Skip to content
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.

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

Remove NVML linking from entrypoint script #2965

Merged
merged 3 commits into from
Jun 9, 2021
Merged

Conversation

jbkyang-nvi
Copy link
Contributor

Remove NVM linking. Since we have moved off of NVML to DCGM, we do not need
special linking in entry script.
This PR should merge after #2924
Solves #2600

CoderHam
CoderHam previously approved these changes Jun 7, 2021
Copy link
Contributor

@CoderHam CoderHam left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Did you check that after removing this the docker can run without root privileges?

@jbkyang-nvi
Copy link
Contributor Author

jbkyang-nvi commented Jun 7, 2021

Did you check that after removing this the docker can run without root privileges?

Yep. The following works:

docker run --rm --net=host -v/tmp:/tmp/host --user 8888 --gpus all -it --sh
m-size=1G gitlab-master.nvidia.com:5005/dl/dgx/tritonserver:kyang-remove-nvml-py3-devel

returns

// Licensing stuff...
find: '/usr/lib/ssl/private': Permission denied
find: File system loop detected; '/usr/bin/X11' is part of the same file system loop as '/usr/bin'.

And running it normally:

docker run --rm --net=host --privileged --gpus all -it --shm-size=1G
 gitlab-master.nvidia.com:5005/dl/dgx/tritonserver:kyang-remove-nvml-py3-devel

returns

// Licensing stuff ...
find: File system loop detected; '/usr/bin/X11' is part of the same file system loop as '/usr/bin'.

@jbkyang-nvi jbkyang-nvi force-pushed the kyang-remove-nvml branch 3 times, most recently from 1c0448a to fba0d89 Compare June 7, 2021 23:40
@deadeyegoodwin deadeyegoodwin linked an issue Jun 8, 2021 that may be closed by this pull request
@deadeyegoodwin deadeyegoodwin changed the base branch from main to kyang-dcgm June 8, 2021 23:55
@deadeyegoodwin deadeyegoodwin changed the base branch from kyang-dcgm to main June 8, 2021 23:56
@jbkyang-nvi jbkyang-nvi merged commit dc049e1 into main Jun 9, 2021
@jbkyang-nvi jbkyang-nvi deleted the kyang-remove-nvml branch June 9, 2021 17:26
@taomanwai
Copy link

Why #2600 is fixed, but running in privileged mode is needed ?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Development

Successfully merging this pull request may close these issues.

Docker image does not run under arbitrary non-root user
4 participants