fix(platform): Fix service health check mechanism on app service (#10401)

## Summary
- Introduced correct health check API for AppService
- Fixed service health check mechanism to properly handle health status
monitoring

## Changes Made
- Updated health check implementation in AppService.
- Make rest service health check checks the health of DatabaseManager
too.

## Test plan
- [x] Verify health check endpoint responds correctly
- [x] Test health check mechanism under various service states
- [x] Validate monitoring and alerting integration

🤖 Generated with [Claude Code](https://claude.ai/code)

autogpt_platform/backend/backend/monitoring/notification_monitor.py

@@ -2,6 +2,7 @@
 
 import logging
 
+from autogpt_libs.utils.cache import thread_cached
 from prisma.enums import NotificationType
 from pydantic import BaseModel
 
@@ -16,6 +17,11 @@ class NotificationJobArgs(BaseModel):
     cron: str
 
 
+@thread_cached
+def get_notification_manager_client():
+    return get_service_client(NotificationManagerClient)
+
+
 def process_existing_batches(**kwargs):
     """Process existing notification batches."""
     args = NotificationJobArgs(**kwargs)
@@ -23,7 +29,7 @@ def process_existing_batches(**kwargs):
         logging.info(
             f"Processing existing batches for notification type {args.notification_types}"
         )
-        get_service_client(NotificationManagerClient).process_existing_batches(
+        get_notification_manager_client().process_existing_batches(
             args.notification_types
         )
     except Exception as e:
@@ -34,6 +40,6 @@ def process_weekly_summary(**kwargs):
     """Process weekly summary notifications."""
     try:
         logging.info("Processing weekly summary")
-        get_service_client(NotificationManagerClient).queue_weekly_summary()
+        get_notification_manager_client().queue_weekly_summary()
     except Exception as e:
         logger.exception(f"Error processing weekly summary: {e}")

autogpt_platform/backend/backend/server/rest_api.py

@@ -14,6 +14,7 @@ from autogpt_libs.feature_flag.client import (
     shutdown_launchdarkly,
 )
 from autogpt_libs.logging.utils import generate_uvicorn_config
+from autogpt_libs.utils.cache import thread_cached
 from fastapi.exceptions import RequestValidationError
 from fastapi.routing import APIRoute
 
@@ -212,8 +213,19 @@ app.include_router(
 app.mount("/external-api", external_app)
 
 
+@thread_cached
+def get_db_async_client():
+    from backend.executor import DatabaseManagerAsyncClient
+
+    return backend.util.service.get_service_client(
+        DatabaseManagerAsyncClient,
+        health_check=False,
+    )
+
+
 @app.get(path="/health", tags=["health"], dependencies=[])
 async def health():
+    await get_db_async_client().health_check_async()
     return {"status": "healthy"}
 
 

autogpt_platform/backend/backend/util/service.py

@@ -216,7 +216,10 @@ class AppService(BaseAppService, ABC):
                     methods=["POST"],
                 )
         self.fastapi_app.add_api_route(
-            "/health_check", self.health_check, methods=["POST"]
+            "/health_check", self.health_check, methods=["POST", "GET"]
+        )
+        self.fastapi_app.add_api_route(
+            "/health_check_async", self.health_check, methods=["POST", "GET"]
         )
         self.fastapi_app.add_exception_handler(
             ValueError, self._handle_internal_http_error(400)
@@ -248,6 +251,9 @@ class AppServiceClient(ABC):
     def health_check(self):
         pass
 
+    async def health_check_async(self):
+        pass
+
     def close(self):
         pass