[dezren39]

Closes #1039.

Problem

PR #843's family sync walks missing ancestors by ID, best effort: when an ancestor session.get fails (the .catch(() => undefined)), the walk breaks and the sync still completes, caching session.family:<id> as complete. Later family syncs — including reopening the same session — never retry the missing ancestor until an unrelated invalidation. resolveRoot then keeps grouping under the furthest-known ancestor: the #631 wrong-root symptom, with no error surfaced. The descendant walk already invalidates on a skipped branch; the ancestor walk had no mirror rule.

Fix

• Track ancestorsComplete: set false only when a parent GET fails or returns a mismatched ID (walk stopped early). Genuine termination (no parentID, seen-set cycle guard, or walking through cached info) still counts as complete — no invalidation loop.
• After publishing, invalidate session.family:<id> when the ancestor walk truncated OR the descendant walk was incomplete. syncDescendants still always runs for its publish side effect (no short-circuit).
• Regression test retries the ancestor walk on the next sync after a parent read failed: first family sync with a 500 on the root GET groups tentatively without root info; second sync re-GETs the root (asserted: 2 root GETs), resolves the true root, and completes the family.

Acceptance

• Truncated ancestor walk invalidates session.family:<id>, mirroring the descendant rule — next sync retries.
• Regression test: failing GET then recovered GET resolves the true root.
• Genuine termination still cached as complete (flag only set on fetch failure/mismatch; existing ci(app): e2e app-2 regression specs fail on subagent navigation/timeline visibility — reproduced on unrelated PR #607 (attempts 1+2) #631 + cycle tests pass unchanged).

Verification

• bun test test/solid-data.test.ts (packages/client): 40/40 pass with fix; new test fails on unpatched code (2nd root GET never happens) — true regression coverage.
• bun typecheck (packages/client): clean.
• Diff: 2 files, +61/-2 (fix + test only).

Notes

• Branch uses the queue worker namespace queue/4118-1039-ancestor-retry per the repo-queue runbook for traceability (repo AGENTS.md prefers slashless short names; runbook governs here).
• Task 4118 (openchamber-scheduled-lane).

---

^{Need help on this PR? Tag @codesmith-bot with what you need.}

[coderabbitai]

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 35 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

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: defaults
• Review profile: CHILL
• Plan: Advanced
• Run ID: 7e6ea9c1-242d-435e-b9ed-d59066f0f1e1

Commits

Reviewing files that changed from the base of the PR and between a1a1e34 and 8677c67.

Files selected for processing: 2

• packages/client/src/solid/data.ts
• packages/client/test/solid-data.test.ts

• 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

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[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 2026-10-31T02:11:53.751556Z	8677c67	PR opened

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.