akadmin/headroom
1
0
Fork 0
mirror of https://github.com/headroomlabs-ai/headroom.git synced 2026-08-27 14:17:10 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions
headroom/scripts
History
JD Davis ff221e6346
ci: scope PR workflow runs by changed paths (#1067) 
## Description

Makes PR workflow runs more selective by routing docs-only changes to
docs validation instead of the full CI workflow, while preserving
workflow validation and existing code/e2e/release gates for applicable
changes.

## Type of Change

- [ ] Bug fix (non-breaking change that fixes an issue)
- [ ] New feature (non-breaking change that adds functionality)
- [ ] Breaking change (fix or feature that would cause existing
functionality to change)
- [ ] Documentation update
- [ ] Performance improvement
- [x] Code refactoring (no functional changes)

## Changes Made

- Added `pull_request.paths-ignore` to `.github/workflows/ci.yml` so
docs/wiki/markdown-only PRs do not queue the general CI workflow.
- Removed `.github/workflows/ci.yml` from the CI internal `code` path
filter so CI-only workflow edits can run workflow validation without
forcing Python/Rust code jobs.
- Added a docs PR validation job to `.github/workflows/docs.yml` for
`docs/**`, `wiki/**`, `mkdocs.yml`, and docs workflow changes.
- Reduced default docs workflow token permissions to `contents: read`,
with `contents: write` scoped only to the deploy job.
- Added docs workflow dry-runs to `scripts/validate-workflows.sh` so
local/CI workflow validation covers the new PR and manual docs paths.

## Testing

- [ ] Unit tests pass (`pytest`)
- [ ] Linting passes (`ruff check .`)
- [ ] Type checking passes (`mypy headroom`)
- [ ] New tests added for new functionality
- [x] Manual testing performed

### Test Output

```text
$ actionlint .github/workflows/ci.yml .github/workflows/docs.yml
# no output

$ act pull_request -W .github/workflows/docs.yml -n
*DRYRUN* [Deploy Documentation/validate] 🏁  Job succeeded

$ act workflow_dispatch -W .github/workflows/docs.yml -n
*DRYRUN* [Deploy Documentation/deploy] 🏁  Job succeeded

$ act pull_request -W .github/workflows/ci.yml -n
*DRYRUN* [CI/changes] 🏁  Job succeeded
*DRYRUN* [CI/commitlint] 🏁  Job succeeded

$ python -m mkdocs build
INFO    -  Documentation built in 1.28 seconds

$ bash scripts/validate-workflows.sh
# completed successfully; act dry-runs passed. Some unsupported runner-platform matrix entries are skipped by local act, as before.

$ git diff --check
# no output
```

## Real Behavior Proof

- Environment: Windows local checkout, branch `smart-pr-runs`, `act`
0.2.87, temporary local `actionlint` installed via `go install`.
- Exact command / steps: Ran `actionlint` against changed workflows,
`act` dry-runs for docs PR/manual paths and CI PR path, actual `python
-m mkdocs build`, full `scripts/validate-workflows.sh`, and `git diff
--check`.
- Observed result: Changed workflows lint cleanly; docs PR and manual
docs workflow paths dry-run successfully; CI PR dry-run still covers
`changes` and `commitlint`; MkDocs builds; repository workflow
validation script completes with the new docs dry-runs included.
- Not tested: Full non-dry-run GitHub Actions execution on hosted
runners before PR creation.

## Review Readiness

- [x] I have performed a self-review
- [x] This PR is ready for human review

## Checklist

- [x] My code follows the project's style guidelines
- [x] I have performed a self-review of my code
- [x] I have commented my code, particularly in hard-to-understand areas
- [x] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [ ] I have added tests that prove my fix is effective or that my
feature works
- [ ] New and existing unit tests pass locally with my changes
- [ ] I have updated the CHANGELOG.md if applicable

## Screenshots (if applicable)

N/A.

