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

Add arm64 image to the Pax nightly build #264

Merged
merged 6 commits into from
Oct 2, 2023
Merged

Conversation

yhtang
Copy link
Collaborator

@yhtang yhtang commented Sep 26, 2023

This will render both the Pax nightlies (pax:nightly-YYYY-MM-DD) and the CI builds multi-arch.

@yhtang yhtang self-assigned this Sep 26, 2023
@nouiz
Copy link
Collaborator

nouiz commented Sep 26, 2023

LGTM, but I can't review the details?
I get an error when I try to download the containers on x86:

$ docker pull ghcr.io/nvidia/jax-toolbox-internal:6316452742-base-multiarch
Error response from daemon: denied

How can I test it?

@yhtang
Copy link
Collaborator Author

yhtang commented Sep 26, 2023

LGTM, but I can't review the details? I get an error when I try to download the containers on x86:

$ docker pull ghcr.io/nvidia/jax-toolbox-internal:6316452742-base-multiarch
Error response from daemon: denied

How can I test it?

Our workflow already assigned a tag for the image, but it has not been built yet. It will be ready when the build-pax job completes.

That's an internal image. Did you docker login ghcr.io using a personal access token with NVIDIA SSO configured?

@yhtang
Copy link
Collaborator Author

yhtang commented Sep 26, 2023

This is the multiarch image produced by the workflow: ghcr.io/nvidia/jax-toolbox-internal:6316452742-pax-multiarch

@yhtang
Copy link
Collaborator Author

yhtang commented Sep 27, 2023

Pax MGMN tests all failed due to

slurmstepd: error: pyxis:     [ERROR] URL https://ghcr.io/v2/nvidia/jax-toolbox-internal/manifests/6316452742-pax-multiarch returned error code: 404 Not Found

There is an incompatibility between the pyxis enroot version available on our CI cluster and the format of the multi-arch manifest. I'll look into it.

@nouiz
Copy link
Collaborator

nouiz commented Oct 2, 2023

LGTM. When the CI pass, you can merge.

@yhtang
Copy link
Collaborator Author

yhtang commented Oct 2, 2023

Pax CI passed. I'll push one more minor fix for tag naming and then merge.

@yhtang yhtang merged commit 48f541f into main Oct 2, 2023
5 of 6 checks passed
@yhtang yhtang deleted the add-pax-arm64-nightly branch October 2, 2023 17:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants