Skip to content

test: AUT-1399 support non-root CI checkouts - #740

Open
svcnemo-autobot wants to merge 12 commits into
NVIDIA-NeMo:mainfrom
svcnemo-autobot:ci/non-root-local-tests
Open

test: AUT-1399 support non-root CI checkouts#740
svcnemo-autobot wants to merge 12 commits into
NVIDIA-NeMo:mainfrom
svcnemo-autobot:ci/non-root-local-tests

Conversation

@svcnemo-autobot

@svcnemo-autobot svcnemo-autobot commented Aug 7, 2026

Copy link
Copy Markdown
Contributor

Background

Export-Deploy launchers assumed /workspace, passwd-backed users, and root-writable runtime state, preventing its CI image from testing arbitrary checkouts under numeric non-root UIDs.

What changed

  • derive coverage paths from the mounted checkout without Git metadata
  • provide fallback identity and writable XDG/UV caches
  • keep launchers rooted in the selected checkout
  • install and validate the ONNX-specific Transformers API in a writable overlay
  • cover /workdir, /tmp, no-Git, passwd-less, cache, and ONNX cases

Details

flowchart LR
  C[Selected commit] --> M[Non-root bind mount]
  M --> I[Existing CI image]
  I --> T[Checkout-relative launchers]
  T --> O[Writable checkout and caches]
Loading

The ONNX launcher resolves Transformers 4.51.3 outside the project, verifies HybridCache, and repairs incomplete cached overlays without rebuilding the CI image.

Tested

  • launcher/coverage regression suite: 11 passed
  • whole-tree Ruff and format checks
  • exact secrets-detector hook and shell syntax
  • non-root wheel build
  • overlay repair and cache-hit probes
  • prior L0 CPU/GPU, vLLM, and in-framework CI passed
  • exact-head CI pending

Signed-off-by: svcnemo-autobot <svcnemo-autobot@nvidia.com>
@copy-pr-bot

copy-pr-bot Bot commented Aug 7, 2026

Copy link
Copy Markdown

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@svcnvidia-nemo-ci

Copy link
Copy Markdown
Contributor

/ok to test ade0be3

Signed-off-by: svcnemo-autobot <svcnemo-autobot@nvidia.com>
@nemo-autobot-origin
nemo-autobot-origin Bot requested a review from a team as a code owner August 7, 2026 14:38
@svcnvidia-nemo-ci svcnvidia-nemo-ci added the force-run-all Apply this to bypass potential skips of CI label Aug 7, 2026
@svcnvidia-nemo-ci

Copy link
Copy Markdown
Contributor

/ok to test 092fe8c

Signed-off-by: svcnemo-autobot <svcnemo-autobot@nvidia.com>
@svcnvidia-nemo-ci

Copy link
Copy Markdown
Contributor

/ok to test dbc0e3b

Signed-off-by: svcnemo-autobot <svcnemo-autobot@nvidia.com>
@svcnvidia-nemo-ci

Copy link
Copy Markdown
Contributor

/ok to test 899d8b9

Signed-off-by: svcnemo-autobot <svcnemo-autobot@nvidia.com>
@svcnvidia-nemo-ci

Copy link
Copy Markdown
Contributor

/ok to test 393ed07

Signed-off-by: svcnemo-autobot <svcnemo-autobot@nvidia.com>
Signed-off-by: svcnemo-autobot <svcnemo-autobot@nvidia.com>
@svcnvidia-nemo-ci

Copy link
Copy Markdown
Contributor

/ok to test 47820b2

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

deploy export force-run-all Apply this to bypass potential skips of CI ONNX tests vLLM

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants