fix(server): stop fsyncing SQLite on every commit so slow disks no longer drop clients

[t3dotgg]

Problem

Users keep seeing "reconnecting" while agents run. SQLite runs at the default synchronous = FULL, so every commit fsyncs the WAL on the main thread, and every provider event is its own commit. On a slow or busy disk (HDD, LUKS/btrfs, ext4 under build load, WSL with discard), one fsync takes 100 ms to several seconds. The event loop stops, heartbeats miss the timeout, and clients drop. The data is in #15707.

Change

Set PRAGMA synchronous = NORMAL with the other connection pragmas. In WAL mode this fsyncs at checkpoints, not on every commit.

Tradeoff: an app crash still loses nothing, and the database cannot corrupt. An OS crash or power loss can roll back commits made since the last checkpoint. That is the standard WAL setting, and macOS FULL already does not use F_FULLFSYNC (#13544).

Scope and approval

Maintainer change. It is the first option in the #15707 triage, and users have measured it on live servers. It does not move SQLite off the main thread (#18736, #14703), so slow reads (#14701) still stall.

Verification

• User A/B on a live server in [Bug]: Orchestrator V2 event writes stall the server on a high-latency disk (156 ms fsync), environment flips Offline #15707: fsync time went from 17.6 s to 0.4 s per minute, the slowest single fsync from 6.7 s to 0.25 s, and the reconnects stopped. On the slow disk from the original report, median commit time went from 41.8 ms to 0.05 ms.
• Local real-disk benchmark (ext4), 500 commits shaped like event writes: fsyncs went from 514 to 11, and median commit time from 1.85 ms to 0.60 ms.
• New test in Sqlite.test.ts: it checks that a file-backed database opens at NORMAL. It fails without the change (expected 2 to equal 1). Sqlite.test.ts passes 5/5. Lint and format pass.
• GPT-6.1 Sol reviewed the change and approved it. It confirmed that every production connection to statev2.sqlite goes through this setup.

Created with Claude Opus 5.5 in Claude Code, running in T3 Code. Reviewed with sol-loop: 1 round with GPT-6.1 Sol.

🤖 Generated with Claude Code

Closes #15707

[macroscopeapp]

Approvability

Verdict: Not approved

Macroscope's review found this PR not approvable — This changes SQLite's automatic durability setting for all server persistence connections, trading per-commit fsync latency for possible loss of commits since the last checkpoint after an OS or power failure. The production default change and its persistence tradeoff warrant human review.

^{You can add or adjust custom eligibility rules. Learn more.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	5.0 KiB	5.0 KiB	0 B (0.0%)	6.8 KiB	✅
Codex	Thread snapshot wire	3.8 KiB	3.8 KiB	0 B (0.0%)	4.9 KiB	✅
Codex	Live turn WebSocket wire	1.2 KiB	1.2 KiB	0 B (0.0%)	2.0 KiB	✅
Codex	Live turn WebSocket decoded	20.9 KiB	20.9 KiB	0 B (0.0%)	29.3 KiB	✅
Codex	Live turn messages	2	2	0 (0.0%)	8	✅
Claude	Total thread wire	5.0 KiB	5.0 KiB	−24 B (−0.5%)	6.8 KiB	✅
Claude	Thread snapshot wire	3.8 KiB	3.8 KiB	0 B (0.0%)	4.9 KiB	✅
Claude	Live turn WebSocket wire	1.2 KiB	1.2 KiB	−24 B (−2.0%)	2.0 KiB	✅
Claude	Live turn WebSocket decoded	21.2 KiB	21.2 KiB	−41 B (−0.2%)	29.3 KiB	✅
Claude	Live turn messages	2	1	−1 (−50.0%)	8	✅

Baseline: aef0e9c · PR result: d72c261 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 108.5 KiB
• Claude decoded thread snapshot: 108.8 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

• Configuration used: Path: .coderabbit.config.ts
• Review profile: CHILL
• Plan: Team
• Run ID: 8b1682fb-1446-4f82-9388-3423f44e5d77

📥 Commits

Reviewing files that changed from the base of the PR and between aef0e9c and d72c261.

📒 Files selected for processing: 2

• apps/server/src/persistence/Sqlite.test.ts
• apps/server/src/persistence/Sqlite.ts

Included review availability: This review used your included allowance. 8 included reviews remain after this review. Your included PR review attempts over the past 7 days set your current allowance at 10 reviews per hour.

📝 Walkthrough

SQLite connections now use synchronous = NORMAL. A persistence test opens a temporary file-backed database and verifies that SQLite reports the setting as 1.

Changes: SQLite persistence

Layer / File(s)	Summary
Set and verify synchronous mode apps/server/src/persistence/Sqlite.ts, apps/server/src/persistence/Sqlite.test.ts	The shared SQLite setup sets synchronous = NORMAL. A test verifies the setting on a temporary file-backed database. Comments describe the WAL durability tradeoff.

Priority: ⬆️ High

Estimated code review effort: 2 (Simple) | ~10 minutes

Change: Bug fix

Suggested reviewers: otavio

Merge Risk: ⚪ Minimal · up to d72c2

This change trades some power-loss durability for lower commit latency, as intended. No merge-blocking issue is established by the supplied evidence.

Pre-merge checks: 4

Passed checks: 4 passed

Check name	Status	Explanation
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.
Title check		The title clearly and concisely describes the main change: changing SQLite commit durability behavior to prevent slow-disk client disconnects.
Description check		The description includes the required Problem, Change, Scope and approval, and Verification sections. It explains the durability tradeoff, approval context, test coverage, benchmark results, and limit…

✨ Finishing Touches

📝 Generate docstrings

• Commit to this branch
• Create a new PR

🧪 Generate unit tests (beta)

• Commit to this branch
• Create a new PR

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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