gpu is not detected

[ena2016]

Hi all,

I believe this is a simple fix but I am stuck on step "Run run_docker.py". I just put the input and output.

I'm running linux mint and I've had this working until the recent nvidia driver update and I performed the workaround by adding the new keys on Dockerfile.

I have an RTX 3090 gpu and have tried using the GPU Enumeration from 0-all on the "--gpu_devices=XX"

Any help is appreciated.

Input:
python3 docker/run_docker.py --fasta_paths=/home/eric/alphafold/SIVmac239_Env.fasta --max_template_date=None --model_preset=multimer --data_dir=/media/eric/NewCupcake --docker_user=0 --output_dir=/home/eric/alphafold/tmp/alphafold --gpu_devices=all

Output:
`I0526 07:57:35.722631 139808385271616 run_docker.py:113] Mounting /home/eric/alphafold -> /mnt/fasta_path_0
I0526 07:57:35.722714 139808385271616 run_docker.py:113] Mounting /media/eric/NewCupcake/uniref90 -> /mnt/uniref90_database_path
I0526 07:57:35.722770 139808385271616 run_docker.py:113] Mounting /media/eric/NewCupcake/mgnify -> /mnt/mgnify_database_path
I0526 07:57:35.722814 139808385271616 run_docker.py:113] Mounting /media/eric/NewCupcake -> /mnt/data_dir
I0526 07:57:35.722855 139808385271616 run_docker.py:113] Mounting /media/eric/NewCupcake/pdb_mmcif/mmcif_files -> /mnt/template_mmcif_dir
I0526 07:57:35.722901 139808385271616 run_docker.py:113] Mounting /media/eric/NewCupcake/pdb_mmcif -> /mnt/obsolete_pdbs_path
I0526 07:57:35.722949 139808385271616 run_docker.py:113] Mounting /media/eric/NewCupcake/uniprot -> /mnt/uniprot_database_path
I0526 07:57:35.722995 139808385271616 run_docker.py:113] Mounting /media/eric/NewCupcake/pdb_seqres -> /mnt/pdb_seqres_database_path
I0526 07:57:35.723045 139808385271616 run_docker.py:113] Mounting /media/eric/NewCupcake/uniclust30/uniclust30_2018_08 -> /mnt/uniclust30_database_path
I0526 07:57:35.723094 139808385271616 run_docker.py:113] Mounting /media/eric/NewCupcake/bfd -> /mnt/bfd_database_path
Traceback (most recent call last):
File "/usr/local/lib/python3.8/dist-packages/docker/api/client.py", line 268, in _raise_for_status
response.raise_for_status()
File "/usr/lib/python3/dist-packages/requests/models.py", line 940, in raise_for_status
raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 500 Server Error: Internal Server Error for url: http+docker://localhost/v1.41/containers/8fcc849fe7dc1991af9a8cec8707ee9b088c55c4fd52e02e849213de39a9cdac/start

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "docker/run_docker.py", line 264, in
app.run(main)
File "/usr/local/lib/python3.8/dist-packages/absl/app.py", line 312, in run
_run_main(main, args)
File "/usr/local/lib/python3.8/dist-packages/absl/app.py", line 258, in _run_main
sys.exit(main(argv))
File "docker/run_docker.py", line 234, in main
container = client.containers.run(
File "/usr/local/lib/python3.8/dist-packages/docker/models/containers.py", line 818, in run
container.start()
File "/usr/local/lib/python3.8/dist-packages/docker/models/containers.py", line 404, in start
return self.client.api.start(self.id, **kwargs)
File "/usr/local/lib/python3.8/dist-packages/docker/utils/decorators.py", line 19, in wrapped
return f(self, resource_id, *args, **kwargs)
File "/usr/local/lib/python3.8/dist-packages/docker/api/container.py", line 1111, in start
self._raise_for_status(res)
File "/usr/local/lib/python3.8/dist-packages/docker/api/client.py", line 270, in _raise_for_status
raise create_api_error_from_http_exception(e)
File "/usr/local/lib/python3.8/dist-packages/docker/errors.py", line 31, in create_api_error_from_http_exception
raise cls(e, response=response, explanation=explanation)
docker.errors.APIError: 500 Server Error for http+docker://localhost/v1.41/containers/8fcc849fe7dc1991af9a8cec8707ee9b088c55c4fd52e02e849213de39a9cdac/start: Internal Server Error ("could not select device driver "nvidia" with capabilities: [[gpu]]")`

[tomwardio]

Hi @ena2016! Thanks for reporting, it looks like there's an issue with your NVIDIA docker installation, maybe some of the suggestions in this NVIDIA issue might help?

Please do re-open if you have any issues with AFOS after resolving the above!

[mohsenumn]

If the issue persists for you, try the following:

pip install --upgrade "jax[cuda11_pip]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html