refactor: optimize Application Insights telemetry and prevent duplicate logging handlers

src/backend/api/event_utils.py

@@ -4,8 +4,6 @@
 
 # Third-party
 from applicationinsights import TelemetryClient
-from applicationinsights.channel import SynchronousQueue, SynchronousSender, TelemetryChannel
-
 from dotenv import load_dotenv
 
 load_dotenv()
@@ -28,12 +26,8 @@ def _get_telemetry_client():
                 instrumentation_key = parts.get('InstrumentationKey')
 
                 if instrumentation_key:
-                    # Create a synchronous channel for immediate sending
-                    sender = SynchronousSender()
-                    queue = SynchronousQueue(sender)
-                    channel = TelemetryChannel(None, queue)
-
-                    _telemetry_client = TelemetryClient(instrumentation_key, channel)
+                    # Use the default (buffered/async) channel configuration
+                    _telemetry_client = TelemetryClient(instrumentation_key)
                     logging.info("Application Insights TelemetryClient initialized successfully")
                 else:
                     logging.error("Could not extract InstrumentationKey from connection string")
@@ -59,7 +53,9 @@ def track_event_if_configured(event_name: str, event_data: dict):
 
                 # Track the custom event
                 client.track_event(event_name, properties=properties)
-                client.flush()  # Ensure immediate sending
+
+                # Flush to ensure events are sent immediately
+                client.flush()
 
                 logging.debug(f"Tracked custom event: {event_name} with data: {event_data}")
             else:

src/backend/app.py

@@ -161,8 +161,14 @@ def create_app() -> FastAPI:
         set_logger_provider(logger_provider)
 
         # Attach OpenTelemetry handler to Python's root logger
-        handler = LoggingHandler(logger_provider=logger_provider)
-        logging.getLogger().addHandler(handler)
+        # Ensure we don't accumulate stale handlers or mismatch logger providers
+        root_logger = logging.getLogger()
+        # Remove any existing OpenTelemetry logging handlers to avoid duplicates
+        for handler in list(root_logger.handlers):
+            if isinstance(handler, LoggingHandler):
+                root_logger.removeHandler(handler)
+        # Add a fresh handler bound to the current logger_provider
+        root_logger.addHandler(LoggingHandler(logger_provider=logger_provider))
 
         # Instrument ONLY FastAPI for HTTP request/response tracing
         # This is safe because it only wraps HTTP handlers, not internal async operations