From 058fa8560674e39e25cf7dd461753877e433c2df Mon Sep 17 00:00:00 2001
From: Bryan Van Deusen <bryan.vandeusen@fabledsword.com>
Date: Thu, 24 Sep 2026 19:26:48 -0400
Subject: [PATCH] fix: backfill_phash runs on the long maintenance lane, not
 the scheduler's quick one (4411)

A whole-library rehash with a 35-minute limit was matched by the
maintenance.* glob. It held a scheduler process for its whole run, and the
minute ticks queued behind it. An exact-name route now sends it to
maintenance_long.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LVjrnpQjRgHdvq95rASoiR
---
 backend/app/celery_app.py    |  7 +++++++
 tests/test_celery_routing.py | 14 ++++++++++++++
 2 files changed, 21 insertions(+)

diff --git a/backend/app/celery_app.py b/backend/app/celery_app.py
index 5f45567..bd0ad2c 100644
--- a/backend/app/celery_app.py
+++ b/backend/app/celery_app.py
@@ -62,6 +62,13 @@ def make_celery() -> Celery:
             # can never starve the quick self-healing sweeps (operator-flagged
             # 2026-06-07: a 2h audit blocked vacuum/backup/normalize for hours).
             "backend.app.tasks.maintenance.*": {"queue": "maintenance"},
+            # The one long job in maintenance.py: a whole-library phash
+            # recompute (35 min hard limit; the library was cleared for
+            # re-hashing by migration 0098). On the quick lane it held a
+            # scheduler process for its whole run, and the minute ticks queued
+            # up behind it (2026-09-24: 7 waiting, "all workers busy for 18
+            # minutes"). An exact name wins over the glob above.
+            "backend.app.tasks.maintenance.backfill_phash": {"queue": "maintenance_long"},
             "backend.app.tasks.backup.*": {"queue": "maintenance_long"},
             "backend.app.tasks.admin.*": {"queue": "maintenance_long"},
             "backend.app.tasks.library_audit.*": {"queue": "maintenance_long"},
diff --git a/tests/test_celery_routing.py b/tests/test_celery_routing.py
index 8a7cb67..e7a24e6 100644
--- a/tests/test_celery_routing.py
+++ b/tests/test_celery_routing.py
@@ -36,3 +36,17 @@ def test_queue_for_mirrors_external_to_download():
         celery.conf.task_routes["backend.app.tasks.external.*"]["queue"]
         == "download"
     )
+
+
+def test_backfill_phash_runs_on_the_long_lane():
+    """It lives in maintenance.py, so the quick-lane glob matches it too —
+    the router must pick the exact name. A 35-minute rehash on the scheduler
+    lane blocked the minute ticks behind it (2026-09-24)."""
+    route = celery.amqp.router.route(
+        {}, "backend.app.tasks.maintenance.backfill_phash",
+    )
+    assert route["queue"].name == "maintenance_long"
+    quick = celery.amqp.router.route(
+        {}, "backend.app.tasks.maintenance.recover_stalled_task_runs",
+    )
+    assert quick["queue"].name == "maintenance"