## Additional Notes

- No issue is linked because this PR was not opened for a specific
tracked issue.
- `mkdocs build` reports existing docs/nav warnings but exits
successfully; strict mode currently fails on existing warnings, so the
PR validation uses the deploy-compatible non-strict build.
- Python unit/lint/type checks are not applicable to this workflow-only
change.
2026-06-16 19:11:45 -07:00
..
fixtures feat(scripts): add Codex proxy reconnect-storm repro harness 2026-04-20 22:02:02 +07:00
tests fix(ci): make PR governance advisory (#1047) 2026-06-16 12:36:39 -05:00
audit_wheel_glibc_symbols.py fix(crusher): shim __libc_single_threaded for glibc < 2.32 + extend audit 2026-05-05 13:59:21 -07:00
build_rust_extension.sh refactor: single-wheel maturin build backend (fixes #355) 2026-05-03 13:16:41 -07:00
changelog-gen.py chore: renormalize line endings to LF 2026-04-24 15:33:30 +02:00
export_kompress_v2_onnx.py feat: switch Kompress default to kompress-v2-base with weight-only int8 ONNX (#799) 2026-06-09 23:28:40 -07:00
install-git-hooks.sh Fix CI lint failure by formatting PR governance scripts (#933) 2026-06-12 17:11:39 -05:00
install.ps1 feat: add lean-ctx context tool support 2026-05-11 17:54:17 -04:00
install.sh feat: add lean-ctx context tool support 2026-05-11 17:54:17 -04:00
pr-governance.py Fix CI lint failure by formatting PR governance scripts (#933) 2026-06-12 17:11:39 -05:00
README.md feat(scripts): add Codex proxy reconnect-storm repro harness 2026-04-20 22:02:02 +07:00
record_fixtures.py feat(rust): scaffold workspace + parity harness (phase-0) 2026-04-24 13:39:48 -07:00
refresh_model_limits.sh fix(rust): wire ICM compressor into Rust proxy on /v1/messages 2026-05-01 16:44:44 -07:00
replay_codex_ws_load.py fix(tests): ship scripts/replay_codex_ws_load.py so CI can import it 2026-05-14 13:44:41 -07:00
repro_codex_replay.py fix: replace asyncio.timeout with 3.10-compat shim in repro harness 2026-04-20 13:41:02 -05:00
smoke_issue_327.py fix(proxy): remove content-keyed TTL walker that conflated content with positional cache (#327) 2026-05-01 12:04:28 -07:00
sync-plugin-versions.py fix(proxy): lazy-import server to avoid fastapi crash (#442) 2026-06-10 12:44:23 -05:00
validate-workflows.sh ci: scope PR workflow runs by changed paths (#1067) 2026-06-16 19:11:45 -07:00
verify-versions.py fix: make proxy upgrades version-aware 2026-05-09 15:58:27 -07:00
version-sync.py fix: make proxy upgrades version-aware 2026-05-09 15:58:27 -07:00

README.md

scripts/

Utility scripts bundled with the Headroom repo. Most are one-off operator tools; a few are runnable as part of development workflows.

Reproducing the reconnect storm

repro_codex_replay.py reproduces the multi-agent Codex reconnect/retry storm against a local Headroom proxy (default http://127.0.0.1:8787), as described in wiki/plans/2026-04-17-codex-proxy-runtime-analysis.md under "Latest Correction". Use it to:

  • Regression-check that /livez stays responsive under a cold-start storm.
  • Empirically tune the Unit 4 pre-upstream semaphore default (HEADROOM_ANTHROPIC_PRE_UPSTREAM_CONCURRENCY).
  • Exercise the Codex WS lifecycle + Anthropic HTTP path simultaneously without needing to replay captured production traffic.

Run

# Default: 8 WS + 4 HTTP clients, 30s storm, p99 /livez must stay <= 500ms.
python scripts/repro_codex_replay.py

# Tighter budget, shorter run:
python scripts/repro_codex_replay.py \
    --url http://127.0.0.1:8787 \
    --ws-clients 16 \
    --anthropic-clients 8 \
    --duration 60 \
    --livez-threshold-ms 100

# Dump the full summary as JSON for downstream tooling:
python scripts/repro_codex_replay.py --json

Exit code:

  • 0 — warmup succeeded (or was skipped), storm ran for the requested duration, and /livez p99 stayed under --livez-threshold-ms.
  • 1 — soft assertion failed, proxy unreachable, or unhandled exception. Proxy-unreachable is detected and reported within ~5 seconds.

Fixtures

The script loads two hand-crafted, fully synthetic JSON fixtures:

  • scripts/fixtures/anthropic_replay_body.json — shape of a large agent reconnect replay /v1/messages?beta=true POST body.
  • scripts/fixtures/codex_response_create_frame.json — first Codex WS frame with the {"type": "response.create", "response": {...}} envelope.

Override via --ws-frame-fixture / --anthropic-body-fixture if you have captured traffic to replay instead.

Interpretation

  • /livez p99 under threshold means the event loop is not starved during the storm. If it rises with the semaphore unbounded (HEADROOM_ANTHROPIC_PRE_UPSTREAM_CONCURRENCY=10000) and drops back under the default, Unit 4's backpressure is working.
  • Codex WS: opened should equal --ws-clients. response.completed typically stays low when upstream auth isn't configured locally — the goal is handshake + relay wiring, not real upstream traffic.
  • Anthropic HTTP: ok_2xx + non_2xx + timed_out + errors should roughly equal attempted. Sustained non-zero timed_out during the storm is the failure signal the plan targets.

A smoke test at tests/test_scripts/test_repro_codex_replay_smoke.py exercises the script against a mock FastAPI server on every PR.

Install scripts

  • install.sh — POSIX installer.
  • install.ps1 — Windows PowerShell installer.

These are generated by the release pipeline; edit with care.