NGC docker container fails to run

[Robert Towster]

I recently got a nano and I wanted to test digits but the docker container fails to run.

The nano is using the latest image. Anyone have any ideas or suggestions?

Thanks

robert@nano1:~$ sudo docker pull nvcr.io/nvidia/digits:19.07-tensorflow
19.07-tensorflow: Pulling from nvidia/digits
Digest: sha256:b0ff402b82e2293b26be7235d6b2660e58f39418180f073de071c88e25b62dc1
Status: Image is up to date for nvcr.io/nvidia/digits:19.07-tensorflow
robert@nano1:~$ sudo nvidia-docker run -it --rm nvcr.io/nvidia/digits:19.07-tensorflow
docker: Error response from daemon: OCI runtime create failed: container_linux.go:344: starting container process caused "process_linux.go:424: container init caused \"process_linux.go:407: 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 --video --require=cuda>=9.0 --pid=26240 /var/lib/docker/overlay2/3e102ace7ef877d710cd27beb43feec03228b2f835a1db35f3c02470f14db031/merged]\\\\nnvidia-container-cli: mount error: mount operation failed: /usr/src/tensorrt: no such file or directory\\\\n\\\"\"": unknown.
robert@nano1:~$

robert@nano1:~$ sudo docker pull nvcr.io/nvidia/digits:19.07-caffe
19.07-caffe: Pulling from nvidia/digits
Digest: sha256:b39590d4c3f1f99b30ed381665d4ca1f251c9b9221e7747d3e24780dd3a36153
Status: Image is up to date for nvcr.io/nvidia/digits:19.07-caffe
robert@nano1:~$ sudo nvidia-docker run -it --rm nvcr.io/nvidia/digits:19.07-caffe
standard_init_linux.go:207: exec user process caused "exec format error"
robert@nano1:~$

[Robert Towster]

This changes the error on the first one but I dont think I should have to do this.. 

robert@nano1:~$ sudo nvidia-docker run -it --rm -v /usr/src:/usr/src nvcr.io/nvidia/digits:19.07-tensorflow
standard_init_linux.go:207: exec user process caused "exec format error"
robert@nano1:~$

[Masato Morita]

Do you PATH in system environment variable?

# I think you should like this
# (You should change *username to your own.)
#refer to https://qiita.com/otomon10/items/ad965d5f8c1cfa8b1a41
https://github.com/dusty-nv/jetson-inference#system-setup

$ mkdir /home/*username/tensorflow_data
$ mkdir /home/*username/tensorflow_digits-jobs

$ sudo docker pull nvcr.io/nvidia/digits:19.07-tensorflow

$ sudo nvidia-docker run -it --rm nvcr.io/nvidia/digits:19.07-tensorflow-d -p 8000:5000 \

 -v /home/*username/tensorflow_data:/data:ro \
 -v /home/*username/tensorflow_digits-jobs:/workspace/jobs nvcr.io/nvidia/digits:19.07-tensorflow

# In browser http://localhost:8000

$ mkdir /home/*username/caffe_data
$ mkdir /home/*username/caffe_digits-jobs

$ sudo docker pull nvcr.io/nvidia/digits:19.07-caffe

$ sudo nvidia-docker run --name digits:19.07-caffe-d -p 8888:5000 \
 -v /home/*username/caffe_data:/data:ro \
 -v /home/*username/caffe_digits-jobs:/workspace/jobs nvcr.io/nvidia/digits:19.07-caffe

# In browser http://localhost:8888

# If does not work, restart the daemon and edit the docker file.

2019年8月29日木曜日 0時11分05秒 UTC+9 Robert Towster: