxrandr-dev
# Install lightdm:
apt-get install lightdm
# Install nvidia-drivers
service lightdm stop
bash ./NVIDIA-Linux-x86_64-460.73.01.run --no-cc-version-check

Configure Xorg and turn off HardDPMS for nvidia cards:

Run nvidia-xconfig --query-gpu-info to obtain the bus ID of the GPU. Example:

GPU #0:
  Name      : Tesla M60
  ...
  PCI BusID : PCI:136:0:0

Create an appropriate xorg.conf file for headless operation:

sudo nvidia-xconfig -a --allow-empty-initial-configuration --virtual=1920x1200 --busid {busid}

Note: Foe headless server use additional flag: --use-display-device=None

Replace {busid} with the bus ID you obtained in Step 1.

If you are using version 440.xx or later of the nVidia proprietary driver, then edit xorg.conf and add

Option "HardDPMS" "false"

under the Device or Screen section.

Install VirtualGL:

# Login as super user
sudo -i
# Stop lightdm service
service lightdm stop
# Download VirtualGL:
wget https://downloads.sourceforge.net/project/virtualgl/2.6.5/virtualgl_2.6.5_amd64.deb
# Install VirtualGL:
dpkg -i virtualgl_2.6.5_amd64.deb
# Turn off nvidia services:
rmmod nvidia_drm
rmmod nvidia_modeset
rmmod nvidia_uvm
rmmod nvidia
# Run VirtualGL configuration:
/opt/VirtualGL/bin/vglserver_config
# Note: For all questions answer "Yes".

Installing VirtualGL creates new group "vglusers", so to use VirtualGL features we need to assosiate user with this group:

usermod -a -G vglusers testuser

Check whether installation is correct:

xauth merge /etc/opt/VirtualGL/vgl_xauth_key
xdpyinfo -display :0
/opt/VirtualGL/bin/glxinfo -display :0 -c

Install TurboVNC:

# Download TurboVNC:
wget https://downloads.sourceforge.net/project/turbovnc/2.2.6/turbovnc_2.2.6_amd64.deb
# Install TurboVNC:
dpkg -i turbovnc_2.2.6_amd64.deb
# Exit sudo mode:
exit

Forwarding graphics:

Run TurboVNC server with command:

/opt/TurboVNC/bin/vncserver

Note: 'vncserver' created a display that it will translate, after running the command, it will ask to set a password (e.g. 'vncpass' as password. Password can be reset by deleting file ~/.vnc/passwd) and created display with name "REMOTESRV:0" (or REMOTESRV:1, REMOTESRV:2, etc.). We need the number after server hostname. The other way to get this display name is to run command:

/opt/TurboVNC/bin/vncserver -list

Set display environment as selected display. E.g. if vncserver creted "REMOTESRV:2" display, set:

export DISPLAY=:2

After this step our server is set up. Let's connect to it remotely. Server creates display accessible by 5900 + {n} port, where n - is the number of display, selected by vncserver. In our example DISPLAY=2 so the port will be 5902. Connect to server by ssh with username/password (testuser/testpass for our example).

ssh testuser@145.15.42.45 -L 5902:localhost:5902

Reminder: IP 145.15.42.45 is an example IP. User you server IP or hostname. Note: this command will forward server port 5902 to your PC local port 5902

Install Turbo VNC Viewer on your machine (https://sourceforge.net/projects/turbovnc/files/2.2.6/)

Run Turbo VNC Viewer and set as "VNC server" the following:

"VNC server": localhost:5902
Password:     vncpass

After succesfull login you will see linux GUI and login screen. To enter set yout login/pass for the system, e.g.:

Login:    testuser
Password: testpass

How to run applications or scripts that uses VTK:

You should run any app or script using vglrun command, e.g. (remember ser propriate $DISPLAY variable value w.r.t. vncserver display):

vglrun glxgears
vglrun python show.py
vglrun gdb app