From b9ee10a83a69a7224accfbf74f1513ed688f4166 Mon Sep 17 00:00:00 2001
From: ImBenji <benjamin.watt@imbenji.net>
Date: Tue, 4 Aug 2026 21:07:42 +0100
Subject: [PATCH] fix: serialize autonomy job leases

---
 src/autonomy/jobs.js             | 10 +++++++++-
 workers/autonomyWorker.js        |  1 +
 workers/calibrationWorker.js     |  1 +
 workers/coordinatorWorker.js     |  1 +
 workers/executionWorker.js       |  1 +
 workers/outcomeAutonomyWorker.js |  1 +
 6 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/src/autonomy/jobs.js b/src/autonomy/jobs.js
index 028d619..8a5e8c4 100644
--- a/src/autonomy/jobs.js
+++ b/src/autonomy/jobs.js
@@ -31,7 +31,15 @@ function leaseNextJob(db, workerId, leaseSeconds = 60, jobTypes = null) {
     `).run(workerId, `+${Math.max(1, Math.floor(leaseSeconds))} seconds`, job.id);
     return updated.changes ? { ...job, status: 'leased', leased_by: workerId } : null;
   });
-  return tx();
+  try {
+    // Acquire the write reservation before selecting. A deferred transaction can
+    // otherwise read a snapshot, lose the writer race, and fail with
+    // SQLITE_BUSY_SNAPSHOT when it attempts the lease update.
+    return tx.immediate();
+  } catch (error) {
+    if (String(error.code || '').startsWith('SQLITE_BUSY')) return null;
+    throw error;
+  }
 }
 
 function completeJob(db, id, workerId) {
diff --git a/workers/autonomyWorker.js b/workers/autonomyWorker.js
index 88509a4..e4a0b0a 100644
--- a/workers/autonomyWorker.js
+++ b/workers/autonomyWorker.js
@@ -71,6 +71,7 @@ async function runAutonomyWorker({ archivePath, intelligencePath, workerId = `au
   const archiveDb = new Database(archivePath, { readonly: true });
   const intelligenceDb = new Database(intelligencePath);
   intelligenceDb.pragma('journal_mode = WAL');
+  intelligenceDb.pragma('busy_timeout = 5000');
   initAutonomySchema(intelligenceDb);
 
   while (true) {
diff --git a/workers/calibrationWorker.js b/workers/calibrationWorker.js
index 220fc5d..e293428 100644
--- a/workers/calibrationWorker.js
+++ b/workers/calibrationWorker.js
@@ -71,6 +71,7 @@ function createDecisions(db, strategyVersion = 'autonomy-1') {
 async function runCalibrationWorker({ intelligencePath, pollMs = 60000, workerId = `calibration-${os.hostname()}-${process.pid}` } = {}) {
   const db = new Database(intelligencePath);
   db.pragma('journal_mode = WAL');
+  db.pragma('busy_timeout = 5000');
   initAutonomySchema(db);
   while (true) {
     try {
diff --git a/workers/coordinatorWorker.js b/workers/coordinatorWorker.js
index 74de840..5737a6e 100644
--- a/workers/coordinatorWorker.js
+++ b/workers/coordinatorWorker.js
@@ -34,6 +34,7 @@ async function runCoordinatorWorker({ archivePath, intelligencePath, workerId =
   const archiveDb = new Database(archivePath, { readonly: true });
   const intelligenceDb = new Database(intelligencePath);
   intelligenceDb.pragma('journal_mode = WAL');
+  intelligenceDb.pragma('busy_timeout = 5000');
   initAutonomySchema(intelligenceDb);
   const config = loadConfig();
   while (true) {
diff --git a/workers/executionWorker.js b/workers/executionWorker.js
index 2e677dd..539b71c 100644
--- a/workers/executionWorker.js
+++ b/workers/executionWorker.js
@@ -10,6 +10,7 @@ async function runExecutionWorker({ intelligencePath, pollMs = 10000, mode = 'sh
   if (!['shadow', 'paper'].includes(mode)) throw new Error(`unsupported execution mode: ${mode}`);
   const db = new Database(intelligencePath);
   db.pragma('journal_mode = WAL');
+  db.pragma('busy_timeout = 5000');
   initAutonomySchema(db);
   const paperClient = mode === 'paper'
     ? createAlpacaPaperClient({ keyId: process.env.ALPACA_PAPER_KEY_ID, secretKey: process.env.ALPACA_PAPER_SECRET_KEY })
diff --git a/workers/outcomeAutonomyWorker.js b/workers/outcomeAutonomyWorker.js
index deb0737..81727b4 100644
--- a/workers/outcomeAutonomyWorker.js
+++ b/workers/outcomeAutonomyWorker.js
@@ -33,6 +33,7 @@ async function history(symbol) {
 async function resolveAutonomyOutcomes({ intelligencePath, workerId = `outcome-${os.hostname()}-${process.pid}`, pollMs = 60000 } = {}) {
   const db = new Database(intelligencePath);
   db.pragma('journal_mode = WAL');
+  db.pragma('busy_timeout = 5000');
   initAutonomySchema(db);
   const cache = new Map();
   while (true) {