test(daemon): initialize rmcp target project before the engine

[ScriptedAlchemy]

Summary

Three commits on agent/863-rmcp-target-fixture → codex/tracedecay-total-redesign-plan-reopened:

1. Fixture isolation: initialize every rmcp_route mounted project before the in-process engine exists.
2. Production: sort clone_bodies after partitioned segment restore; keep incompatible-cursor rebuild on the V14 admission writer; publish text Ready before seating owners.
3. Tests: align page-alignment / invalid-cursor / graph-off decode probes with clone-body pages and V14 successors.

Motivation (#863)

• Isolation cluster: selected_target_rmcp_… failed 2/40 under --test-threads=16 with typed database is locked because mount_rmcp_target ran exclusive maintenance while the fixture engine was live on the same profile.
• Tip-alone runtime reds (fail alone on tip binary; blocked Linux runtime nextest): clone-body page packing, unsorted clone bodies on cold restore, wrong cursor corruption index, probe that self-decodes, Ready published after owner install, artifact count vs attached pointer.

Evidence

• rmcp before/after: 2/40 → 0/40 database is locked on the target identity.
• Tip-alone: hauler cc-9…cc-14 each 1/1 alone-green after the fixes.
• Codex threads: 0 unresolved.
• Exact-head CI: requeued on 7f991b120 (supersedes 86292a42e).

Checklist

• No secrets
• Exact-head CI green on 7f991b120
• Merge target: campaign base only (never master / never chore(release): release 0.1.0 #745)
• Full nextest CI lane has systemic cross-test isolation failures #863 stays open until two clean Linux full-lane runs (or issue-write 403)

  

[changeset-bot]

⚠️ No Changeset found

Latest commit: 7f991b1

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[github-actions]

Performance Comparison codex/tracedecay-total-redesign-plan-reopened → agent/863-rmcp-target-fixture

Total Elapsed Time: 8.41s → 8.23s (-2.1%)
CPU Baseline: 90.46µs → 90.24µs (-0.2%)
Benchmark ID: index-bench-timing

timing - Execution duration of functions.

+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| Function                                 | Calls                      | Avg                            | P95                            | Total                          | % Total                      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| tracedecay-index-bench                   | 1 → 1 (+0.0%)              | 8.41s → 8.23s (-2.1%)          | 8.41s → 8.23s (-2.1%)          | 8.41s → 8.23s (-2.1%)          | 100.00% → 100.00% (+0.0%)    |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| domain.canonical.sha256                  | 85787 → 85787 (+0.0%)      | 35.42µs → 34.86µs (-1.6%)      | 105.34µs → 103.68µs (-1.6%)    | 3.04s → 2.99s (-1.6%)          | 36.14% → 36.33% (+0.5%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| code_index.workers.install               | 81 → 81 (+0.0%)            | 36.22ms → 36.05ms (-0.5%)      | 196.74ms → 208.01ms (+5.7%)    | 2.93s → 2.92s (-0.3%)          | 34.89% → 35.47% (+1.7%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| query.artifact.batch.sqlite              | 5 → 5 (+0.0%)              | 306.97ms → 299.73ms (-2.4%)    | 362.81ms → 350.75ms (-3.3%)    | 1.53s → 1.50s (-2.0%)          | 18.26% → 18.20% (-0.3%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| code_index.build.and_publish             | 2 → 2 (+0.0%)              | 714.15ms → 703.52ms (-1.5%)    | 975.70ms → 966.26ms (-1.0%)    | 1.43s → 1.41s (-1.4%)          | 16.99% → 17.09% (+0.6%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| query.artifact.finalization.advance_wake | 28 → 28 (+0.0%)            | 47.72ms → 46.53ms (-2.5%)      | 320.34ms → 316.15ms (-1.3%)    | 1.34s → 1.30s (-3.0%)          | 15.89% → 15.83% (-0.4%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| query.artifact.batch.parallel_prepare    | 5 → 5 (+0.0%)              | 241.10ms → 235.26ms (-2.4%)    | 293.60ms → 284.16ms (-3.2%)    | 1.21s → 1.18s (-2.5%)          | 14.34% → 14.29% (-0.3%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| query.artifact.batch.postings            | 5 → 5 (+0.0%)              | 222.73ms → 219.72ms (-1.4%)    | 255.59ms → 250.87ms (-1.8%)    | 1.11s → 1.10s (-0.9%)          | 13.25% → 13.34% (+0.7%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| domain.canonical.json_bytes              | 206681 → 206681 (+0.0%)    | 5.15µs → 5.26µs (+2.1%)        | 4.31µs → 4.25µs (-1.4%)        | 1.06s → 1.09s (+2.8%)          | 12.65% → 13.20% (+4.3%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| code_index.extract.parser_artifact       | 276 → 276 (+0.0%)          | 3.76ms → 3.77ms (+0.3%)        | 6.05ms → 6.08ms (+0.5%)        | 1.04s → 1.04s (+0.0%)          | 12.35% → 12.64% (+2.3%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| code_index.restore.file_admit            | 260 → 260 (+0.0%)          | 3.77ms → 3.76ms (-0.3%)        | 6.78ms → 6.88ms (+1.5%)        | 979.43ms → 977.43ms (-0.2%)    | 11.65% → 11.87% (+1.9%)      |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| code_index.lexical_source.batch_stage    | 6 → 6 (+0.0%)              | 120.99ms → 121.94ms (+0.8%)    | 196.21ms → 193.07ms (-1.6%)    | 725.96ms → 731.61ms (+0.8%)    | 8.63% → 8.89% (+3.0%)        |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| code_index.build.materialize_full        | 1 → 1 (+0.0%)              | 691.57ms → 685.25ms (-0.9%)    | 692.06ms → 685.77ms (-0.9%)    | 691.57ms → 685.25ms (-0.9%)    | 8.23% → 8.32% (+1.1%)        |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| code_index.collect.materialize_full      | 1 → 1 (+0.0%)              | 677.08ms → 671.41ms (-0.8%)    | 677.38ms → 671.61ms (-0.9%)    | 677.08ms → 671.41ms (-0.8%)    | 8.05% → 8.16% (+1.4%)        |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+
| query.artifact.batch.postings.ngram_rows | 5 → 5 (+0.0%)              | 127.56ms → 125.83ms (-1.4%)    | 144.97ms → 145.88ms (+0.6%)    | 637.78ms → 629.15ms (-1.4%)    | 7.59% → 7.64% (+0.7%)        |
+------------------------------------------+----------------------------+--------------------------------+--------------------------------+--------------------------------+------------------------------+

---

Generated with hotpath-rs

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-16T00:03:54.920028Z	7f991b1	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 7f991b1205

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Publish Ready only after owner installation succeeds

If install_artifact_owners returns an error—for example because the reader exceeds its admitted ceiling or the reservation cannot shrink—the progress snapshot has already been permanently advanced to Ready, while no query owners were installed. The caller latches the projection failure but does not roll this phase back, so dashboard/MCP progress reports a ready generation that cannot serve queries; coordinate these transitions atomically or publish Ready only after successful installation.

AGENTS.md reference: AGENTS.md:L9-L12

Useful? React with 👍 / 👎.