feat(observability): add Prometheus/Grafana/Loki/Alertmanager/Alloy stack and remove legacy Monitoring UI

backend/src/media_library_viewer_api/routers/backups.py

@@ -8,6 +8,7 @@ from ..models.backups import (
     BackupReportRequest,
     BackupRunResponse,
 )
+from ..observability import record_backup_run
 from ..services.backup_alert_engine import generate_alerts_for_run
 from ..services.settings_store import SettingsStore, get_settings_store
 
@@ -17,20 +18,24 @@ router = APIRouter(prefix="/api/backups", tags=["backups"])
 def _get_or_create_job(store: SettingsStore, report: BackupReportRequest) -> dict[str, Any]:
     job = store.get_backup_job_by_name(report.name)
     if not job:
-        job = store.upsert_backup_job({
-            "name": report.name,
-            "source": report.source,
-            "target": report.target,
-            "schedule_interval_seconds": report.schedule_interval_seconds,
-        })
+        job = store.upsert_backup_job(
+            {
+                "name": report.name,
+                "source": report.source,
+                "target": report.target,
+                "schedule_interval_seconds": report.schedule_interval_seconds,
+            }
+        )
     elif report.schedule_interval_seconds:
-        store.upsert_backup_job({
-            "id": job["id"],
-            "name": report.name,
-            "source": report.source,
-            "target": report.target,
-            "schedule_interval_seconds": report.schedule_interval_seconds,
-        })
+        store.upsert_backup_job(
+            {
+                "id": job["id"],
+                "name": report.name,
+                "source": report.source,
+                "target": report.target,
+                "schedule_interval_seconds": report.schedule_interval_seconds,
+            }
+        )
         job = store.get_backup_job(job["id"])
     return job
 
@@ -61,6 +66,7 @@ def post_backup_report(
         "details": report.details,
     }
     run = store.create_backup_run(run_data)
+    record_backup_run(job_name=report.name, status=report.status, success=report.status == "success")
 
     # Generate alerts
     previous_runs = store.list_backup_runs(job_id=job["id"], status="success", limit=20)
@@ -93,6 +99,7 @@ def post_backup_start(
         "status": "in_progress",
     }
     run = store.create_backup_run(run_data)
+    record_backup_run(job_name=report.name, status="in_progress")
     # Map details -> details_json for response model
     run["details_json"] = run.pop("details", None)
     return BackupRunResponse(**run)