Added error loggers in quart error_handler.

Also fixed the default quart error handler (500) printing the stacktrace. MIP-476

exareme2/controller/api/endpoint.py

@@ -70,13 +70,11 @@ async def post_algorithm(algorithm_name: str) -> str:
     if not algorithm_request_dto.request_id:
         algorithm_request_dto.request_id = UIDGenerator().get_a_uid()
 
-    # request_id = algorithm_request_dto.request_id or UIDGenerator().get_a_uid()
     controller.validate_algorithm_execution_request(
         algorithm_name=algorithm_name, algorithm_request_dto=algorithm_request_dto
     )
 
     algorithm_result = await controller.exec_algorithm(
-        # request_id=request_id,
         algorithm_name=algorithm_name,
         algorithm_request_dto=algorithm_request_dto,
     )

exareme2/controller/api/error_handlers.py

@@ -1,12 +1,12 @@
 import enum
+import traceback
 
 from quart import Blueprint
 
-from exareme2.controller import controller_logger as ctrl_logger
 from exareme2.controller.api.validator import BadRequest
-from exareme2.controller.celery_app import CeleryTaskTimeoutException
 from exareme2.controller.controller import NodeTaskTimeoutException
 from exareme2.controller.controller import NodeUnresponsiveException
+from exareme2.controller.controller_logger import get_background_service_logger
 from exareme2.exceptions import BadUserInput
 from exareme2.exceptions import DataModelUnavailable
 from exareme2.exceptions import DatasetUnavailable
@@ -35,47 +35,67 @@ class HTTPStatusCode(enum.IntEnum):
 
 @error_handlers.app_errorhandler(BadRequest)
 def handle_bad_request(error: BadRequest):
+    get_background_service_logger().info(
+        f"Request Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
     return error.message, HTTPStatusCode.BAD_REQUEST
 
 
 @error_handlers.app_errorhandler(FilterError)
 def handle_bad_request(error: FilterError):
+    get_background_service_logger().info(
+        f"Request Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
     return error.message, HTTPStatusCode.BAD_REQUEST
 
 
 @error_handlers.app_errorhandler(BadUserInput)
 def handle_bad_user_input(error: BadUserInput):
+    get_background_service_logger().info(
+        f"Request Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
     return error.message, HTTPStatusCode.BAD_USER_INPUT
 
 
 @error_handlers.app_errorhandler(DataModelUnavailable)
 def handle_bad_user_input(error: DataModelUnavailable):
+    get_background_service_logger().info(
+        f"Request Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
     return error.message, HTTPStatusCode.BAD_USER_INPUT
 
 
 @error_handlers.app_errorhandler(DatasetUnavailable)
 def handle_bad_user_input(error: DatasetUnavailable):
+    get_background_service_logger().info(
+        f"Request Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
     return error.message, HTTPStatusCode.BAD_USER_INPUT
 
 
 @error_handlers.app_errorhandler(InsufficientDataError)
 def handle_privacy_error(error: InsufficientDataError):
-    # TODO: Add proper context id. Related JIRA issue: https://team-1617704806227.atlassian.net/browse/MIP-486
-    # ctrl_logger.get_request_logger("demoContextId123").info(
-    #     f"Insufficient Data Error: \n " + error.message
-    # )
+    get_background_service_logger().info(
+        f"Request Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
     return INSUFFICIENT_DATA_ERROR_MESSAGE, HTTPStatusCode.INSUFFICIENT_DATA_ERROR
 
 
 @error_handlers.app_errorhandler(SMPCUsageError)
 def handle_smpc_error(error: SMPCUsageError):
+    get_background_service_logger().info(
+        f"Request Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
     return error.message, HTTPStatusCode.SMPC_USAGE_ERROR
 
 
 @error_handlers.app_errorhandler(NodeUnresponsiveException)
 def handle_node_unresponsive_algorithm_excecution_exception(
     error: NodeUnresponsiveException,
 ):
+    get_background_service_logger().error(
+        f"Internal Server Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
     return (
         error.message,
         HTTPStatusCode.NODE_UNRESPONSIVE_ALGORITHM_EXECUTION_ERROR,
@@ -86,24 +106,19 @@ def handle_node_unresponsive_algorithm_excecution_exception(
 def handle_node_task_timeout_algorithm_execution_exception(
     error: NodeTaskTimeoutException,
 ):
+    get_background_service_logger().error(
+        f"Internal Server Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
     return (
         error.message,
         HTTPStatusCode.NODE_TASK_TIMEOUT_ALGORITHM_EXECUTION_ERROR,
     )
 
 
-# TODO BUG https://team-1617704806227.atlassian.net/browse/MIP-476
-#  Default error handler doesn't contain enough error information.
-#  It's better to propagate, the error it's at least visible
-# @error_handlers.app_errorhandler(Exception)
-# def handle_unexpected_exception(error: Exception):
-# TODO: Add proper context id. Related JIRA issue: https://team-1617704806227.atlassian.net/browse/MIP-486
-
-#     ctrl_logger.getRequestLogger("demoContextId123").error(
-#         f"Internal Server Error."
-#         f"\nErrorType: {type(error)}"
-#         f"\nError: {error}"
-#         f"\nTraceback: {traceback.print_tb(error.__traceback__)}"
-#     )
-#
-#     return "", HTTPStatusCode.UNEXPECTED_ERROR
+@error_handlers.app_errorhandler(Exception)
+def handle_unexpected_exception(error: Exception):
+    get_background_service_logger().error(
+        f"Internal Server Error. Type: '{type(error).__name__}' Message: '{error}'"
+    )
+    traceback.print_tb(error.__traceback__)
+    return "", HTTPStatusCode.UNEXPECTED_ERROR

exareme2/controller/controller_logger.py

@@ -34,7 +34,7 @@ def init_logger(request_id, log_level=None):
     return logger
 
 
-def get_background_service_logger():
+def get_background_service_logger() -> logging.Logger:
     """
     Used for logging information produced by any background service.
     """