Implement the full ADR-0006 plan: three-layer separation of transaction
correctness, retry policy, and cross-process writer coordination.
Layer 1 — scripts/tx.ts (transaction correctness):
- runWriteTransaction executes work exactly once; no internal retry.
- BEGIN IMMEDIATE takes the write lock up front (avoids SQLITE_BUSY_SNAPSHOT).
- Guarded rollback: checks inTransaction() via adapter before attempting
ROLLBACK; never masks the primary exception.
- WriteTxDiagnostics attached to errors: phase, code, label,
rollbackSucceeded, rollbackError, transactionActive.
- Binding adapters (betterSqliteTransactionAdapter, nodeSqliteTransactionAdapter)
mapping better-sqlite3's `.inTransaction` and node:sqlite's `.isTransaction`.
- configureConnection centralizes WAL + synchronous + busy_timeout.
Layer 2 — scripts/write-coordinator.ts (retry policy):
- runRetryableWriteTransaction: bounded retry with total time budget.
- Only retries when the transaction confirmed ended (transactionActive=false)
and the error is SQLITE_BUSY during work/commit phase.
- BEGIN-phase BUSY = abort entire build (isBeginBusyFailure); the caller
returns `{ deferred: true, reason: 'writer_busy' }` instead of waiting.
- hasUnusableTransaction detects a still-active transaction after failure;
aborts the build immediately, never retries.
Layer 3 — scripts/writer-lease.ts (cross-process coordination):
- acquireWriterLease: dedicated writer.lock.sqlite with busy_timeout=0 +
BEGIN IMMEDIATE. Non-blocking attempt; bounded wait with retryDelayMs.
- writerLockPathFor derives lock path from the target DB path.
- Lease held for the entire build; released on completion or failure.
- Lock DB uses DELETE journal (not WAL); crash/close auto-releases.
- All consumers obey: skill acquires at build start (returns deferred if
unavailable); app daemon (via worker) acquires for its build cycle.
Build semantics changes:
- affectedSessionIds updated only after successful commit.
- BuildIndexResult gains skipped/skippedFiles for observability.
- Skill finalize failure now fails the build (was silently warned).
- Checkpoint changed to PASSIVE (TRUNCATE reserved for maintenance/exit).
- Skill buildIndex returns { deferred, reason } on lease contention;
indexer-service reschedules the build (deferredRetryMs) without publishing
a heartbeat (so the build-deferred state is visible to cross-process
arbitration).
- Service publishes heartbeat immediately on start() for correct arbitration.
Tests:
- tests/write-transaction.test.mjs: single-shot execution, diagnostics
propagation, auto-rolled-back transaction detected, rollback failure
captured as metadata, BEGIN IMMEDIATE semantics.
- tests/writer-lease.test.mjs: acquire/release, contention returns null,
bounded wait with release during budget.
- tests/app-writer-lease.test.mjs: better-sqlite3 adapter integration.
- tests/app-rollback-guard.test.mjs: rewritten — transient BUSY recovered
by coordinator, persistent BUSY skips file, begin-busy aborts build,
live-transaction aborts build, phantom affectedSessionIds prevented.
- tests/daemon-arbitration.test.mjs: skill defers to fresh app heartbeat,
builds when heartbeat is stale.
- tests/app-indexer-service.test.mjs: new cases for deferred-retry
scheduling and immediate heartbeat on start.
- app/tests/electron-concurrency.mjs + child: dual-child IPC structure for
real better-sqlite3 contention (holder acquires lock → build child starts
→ delayed release → result collected; persistent contention bounded).
ADR-0006 updated to reflect the implemented design.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
167 lines
5.2 KiB
JavaScript
167 lines
5.2 KiB
JavaScript
import { test } from 'node:test';
|
|
import assert from 'node:assert/strict';
|
|
import { createRequire } from 'node:module';
|
|
import { mkdtempSync } from 'node:fs';
|
|
import { tmpdir } from 'node:os';
|
|
import { join } from 'node:path';
|
|
|
|
import { nodeSqliteTransactionAdapter, runWriteTransaction } from '../scripts/tx.ts';
|
|
import { hasUnusableTransaction, isBeginBusyFailure, isRetryableWriteFailure } from '../scripts/write-coordinator.ts';
|
|
|
|
const require = createRequire(import.meta.url);
|
|
const { DatabaseSync } = require('node:sqlite');
|
|
|
|
test('a failed rollback with an active transaction never retries or masks the primary error', () => {
|
|
const primary = Object.assign(new Error('database is locked'), { code: 'SQLITE_BUSY' });
|
|
let active = false;
|
|
let workCalls = 0;
|
|
|
|
const db = {
|
|
exec(sql) {
|
|
if (sql.startsWith('BEGIN')) {
|
|
