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.

Sign up for GitHub

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

Jump to bottom

Pin nvidia-driver and cuda packages to working packages #496

Merged
merged 19 commits into from
Dec 19, 2024
Merged

Pin nvidia-driver and cuda packages to working packages #496

merged 19 commits into from
Dec 19, 2024

Conversation

sjpb
Copy link
Collaborator

@sjpb sjpb commented Dec 13, 2024 •
edited
Loading

  • Pins nvidia-driver and cuda packages to avoid a dnf depsolve error on install
  • Makes installing cuda packages optional
  • Only run nvidia-driver/cuda installs during image build, but can be done from an "extra" build on top of an existing fat image
  • Will error if the install changes the kernel, as this is assumed to break things
  • Extends CI extra build test to also include cuda role

Depsolve error with 565-open nvidia-driver stream was:

Depsolve Error occurred: 
 Problem: package cuda-12-6-12.6.3-1.x86_64 from cuda-rhel9-x86_64 requires cuda-runtime-12-6 >= 12.6.3, but none of the providers can be installed
...

Replaces https://github.com/stackhpc/ansible-slurm-appliance/pull/471/files

@sjpb sjpb marked this pull request as ready for review December 13, 2024 15:06
@sjpb sjpb requested a review from a team as a code owner December 13, 2024 15:06
@sjpb sjpb mentioned this pull request Dec 13, 2024
Closed
wtripp180901
wtripp180901 requested changes Dec 16, 2024
View reviewed changes
Copy link
Contributor

@wtripp180901 wtripp180901 left a comment

Choose a reason for hiding this comment

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

some nitpicks but otherwise LGTM

ansible/roles/cuda/defaults/main.yml Show resolved Hide resolved
ansible/roles/cuda/defaults/main.yml Outdated Show resolved Hide resolved
@wtripp180901
Copy link
Contributor

have we been able to test cuda on any GPU nodes yet btw?

@sjpb
Copy link
Collaborator Author

sjpb commented Dec 18, 2024

Image build: https://github.com/stackhpc/ansible-slurm-appliance/actions/runs/12397634280

@sjpb sjpb requested a review from wtripp180901 December 19, 2024 08:51
wtripp180901
wtripp180901 approved these changes Dec 19, 2024
View reviewed changes
Copy link
Contributor

@wtripp180901 wtripp180901 left a comment

Choose a reason for hiding this comment

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

LGTM assuming CI passes

@sjpb sjpb merged commit fed2d6e into main Dec 19, 2024
7 checks passed
@sjpb sjpb deleted the feat/cuda-pin branch December 19, 2024 10:09
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@wtripp180901 wtripp180901 wtripp180901 approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@sjpb @wtripp180901