test: [YW-266] stabilize native-DB-concurrency test flakes (WsManager + mapping-store)

[youhaowei]

Problem

Two native-DB (PGlite/WASM) test races ate CI re-runs across unrelated PRs this cycle (e.g. #138 hit the mapping-store flake despite touching zero server-core files). They are a class, not one test:

1. @wystack/client WsManager > does not reconnect on close code 4001 — original report.
2. @wystack/server-core mapping-store.test.ts > DrizzleMappingStore > get returns the record written by set — flaked on a PR touching zero server-core files; 53s file duration = native-DB timing contention.

Both are test-setup races, fixed at the mechanism level — no test.retry(), no bare timeout bumps to mask the race.

Fixes

server-core — vitest.config.ts (this repo)

vitest 4.x runs all five server-core test files in parallel forks. Every file spins up PGlite (WASM Postgres), so they cold-start the WASM module simultaneously and contend for FDs/memory on a loaded runner. The first test in each file runs before any warm-up query, so it's the victim. Serialize the suite (maxWorkers: 1 / minWorkers: 1) so files cold-start one at a time. Tests are I/O-bound — small wall-clock cost.

client / WsManager — wystack submodule bump

ws.test.ts leaked ~12 PGlite WASM instances (sync afterEach never closed them), squeezing the 2500ms-sleep no-reconnect tests against the 5s bun timeout. The submodule commit adds a tracked-handle registry drained in an async afterEach, plus timeout headroom. Detail in youhaowei/wystack#46.

Verification

De-flake proven by looped local runs (a single green run does not prove a flake fixed):

• mapping-store.test.ts: 20/20 pass, 0 fail
• ws.test.ts (submodule): 20/20 pass, 0 fail
• Full server-core suite (maxWorkers:1): 70 passed, 1 todo
• typecheck + lint clean (DashFrame + wystack)

Submodule sequencing

The libs/wystack pointer references the YW-266 branch commit so CI runs green against the fix now. Re-point to the merged wystack main commit before merging this PR (depends on youhaowei/wystack#46).

Tracked internally as YW-266.

🤖 Generated with Claude Code

Greptile Summary

This PR stabilizes two recurring CI flakes caused by PGlite/WASM test-setup races — no retries or timeout masking. Both fixes target the mechanism: serializing the server-core vitest worker pool so five PGlite files no longer cold-start simultaneously, and bumping the wystack submodule to pick up an async afterEach that properly drains leaked WASM handles in ws.test.ts.

• packages/server-core/vitest.config.ts: adds maxWorkers: 1 / minWorkers: 1 with a detailed comment; forces sequential file execution and eliminates the FD/memory contention that caused the first test in each file to be the flake victim.
• libs/wystack: advances the submodule pointer to the wystack YW-266 branch commit (4bed431) which adds a tracked-handle registry drained via async afterEach; the PR description explicitly notes this must be re-pointed to the merged main SHA before merging.

Confidence Score: 5/5

Safe to merge once the wystack submodule is re-pointed to the merged main commit as documented.

Both changes are targeted and low-risk: the vitest config serializes existing test files (no production code touched), and the submodule bump only affects test infrastructure. The pre-merge gate (re-pointing the submodule to wystack main after wystack#46 merges) is clearly documented in the PR description and acknowledged by the author.

No files require special attention beyond the documented pre-merge step of re-pointing libs/wystack to the merged wystack main SHA.

Important Files Changed

Filename	Overview
packages/server-core/vitest.config.ts	Adds maxWorkers: 1 / minWorkers: 1 to serialize PGlite-based test files and eliminate WASM cold-start contention; well-commented and correct.
libs/wystack	Submodule pointer bumped from 6b5f7c1 to 4bed431 (wystack YW-266 branch) to pick up the async afterEach PGlite handle drain fix; intentionally on a pre-merge branch SHA per PR description.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[CI: server-core test run] --> B{Before fix\nmaxWorkers default}
    B --> C[Fork 1: mapping-store.test.ts\nPGlite WASM cold-start]
    B --> D[Fork 2: other.test.ts\nPGlite WASM cold-start]
    B --> E[Fork 3: ...\nPGlite WASM cold-start]
    C & D & E --> F[Contend for FDs/memory\non loaded CI runner]
    F --> G[First test in each file\nflakes / times out]
    A2[CI: server-core test run] --> B2{After fix\nmaxWorkers: 1}
    B2 --> H[Fork 1: mapping-store.test.ts\nPGlite WASM cold-start]
    H --> I[Fork 1 done]
    I --> J[Fork 2: other.test.ts\nPGlite WASM cold-start]
    J --> K[... sequential ...]
    K --> L[All tests pass]
    style G fill:#f88,stroke:#c00
    style L fill:#8f8,stroke:#0c0

Loading

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
flowchart TD
    A[CI: server-core test run] --> B{Before fix\nmaxWorkers default}
    B --> C[Fork 1: mapping-store.test.ts\nPGlite WASM cold-start]
    B --> D[Fork 2: other.test.ts\nPGlite WASM cold-start]
    B --> E[Fork 3: ...\nPGlite WASM cold-start]
    C & D & E --> F[Contend for FDs/memory\non loaded CI runner]
    F --> G[First test in each file\nflakes / times out]
    A2[CI: server-core test run] --> B2{After fix\nmaxWorkers: 1}
    B2 --> H[Fork 1: mapping-store.test.ts\nPGlite WASM cold-start]
    H --> I[Fork 1 done]
    I --> J[Fork 2: other.test.ts\nPGlite WASM cold-start]
    J --> K[... sequential ...]
    K --> L[All tests pass]
    style G fill:#f88,stroke:#c00
    style L fill:#8f8,stroke:#0c0

Loading

_{Reviews (3): Last reviewed commit: "chore: re-point libs/wystack to main (YW..." | Re-trigger Greptile}

[linear-code]

YW-266

[coderabbitai]

📝 Walkthrough

Walkthrough

Updates the libs/wystack git submodule to a new commit and modifies packages/server-core/vitest.config.ts to force test execution to a single worker (maxWorkers: 1, minWorkers: 1) with comments noting the constraint is for PGlite (WASM) integration tests on CI.

Changes

CI and Test Configuration

Layer / File(s)	Summary
Vitest single-worker constraint packages/server-core/vitest.config.ts	Adds maxWorkers: 1 and minWorkers: 1 with explanatory comments to serialize PGlite-bound integration test execution and avoid parallel cold-start contention on CI.
wystack submodule bump libs/wystack	Advances the submodule pointer from one commit SHA to another.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

🚥 Pre-merge checks | ✅ 4

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately summarizes the main change: stabilizing native-DB-concurrency test flakes in WsManager and mapping-store via Vitest config and submodule updates.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description check	✅ Passed	The PR description is comprehensive and well-structured, covering problem statement, fixes, verification details, and important pre-merge requirements.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

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 and usage tips.}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: c3854eb69b

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".