if (active) throw new Error('cannot start a transaction within a transaction');
|
|
active = true;
|
|
} else if (sql === 'ROLLBACK') {
|
|
throw new Error('rollback failed with I/O error');
|
|
} else if (sql === 'COMMIT') {
|
|
active = false;
|
|
}
|
|
},
|
|
inTransaction() {
|
|
return active;
|
|
},
|
|
};
|
|
|
|
assert.throws(
|
|
() => runWriteTransaction(db, () => {
|
|
workCalls += 1;
|
|
throw primary;
|
|
}),
|
|
error => error === primary,
|
|
);
|
|
assert.equal(workCalls, 1);
|
|
assert.equal(active, true);
|
|
});
|
|
|
|
test('an automatic rollback rethrows the primary error without issuing another rollback', () => {
|
|
const primary = Object.assign(new Error('database is locked'), { code: 'SQLITE_BUSY' });
|
|
let active = false;
|
|
let rollbackCalls = 0;
|
|
const db = {
|
|
exec(sql) {
|
|
if (sql.startsWith('BEGIN')) active = true;
|
|
if (sql === 'ROLLBACK') rollbackCalls += 1;
|
|
},
|
|
inTransaction() {
|
|
return active;
|
|
},
|
|
};
|
|
|
|
assert.throws(
|
|
() => runWriteTransaction(db, () => {
|
|
active = false;
|
|
throw primary;
|
|
}),
|
|
error => error === primary,
|
|
);
|
|
assert.equal(rollbackCalls, 0);
|
|
assert.equal(primary.obelisk.transactionActive, false);
|
|
});
|
|
|
|
test('an active transaction is rolled back once before the primary error is rethrown', () => {
|
|
const primary = new Error('persist failed');
|
|
let active = false;
|
|
let rollbackCalls = 0;
|
|
const db = {
|
|
exec(sql) {
|
|
if (sql.startsWith('BEGIN')) active = true;
|
|
if (sql === 'ROLLBACK') {
|
|
rollbackCalls += 1;
|
|
active = false;
|
|
}
|
|
},
|
|
inTransaction() {
|
|
return active;
|
|
},
|
|
};
|
|
|
|
assert.throws(() => runWriteTransaction(db, () => { throw primary; }), error => error === primary);
|
|
assert.equal(rollbackCalls, 1);
|
|
assert.equal(primary.obelisk.rollbackSucceeded, true);
|
|
assert.equal(primary.obelisk.transactionActive, false);
|
|
});
|
|
|
|
test('an unknown post-error transaction state is unsafe for the next file', () => {
|
|
const primary = new Error('transaction state unavailable');
|
|
const db = {
|
|
exec() {},
|
|
inTransaction() {
|
|
throw new Error('binding cannot report transaction state');
|
|
},
|
|
};
|
|
|
|
assert.throws(() => runWriteTransaction(db, () => { throw primary; }), error => error === primary);
|
|
assert.equal(primary.obelisk.transactionActive, null);
|
|
assert.equal(hasUnusableTransaction(primary), true);
|
|
});
|
|
|
|
test('node:sqlite generic error codes preserve BUSY classification from the message', () => {
|
|
const primary = Object.assign(new Error('database is locked'), { code: 'ERR_SQLITE_ERROR' });
|
|
let active = false;
|
|
const db = {
|
|
exec(sql) {
|
|
if (sql === 'BEGIN IMMEDIATE') active = true;
|
|
},
|
|
inTransaction() { return active; },
|
|
};
|
|
|
|
assert.throws(() => runWriteTransaction(db, () => {
|
|
active = false;
|
|
throw primary;
|
|
}), error => error === primary);
|
|
assert.equal(primary.obelisk.code, 'SQLITE_BUSY');
|
|
assert.equal(isRetryableWriteFailure(primary), true);
|
|
});
|
|
|
|
test('a real node:sqlite BEGIN lock is classified as a deferrable BUSY', () => {
|
|
const dbPath = join(mkdtempSync(join(tmpdir(), 'obelisk-node-sqlite-busy-')), 'index.sqlite');
|
|
const holder = new DatabaseSync(dbPath);
|
|
const contender = new DatabaseSync(dbPath);
|
|
holder.exec('PRAGMA busy_timeout=0; CREATE TABLE test (value TEXT); BEGIN IMMEDIATE');
|
|
contender.exec('PRAGMA busy_timeout=0');
|
|
|
|
try {
|
|
assert.throws(
|
|
() => runWriteTransaction(nodeSqliteTransactionAdapter(contender), () => {}),
|
|
error => isBeginBusyFailure(error) && error.obelisk.code === 'SQLITE_BUSY',
|
|
);
|
|
} finally {
|
|
holder.exec('ROLLBACK');
|
|
holder.close();
|
|
contender.close();
|
|
}
|
|
});
|
|
|
|
test('a BEGIN failure with an active transaction is not deferrable', () => {
|
|
const primary = Object.assign(new Error('database is locked'), { code: 'SQLITE_BUSY' });
|
|
let rollbackCalls = 0;
|
|
const db = {
|
|
exec(sql) {
|
|
if (sql === 'BEGIN IMMEDIATE') throw primary;
|
|
if (sql === 'ROLLBACK') {
|
|
rollbackCalls += 1;
|
|
throw new Error('rollback failed with I/O error');
|
|
}
|
|
},
|
|
inTransaction() {
|
|
return true;
|
|
},
|
|
};
|
|
|
|
assert.throws(() => runWriteTransaction(db, () => {}), error => error === primary);
|
|
assert.equal(rollbackCalls, 1);
|
|
assert.equal(primary.obelisk.transactionActive, true);
|
|
assert.equal(hasUnusableTransaction(primary), true);
|
|
assert.equal(isBeginBusyFailure(primary), false);
|
|
});
|