Skip to content

mujoco_py's MjSim doesn't render (Missing GL version) whereas gym's robotic env works fine #432

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
ghost opened this issue Jul 3, 2019 · 5 comments
Open

mujoco_py's MjSim doesn't render (Missing GL version) whereas gym's robotic env works fine #432

ghost opened this issue Jul 3, 2019 · 5 comments

Comments

@ghost
Copy link

ghost commented Jul 3, 2019 •
edited by ghost
Loading

Describe the bug
When I try mujoco_py's MjSim and try to render the simulator, I get the classic "ERROR: GLEW initalization error: Missing GL version" error which has been raised by many people, however I failed to find a solution that worked out for me. This is even stranger considering that my gym environments using mujoco work completely fine (see following picture).
image

EDIT (comment on nvidia drivers)
As you can see on my .bashrc config, I don't have any path to nvidia drivers. I don't see any nvidia drivers being installed in my /usr/lib folder (however, I have a /usr/src/nvidia-390.116 folder with the different patches from the installation?). It doesn't seem to be a problem because I can still run gym's robotic envs and run all my experiment on my GPU.

Here is the easiest code to reproduce the error I get.
To Reproduce
import mujoco_py
import os
mj_path, _ = mujoco_py.utils.discover_mujoco()
print(mj_path)
xml_path = os.path.join(mj_path, 'model', 'humanoid.xml')
model = mujoco_py.load_model_from_path(xml_path)
sim = mujoco_py.MjSim(model)
print(sim.data.qpos)
sim.step()
print(sim.data.qpos)
sim.render()

Error Messages
/home/usernmame/.mujoco/mujoco200
[0. 0. 1.4 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ]
[-1.12164337e-05 7.29847036e-22 1.39975300e+00 9.99999999e-01
1.80085466e-21 4.45933954e-05 -2.70143345e-20 1.30126513e-19
-4.63561234e-05 -1.88020744e-20 -2.24492958e-06 4.79357124e-05
-6.38208396e-04 -1.61130312e-03 -1.37554006e-03 5.54173825e-05
-2.24492958e-06 4.79357124e-05 -6.38208396e-04 -1.61130312e-03
-1.37554006e-03 -5.54173825e-05 -5.73572648e-05 7.63833991e-05
-2.12765194e-05 5.73572648e-05 -7.63833991e-05 -2.12765194e-05]
ERROR: GLEW initalization error: Missing GL version

Desktop (please complete the following information):

  • OS: Ubuntu 18.04.01 LTS
  • Python version: 3.6.7
  • Mujoco version: 200
  • mujoco_py version: 2.0.2.2
  • GPU: Nvidia GeForce GTX 960M
    Environment
    My .bashrc file is the following:

export PATH=$PATH:/usr/local/cuda-9.0/bin
export LD_LIBRARY_PATH=/home/username/.mujoco/mujoco200/bin
export LD_LIBRARY_PATH=/usr/local/cuda-9.0/lib64:$LD_LIBRARY_PATH
export LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libGLEW.so:/usr/lib/x86_64-linux-gnu/libGL.so
export MUJOCO_LICENSE_PATH=$HOME/.mujoco/mjkey.txt

Additional context
I also tried using the 150 version of Mujoco paired with 1.50 version of mujoco_py but this lead to exactly the same problem.
Thanks for your help!! Let me know if I missed anything worth mentioning to debug this.
@ZiwenZhuang
Copy link

Have you checked your OpenGL version? It seems like the pyopengl only supports OpenGL version v1.1 to 4.4 (refer to here), although based on your Cuda version, that should not be the problem

@ghost
Copy link
Author

ghost commented Jul 7, 2019

Have you checked your OpenGL version? It seems like the pyopengl only supports OpenGL version v1.1 to 4.4 (refer to here), although based on your Cuda version, that should not be the problem

Thanks for your answer! Just checked, and it seems my OpenGL is too recent?
OpenGL version string: 4.6.0 NVIDIA 390.116

Should I downgrade my OpenGL?

@ZiwenZhuang
Copy link

Have you checked your OpenGL version? It seems like the pyopengl only supports OpenGL version v1.1 to 4.4 (refer to here), although based on your Cuda version, that should not be the problem

Thanks for your answer! Just checked, and it seems my OpenGL is too recent?
OpenGL version string: 4.6.0 NVIDIA 390.116

Should I downgrade my OpenGL?

Thank you for the version string. I thought the Nvidia driver version and OpenGL version are connected. That is only a guess based on what I have searched. I haven't tried that yet because I was worrying about reinstalling not only OpenGL, but also Nvidia driver and Cuda.
Now, it seems like it is only needed to downgrade the OpenGL version.

BTW: Have you encountered with other programs which use OpenGL under current settings?

@ghost
Copy link
Author

ghost commented Jul 7, 2019

Have you checked your OpenGL version? It seems like the pyopengl only supports OpenGL version v1.1 to 4.4 (refer to here), although based on your Cuda version, that should not be the problem

Thanks for your answer! Just checked, and it seems my OpenGL is too recent?
OpenGL version string: 4.6.0 NVIDIA 390.116
Should I downgrade my OpenGL?

Thank you for the version string. I thought the Nvidia driver version and OpenGL version are connected. That is only a guess based on what I have searched. I haven't tried that yet because I was worrying about reinstalling not only OpenGL, but also Nvidia driver and Cuda.
Now, it seems like it is only needed to downgrade the OpenGL version.

BTW: Have you encountered with other programs which use OpenGL under current settings?

thanks again for the time :) No i've never seen OpenGL give any problem. This is still weird that when I'm using gym (that uses mujoco) I don't have any problem (see above pic), and this only happens when using the MjSim from mujoco_py.
By the way, I did not have the pyopengl package but it didn't seem to be a problem before, it doesn't look like mujoco_py was depending on it.
If I can dowgrade the OpenGL version without changing the drivers that would be perfect, i'll try that and keep you updated on that.

@pvskand
Copy link

pvskand commented Jul 24, 2019

@PierreAlexSW any update on this?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@pvskand @ZiwenZhuang