Skip to content

fix(qualification): detect hash-only rescue by tensor absence, not any file - #13

Merged
Mathis (echobt) merged 2 commits into
mainfrom
fix/hash-only-detect
Oct 15, 2026
Merged

Mathis (echobt) merged 2 commits into
mainfrom
fix/hash-only-detect

Conversation

@echobt

@echobt Mathis (echobt) commented Oct 15, 2026 •

Copy link
Copy Markdown
Contributor

Hash-only rescue carries job.json and summary.json, so the "no file under
published/" test sent real hash-only trees down the full-rescue branch and
rejected them with qualification_artifact_custody_changed.

Summary by CodeRabbit

  • Bug Fixes
    • Hash-only comparisons now use the host inventory when expected round artifacts are absent. Unrelated files in published rounds no longer incorrectly trigger a full-rescue comparison.
    • When expected artifacts are present, comparisons verify the rescued artifacts.

…y file

Hash-only rescue carries job.json and summary.json, so the "no file under
published/" test sent real hash-only trees down the full-rescue branch and
rejected them with qualification_artifact_custody_changed.
@coderabbitai

coderabbitai Bot commented Oct 15, 2026 •

Copy link
Copy Markdown

Review in Change Stack →

Warning

Review limit reached

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Next included review available in 43 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used the included review currently available.

Learn how review limits work.

Review configuration:

⚙️ Run configuration
  • Configuration used: defaults
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: b4ea6d4e-a8ee-48c6-8e83-5055076705aa
📥 Commits

Reviewing files that changed from the base of the PR and between f298533 and ee7aa3c.

📒 Files selected for processing: 1
  • scripts/launch_cuda_run.sh

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info
⚙️ Run configuration
  • Configuration used: defaults
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: 291439e5-a2fa-47d3-be77-d376f2357318
📥 Commits

Reviewing files that changed from the base of the PR and between 419ec8b and f298533.

📒 Files selected for processing: 2
  • src/hypertrain/gpu_ops/network_qualification.py
  • tests/gpu_ops/test_network_gpu_qualification.py

Included review availability: This review used your included allowance. Your plan provides up to 1 included review per hour; 0 remain after this review.

📝 Walkthrough

The comparison now checks expected round artifacts to select hash-only rescue handling. The test fixture writes per-rank summary.json files and a job.json file in each published round.

Changes: Hash-only rescue

Layer / File(s) Summary
Rescue selection and test fixture
src/hypertrain/gpu_ops/network_qualification.py, tests/gpu_ops/test_network_gpu_qualification.py
compare checks for expected artifacts across result rounds when selecting hash-only handling. The test fixture writes a CUDA-labelled summary for each rank and an empty job.json for each published round.

Priority: ⬇️ Low

Estimated code review effort: 2 (Simple) | ~10 minutes

Change: Bug fix

Merge Risk: ⚪ Minimal · up to f2985

The rescue-mode check matches the repository’s hash-only artifact contract, so no actionable merge-blocking risk remains.

Pre-merge checks: Passed 4 · Failed 1

Failed Failed checks: 1 warning

Check name Status Explanation Resolution
Docstring Coverage Warning Docstring coverage is 66.67% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 3 functions across 2 files. Write docstrings for the functions missing them to satisfy the coverage threshold.
Passed 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 clearly and concisely identifies the main change: detecting hash-only rescues by tensor absence instead of any file.
Linked Issues check Passed Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check Passed Check skipped because no linked issues were found for this pull request.
✨ Finishing Touches: 💡 1
📝 Generate docstrings: 💡
  • Commit to this branch
  • Create a new PR
🧪 Generate unit tests (beta)
  • Commit to this branch
  • Create a new PR
  • Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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.

…t rescue ports, pinned miner host key

- die when the frozen tree is missing instead of using /data/hypertrain-tree
- refuse existing /data/hypertrain-h0/h1 before any spend (materialize would refuse an hour in)
- h1 rescue base port 18600: 16 workers bind base..base+15, 18570 overlapped h0
- record launcher.pid: abort is TERM on it (setsid ignores INT/HUP)
- default h1 miner ssh uses the admission-pinned known_hosts with StrictHostKeyChecking=yes
@echobt
Mathis (echobt) merged commit c372ca7 into main Oct 15, 2026
13 checks passed
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.

1 participant