Hi, I’m trying to use one of the ngc containers images on a HPC center. I’m facing a driver version issue.
I followed the instructions here :
When it gets to starting the container, I ran :
singularity run --nv -B $(pwd):/host_pwd lammps_24Oct2018.simg /bin/bash
(By the way, there’s a typo in the command in the documentation, a whitespace is missing after the -B flag)
What I got is the process stopped with :
WARNING: Underlay of /etc/hosts required more than 50 (78) bind mounts
WARNING: Underlay of /usr/bin/nvidia-cuda-mps-server required more than 50 (233) bind mounts
WARNING: Could not chdir to home: /pbs/home/a/myname
2019/10/22 15:36:49 GPU driver verification failed: Host driver 418.39 not compatible with container: >=410.48, ==384.00
The output of
nvidia-smi
is :
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 418.39 Driver Version: 418.39 CUDA Version: 10.1 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Tesla V100-PCIE... On | 00000000:3B:00.0 Off | 0 |
| N/A 28C P0 25W / 250W | 0MiB / 32480MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
As my driver version is 418.39, which is >=410.48, I don’t understand why it fails ?
Is there a possible workaround ?
If this post is not relevant to this section, feel free to move it.
Thanks !