Scheduler and worker loops survive a database restart

index-scheduler's loop ended when Postgres restarted (the active-groups
query was outside its try), leaving live indexing stopped for ~3 hours
while the process stayed up. Each step's error is now caught and retried
next round; worker:spots and worker:deferred wait 30 s and claim again.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

commands/IndexScheduler.ts

@@ -117,11 +117,18 @@ export default class IndexScheduler extends BaseCommand {
       }
     }
 
-    // Run immediately, then wait a full interval after each run so runs never overlap.
+    // Run immediately, then wait a full interval after each run so runs never overlap. A failed
+    // round (e.g. Postgres restarting) is logged and the next one tries again: an error that escaped
+    // here once ended the loop while the process stayed up, and live indexing stopped for ~3 hours
+    // (LOG.md 2026-10-07 (3)).
     while (!Shutdown.requested) {
-      await schedule()
+      for (const step of [schedule, backfill, split]) {
-      await backfill()
+        try {
-      await split()
+          await step()
+        } catch (err: any) {
+          this.logger.error(`Scheduler round failed, retrying next round: ${err.message}`)
+        }
+      }
       await Shutdown.sleep(indexerConfig.intervalMs)
     }
     await this.pool.shutdown()

commands/workers/Deferred.ts

@@ -31,6 +31,7 @@ export default class DeferredWorker extends BaseCommand {
     let done = 0
     try {
       while (!Shutdown.requested && (!this.limit || done < this.limit)) {
+        try {
           const batch = await DeferredService.claim(Math.min(concurrency * 4, this.limit ? this.limit - done : Infinity))
           if (batch.length === 0) {
             if (this.limit) break
@@ -48,6 +49,11 @@ export default class DeferredWorker extends BaseCommand {
           )
           done += batch.length
           this.logger.info(`${done} deferred files processed: ${Object.entries(counts).map(([k, v]) => `${v} ${k}`).join(', ')}.`)
+        } catch (err: any) {
+          // E.g. Postgres restarting: wait and claim again, instead of ending the loop (LOG.md 2026-10-07 (3)).
+          this.logger.error(`Round failed, retrying in 30 s: ${err.message}`)
+          await Shutdown.sleep(30_000)
+        }
       }
     } finally {
       await NntpService.shutdown()

commands/workers/Spots.ts

@@ -33,6 +33,7 @@ export default class SpotsWorker extends BaseCommand {
     let done = 0
     try {
       while (!Shutdown.requested && (!this.limit || done < this.limit)) {
+        try {
           const batch = await SpotService.claim(Math.min(BATCH, this.limit ? this.limit - done : Infinity))
           if (batch.length === 0) {
             if (this.limit) break
@@ -55,6 +56,11 @@ export default class SpotsWorker extends BaseCommand {
           )
           done += batch.length
           this.logger.info(`${done} spots: ${Object.entries(counts).map(([k, v]) => `${v} ${k}`).join(', ')}.`)
+        } catch (err: any) {
+          // E.g. Postgres restarting: wait and claim again, instead of ending the loop (LOG.md 2026-10-07 (3)).
+          this.logger.error(`Round failed, retrying in 30 s: ${err.message}`)
+          await Shutdown.sleep(30_000)
+        }
       }
     } finally {
       await NntpService.shutdown()