docs(handover): Phase C restart note — bash tee + python buffering trap#691
Merged
Conversation
Update next-session entry doc 2026-06-02 evening with the buffering lesson: bash `python ... | tee log` pipeline + Python default stdout buffering held 30+ min of matrix runner output in pipe buffers, producing a 0-byte log and zero per-cell JSON output. Process killed at session close. Resolution for next session: - Use `python -u` (unbuffered) when piping to tee for long-running runs. PowerShell `Tee-Object` works similarly. - Phase C row in §2 table downgraded from ⏳ to ❌ NOT STARTED. - §1 first-commands updated with the -u flag + correct cells dir (workspaces/hotpot_eval/... not reports/research-runs/...). Lesson worth promoting to memory in next session if pattern recurs (currently 1 occurrence — not yet pattern grade). Quality delta: exempt (label: docs) Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Update next-session entry doc with Phase C restart instructions.
What happened
Phase C N=3 baseline at M_M was launched in background during session 2026-06-02 evening. After 30+ minutes, the log file (
reports/research-runs/alpha8-phase-c-sanity.log) and the bg task output were both 0 bytes, despite Python processes consuming significant memory (~2.4 GB resident for the JAMES server). Process killed at session close.Root cause (likely)
The launch command was:
```bash
python scripts/qvt_ablation_matrix.py ... | tee log
```
Python's default stdout buffering holds output until flush/close. When piped to
tee, the pipe buffer accumulated all matrix runner output but never flushed because Python wasn't told to flush per-line. The matrix runner usesprint()extensively and likely produced output, but it stayed buffered.Fix for next session
Use `python -u` (unbuffered):
```powershell
python -u scripts/qvt_ablation_matrix.py
--sector-cells C_rag-graph,C_rag-ontology--tiers M_M --n-runs 3 2>&1 | Tee-Object reports/research-runs/alpha8-phase-c-sanity.log
```
Also corrected the cells output directory in the entry doc — it lives at `workspaces/hotpot_eval/reports/research-runs/qvt-ablation-cells/` not `reports/research-runs/qvt-ablation-cells/`.
Changes
`Quality delta: exempt (label: docs)`
🤖 Generated with Claude Code