feat(fc2a): add maintenance tasks (recovery + cleanup) with beat schedule

recover_interrupted_tasks runs every 5 minutes, finds ImportTask rows stuck in 'processing' for >30 minutes (well above any legitimate import duration), and re-queues them. cleanup_old_tasks runs daily and deletes finished tasks older than 7 days so the task table stays an operational view rather than an archive. Both thresholds match ImageRepo's precedent. The 30-min stuck threshold is documented inline so a future reader can adjust it intentionally rather than mistaking it for a 'magic number'. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-14 12:09:09 -04:00
parent d3bb8c509a
commit 509c19ce86
3 changed files with 162 additions and 0 deletions
@@ -0,0 +1,77 @@
+"""Unit tests for the maintenance tasks. Eager-mode Celery so the task
+function runs synchronously inside the same DB transaction as the test.
+"""
+
+from datetime import datetime, timedelta, timezone
+
+import pytest
+
+from backend.app.celery_app import celery
+from backend.app.models import ImportBatch, ImportTask
+
+
+@pytest.fixture(autouse=True)
+def eager():
+    celery.conf.task_always_eager = True
+    yield
+    celery.conf.task_always_eager = False
+
+
+def _make_batch(session) -> int:
+    batch = ImportBatch(triggered_by="manual", source_path="/import", scan_mode="quick")
+    session.add(batch)
+    session.flush()
+    return batch.id
+
+
+def test_recover_interrupted_only_old(db_sync):
+    batch_id = _make_batch(db_sync)
+    now = datetime.now(timezone.utc)
+
+    fresh = ImportTask(
+        batch_id=batch_id, source_path="/import/a.jpg", task_type="media",
+        status="processing", started_at=now - timedelta(minutes=5),
+    )
+    stale = ImportTask(
+        batch_id=batch_id, source_path="/import/b.jpg", task_type="media",
+        status="processing", started_at=now - timedelta(hours=2),
+    )
+    db_sync.add_all([fresh, stale])
+    db_sync.commit()
+
+    from backend.app.tasks.maintenance import recover_interrupted_tasks
+    recovered = recover_interrupted_tasks.apply().get()
+    assert recovered == 1
+
+    db_sync.refresh(fresh)
+    db_sync.refresh(stale)
+    assert fresh.status == "processing"
+    assert stale.status == "queued"
+    assert stale.started_at is None
+
+
+def test_cleanup_old_deletes_finished_old(db_sync):
+    batch_id = _make_batch(db_sync)
+    now = datetime.now(timezone.utc)
+
+    old_complete = ImportTask(
+        batch_id=batch_id, source_path="/import/a.jpg", task_type="media",
+        status="complete", finished_at=now - timedelta(days=10),
+    )
+    recent_complete = ImportTask(
+        batch_id=batch_id, source_path="/import/b.jpg", task_type="media",
+        status="complete", finished_at=now - timedelta(days=2),
+    )
+    old_pending = ImportTask(
+        batch_id=batch_id, source_path="/import/c.jpg", task_type="media",
+        status="pending",
+    )
+    db_sync.add_all([old_complete, recent_complete, old_pending])
+    db_sync.commit()
+
+    from backend.app.tasks.maintenance import cleanup_old_tasks
+    deleted = cleanup_old_tasks.apply().get()
+    assert deleted == 1
+
+    remaining = {t.source_path for t in db_sync.query(ImportTask).all()}
+    assert remaining == {"/import/b.jpg", "/import/c.jpg"}