docker: Error response from daemon unsatisfied condition: brand = tesla\\\\n\\\"\"": unknown.

[Wutever]

---

1. Issue or feature description

I run the command docker run --runtime=nvidia --rm nvidia/cuda nvidia-smi

2. Steps to reproduce the issue

hit enter

3. Information to attach (optional if deemed irrelevant)

docker: Error response from daemon: OCI runtime create failed: container_linux.go:348: starting container process caused "process_linux.go:402: container init caused \"process_linux.go:385: running prestart hook 1 caused \\\"error running hook: exit status 1, stdout: , stderr: exec command: [/usr/bin/nvidia-container-cli --load-kmods configure --ldconfig=@/sbin/ldconfig.real --device=all --compute --utility --require=cuda>=10.0 brand=tesla,driver>=384,driver<385 --pid=3088 /var/lib/docker/overlay2/681b52ecc456f043441a3ea22321c4b3842a182ec7ab15b094a24e614eb481c2/merged]\\\\nnvidia-container-cli: requirement error: unsatisfied condition: brand = tesla\\\\n\\\"\"": unknown.

[asolano]

Hi,

I was getting the same error on an older GTX 1070 card.
Have you tried using a previous tag, like nvidia/cuda:9.0-runtime instead?

Best,

Alfredo

[Wutever]

Hi,

I was getting the same error on an older GTX 1070 card.
Have you tried using a previous tag, like nvidia/cuda:9.0-runtime instead?

Best,

Alfredo

Hi, I am using GTX1080, will it also cause problem?

[RenaudWasTaken]

Hello!

Cuda 10 has some driver and brand requirements, as highlighted by this table:
https://github.com/NVIDIA/nvidia-docker/wiki/CUDA#requirements

It seems here that you don't have the correct driver. You should either:

• Run cuda-9.0-runtime as suggested above
• Upgrade your driver