Expire backend-negative availability observations per slot

[mjc]

Summary

• store negative-availability timestamps per backend slot
• expire each backend observation independently
• retain compatibility for older on-disk cache versions
• retire stateful session connections on client disconnect as part of this stacked branch's lifecycle policy

Branch-added tests

• A deterministic cache test updates one backend’s negative observation, advances time, and verifies expiration does not incorrectly preserve or invalidate other slots.
• A codec roundtrip test verifies V8 persists and restores per-backend negative timestamps.
• Legacy serialized availability versions and property-based availability behavior remain covered.

The stateful lifecycle adjustment is called out here because it is included in this branch's diff; it ensures a client-disconnected stateful lease cannot be reused while the per-slot cache state is being updated.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository UI (base), Organization UI (inherited)

Review profile: ASSERTIVE

Plan: Team

Run ID: fed68932-bcb6-40ae-84d5-93654372e917

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

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.}

[Copilot]

🟡 Changes recommended

It includes an unmentioned stateful-connection lifecycle change (scope/rollout risk) and needs a small follow-up adjustment to improve/verify the new negative-timestamp persistence behavior.

Once you've addressed the issues Copilot identified, you can request another Copilot review.

Pull request overview

This PR updates hybrid-cache negative availability tracking to store and expire authoritative “missing” observations per backend slot (rather than as a single shared timestamp), while keeping decode compatibility with older on-disk cache entry versions.

Changes:

• Add per-backend-slot negative-availability timestamps to DiskCachedArticle, persist them as a new on-disk format version, and expire missing bits independently per slot.
• Ensure hybrid-cache reads/mutations apply per-slot availability expiration before using entries, and preserve negative timestamps across ingest upserts.
• Remove stateful-session connection reuse (stateful sessions now always retire a pooled backend connection on exit).

File summaries

File	Description
src/session/handlers/stateful.rs	Removes “reusable” stateful disposition and retires pooled connections on stateful session exit.
src/session/handlers/hybrid.rs	Aligns hybrid handoff finalization with stateful disposition changes (no reusable success completion).
src/session/backend.rs	Removes now-unused BackendResponseComplete::stateful_session() helper.
src/cache/hybrid.rs	Expires stale per-slot availability before treating cache entries as fresh; preserves timestamps across upserts.
src/cache/hybrid_codec.rs	Introduces V8 disk format with per-slot timestamps; adds per-slot expiration logic and a new regression test.
src/cache/availability_identity.rs	Adds AvailabilitySlot::index() to support per-slot timestamp indexing.

Review details

• Files reviewed: 6/6 changed files
• Comments generated: 3
• Review effort level: Lite

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.