Přidal jsem Health Monitoring do portálu:

Přehled má nové karty: Zdravé služby, Nezdravé služby, Nedostupné služby.
Přehled zobrazuje sekci Služby vyžadující pozornost pro unhealthy a unreachable.
Snapshot /portal/ws/operations teď obsahuje health data a stránka je aktualizuje přes existující realtime vrstvu.
Seznam služeb zobrazuje aktuální health status a poslední kontrolu, včetně řazení Podle zdraví.
Detail služby má sekce Zdraví služby a Historie kontrol s posledními 50 záznamy.
Přidal jsem audit event service.health.view.
Browser title teď používá formát CSBot Services Portal - ..., takže detail služby odpovídá požadavku.
Nepoužil jsem ORM, vše je přes SQLite dotazy.

app/db/health.py

@@ -0,0 +1,102 @@
+from app.db.database import get_connection
+
+
+HEALTH_STATUSES = ("healthy", "unhealthy", "unreachable")
+
+
+def get_latest_service_health():
+    con = get_connection()
+
+    try:
+        rows = con.execute(
+            """
+            SELECT
+                h.service_id,
+                COALESCE(a.name, h.service_id) AS service_name,
+                h.status,
+                h.http_status,
+                h.response_time_ms,
+                h.error_text,
+                h.checked_at
+            FROM service_health h
+            LEFT JOIN apps a ON a.id = h.service_id
+            INNER JOIN (
+                SELECT service_id, MAX(id) AS max_id
+                FROM service_health
+                GROUP BY service_id
+            ) latest ON latest.max_id = h.id
+            """
+        ).fetchall()
+    except Exception:
+        rows = []
+
+    con.close()
+    return {row["service_id"]: dict(row) for row in rows}
+
+
+def get_service_health(service_id: str):
+    con = get_connection()
+
+    try:
+        row = con.execute(
+            """
+            SELECT
+                h.service_id,
+                COALESCE(a.name, h.service_id) AS service_name,
+                h.status,
+                h.http_status,
+                h.response_time_ms,
+                h.error_text,
+                h.checked_at
+            FROM service_health h
+            LEFT JOIN apps a ON a.id = h.service_id
+            WHERE h.service_id = ?
+            ORDER BY h.id DESC
+            LIMIT 1
+            """,
+            (service_id,),
+        ).fetchone()
+    except Exception:
+        row = None
+
+    con.close()
+    return dict(row) if row else None
+
+
+def get_service_health_history(service_id: str, limit: int = 50):
+    con = get_connection()
+
+    try:
+        rows = con.execute(
+            """
+            SELECT id, service_id, status, http_status, response_time_ms, error_text, checked_at
+            FROM service_health
+            WHERE service_id = ?
+            ORDER BY id DESC
+            LIMIT ?
+            """,
+            (service_id, limit),
+        ).fetchall()
+    except Exception:
+        rows = []
+
+    con.close()
+    return [dict(row) for row in rows]
+
+
+def get_health_summary():
+    latest = get_latest_service_health()
+    summary = {"healthy": 0, "unhealthy": 0, "unreachable": 0}
+    for item in latest.values():
+        status = item.get("status")
+        if status in summary:
+            summary[status] += 1
+    return summary
+
+
+def get_problem_service_health():
+    latest = get_latest_service_health()
+    return [
+        item for item in latest.values()
+        if item.get("status") in {"unhealthy", "unreachable"}
+    ]

app/db/operations.py

@@ -1,4 +1,5 @@
 from app.db.database import get_connection
+from app.db.health import get_health_summary, get_problem_service_health
 from app.db.migrations import run_migrations
 from app.db.workers import is_worker_online
 
@@ -197,6 +198,8 @@ def get_operations_snapshot():
         WHERE LOWER(COALESCE(status, '')) IN ('failed', 'error', 'disabled', 'deleted', 'archived')
         """
     ).fetchone()["count"]
+    health_summary = get_health_summary()
+    health_problems = get_problem_service_health()
 
     jobs_summary = con.execute(
         """
@@ -296,6 +299,12 @@ def get_operations_snapshot():
             "active": active_applications,
             "problematic": problematic_applications,
         },
+        "health": {
+            "healthy": health_summary["healthy"],
+            "unhealthy": health_summary["unhealthy"],
+            "unreachable": health_summary["unreachable"],
+            "problems": health_problems,
+        },
         "workers": {
             "online": workers_online,
             "offline": max(0, total_workers - workers_online),