Skip to content

Latest commit

 

History

History
51 lines (35 loc) · 1.38 KB

xpra.md

File metadata and controls

51 lines (35 loc) · 1.38 KB
Raw
How to run TDW on a Linux server

Remote rendering with xpra

It is desirable while debugging to be able to view a remote rendering process. In order to forward the X server to your local screen, we recommend Xpra in combination with VirtualGL.

Installation instructions

  1. Install Xpra on both client and server
  2. Install VirtualGL on server. Note: if using Docker, this is already included in the Docker image

Usage instructions

On server:

  1. Start xpra
xpra start :80
  1. Start TDW using VirtualGL
DISPLAY=:80 vlgrun -d :0 ./tdw.x86_64

On client:

xpra attach --ssh=ssh ssh/server_hostname/80

See tdw/Docker/start_container_xpra.sh for an example script that executes the server side steps.

Note, if using Docker, xhost +local:root needs to be set for the virtual display (xpra) and the X server used to run VirtualGL

DISPLAY=:0 
xhost +local:root

DISPLAY=:80
xhost +local:root

Notes: On newer Nvidia driver versions (>440), applications may be locked at 1 fps. For the fix, see here.

This is the last document in the "Misc. remote server topics" tutorial.

Next: X11 forwarding.

Return to the README