Skip to content

Add container-writable flag.#2084

Merged
nv-mollys merged 2 commits intoNVIDIA-NeMo:mainfrom
sudostock:container-writable-main
Jan 28, 2026
Merged

Add container-writable flag.#2084
nv-mollys merged 2 commits intoNVIDIA-NeMo:mainfrom
sudostock:container-writable-main

Conversation

@sudostock
Copy link
Copy Markdown
Contributor

@sudostock sudostock commented Jan 27, 2026

Many of the performance recipes require the container environment to be writable, the enroot default config is disabled.

Prevent errors out of box.

This is set in Nemo2 but did not get carried over: https://github.com/NVIDIA-NeMo/NeMo/blob/main/scripts/performance/executors.py#L83

Summary by CodeRabbit

Release Notes

Chores

  • Enhanced Slurm job execution configuration to enable write-accessible containers during performance testing runs. Containers now fully support file system write operations, providing improved flexibility for stateful and write-intensive containerized workloads. This enables applications requiring data persistence and file modifications during execution to function properly within containerized environments.

✏️ Tip: You can customize this high-level summary in your review settings.

Many of the performance recipes require the container environment to be
writable, the enroot default config is disabled. Prevent errors out of
box.

Signed-off-by: Alex Filby <afilby@nvidia.com>
@copy-pr-bot
Copy link
Copy Markdown

copy-pr-bot bot commented Jan 27, 2026

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.

@bdubauski bdubauski requested a review from malay-nagda January 27, 2026 17:11
@coderabbitai
Copy link
Copy Markdown
Contributor

coderabbitai bot commented Jan 27, 2026

📝 Walkthrough

Walkthrough

A single line was added to the slurm_executor function in the performance utilities module. The change introduces a container-writable flag to the srun_args list, enabling writable container access during Slurm job execution.

Changes

Cohort / File(s) Summary
SLURM Container Configuration
scripts/performance/utils/executors.py
Added --container-writable flag to srun_args list in slurm_executor to enable writable container access during Slurm runs

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

🚥 Pre-merge checks | ✅ 4
✅ Passed checks (4 passed)
Check name Status Explanation
Description Check ✅ Passed Check skipped - CodeRabbit’s high-level summary is enabled.
Title check ✅ Passed The title 'Add container-writable flag' accurately describes the main change: adding a container-writable flag to the srun_args list in the slurm_executor.
Docstring Coverage ✅ Passed Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Test Results For Major Changes ✅ Passed Single-line configuration flag addition to enable container writeability, addressing known issues with performance recipes and aligning with NeMo v2.

✏️ Tip: You can configure your own custom pre-merge checks in the settings.

✨ Finishing touches
  • 📝 Generate docstrings

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands and usage tips.

@malay-nagda malay-nagda requested a review from ko3n1g January 28, 2026 07:35
@nv-mollys nv-mollys enabled auto-merge (squash) January 28, 2026 17:58
@nv-mollys
Copy link
Copy Markdown
Contributor

/ok to test 28bd1c2

@nv-mollys nv-mollys merged commit 0d93dce into NVIDIA-NeMo:main Jan 28, 2026
48 checks passed
conver334 pushed a commit to conver334/Megatron-Bridge that referenced this pull request Jan 30, 2026
Signed-off-by: Alex Filby <afilby@nvidia.com>
Co-authored-by: nv-mollys <149841089+nv-mollys@users.noreply.github.com>
Signed-off-by: conver334 <conver334@gmail.com>
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.

4 participants