chore(ci): try another gpu label#358
Conversation
There was a problem hiding this comment.
Pull request overview
Updates GPU CI runner labels and tidies a minor formatting issue in the troubleshooting guide.
Changes:
- Switch GPU test jobs to run on the
nemo-ci-aws-gpu-x2-containerrunner label. - Remove trailing whitespace in the troubleshooting documentation.
Reviewed changes
Copilot reviewed 1 out of 2 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| docs/user-guide/troubleshooting.md | Removes trailing whitespace in a paragraph for cleaner markdown formatting. |
| .github/workflows/gpu-tests.yml | Updates runs-on labels for GPU smoke/E2E jobs to target a different GPU runner pool. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| if: ${{ needs.changes.outputs.src == 'true' || needs.changes.outputs.test == 'true' || github.event_name == 'workflow_dispatch' }} | ||
| timeout-minutes: 30 | ||
| runs-on: nemo-ci-aws-gpu-x2 | ||
| runs-on: nemo-ci-aws-gpu-x2-container |
There was a problem hiding this comment.
runs-on was changed to nemo-ci-aws-gpu-x2-container, but the internal workflow documentation still lists the GPU Tests runner as nemo-ci-aws-gpu-x2 (see .github/workflows/README.md:157-158). Please update that table (or other relevant docs) so the documented runner labels match what the workflow actually uses.
| if: ${{ needs.changes.outputs.src == 'true' || needs.changes.outputs.test == 'true' || github.event_name == 'workflow_dispatch' }} | ||
| timeout-minutes: 55 | ||
| runs-on: nemo-ci-aws-gpu-x2 | ||
| runs-on: nemo-ci-aws-gpu-x2-container |
There was a problem hiding this comment.
This job’s runs-on label was updated, but .github/workflows/README.md still documents the GPU runner label as nemo-ci-aws-gpu-x2 (e.g. .github/workflows/README.md:157-158). Please update the workflow docs to reflect nemo-ci-aws-gpu-x2-container so future CI changes/debugging aren’t based on stale runner info.
Signed-off-by: Matt Kornfield <mkornfield@nvidia.com>
mckornfield
force-pushed
the
gpu-runner-setup-again/mck
branch
from
ebbe7e4 to
226570b
Compare
|
blegh, this doesn't work either |
No description provided.