feat: tenant app invocations limiter (#16221)

api/controllers/console/app/workflow.py

@@ -10,9 +10,14 @@ from werkzeug.exceptions import Forbidden, InternalServerError, NotFound
 import services
 from configs import dify_config
 from controllers.console import api
-from controllers.console.app.error import ConversationCompletedError, DraftWorkflowNotExist, DraftWorkflowNotSync
+from controllers.console.app.error import (
+    ConversationCompletedError,
+    DraftWorkflowNotExist,
+    DraftWorkflowNotSync,
+)
 from controllers.console.app.wraps import get_app_model
 from controllers.console.wraps import account_initialization_required, setup_required
+from controllers.web.error import InvokeRateLimitError as InvokeRateLimitHttpError
 from core.app.apps.base_app_queue_manager import AppQueueManager
 from core.app.entities.app_invoke_entities import InvokeFrom
 from extensions.ext_database import db
@@ -27,6 +32,7 @@ from models.account import Account
 from models.model import AppMode
 from services.app_generate_service import AppGenerateService
 from services.errors.app import WorkflowHashNotEqualError
+from services.errors.llm import InvokeRateLimitError
 from services.workflow_service import DraftWorkflowDeletionError, WorkflowInUseError, WorkflowService
 
 logger = logging.getLogger(__name__)
@@ -168,6 +174,8 @@ class AdvancedChatDraftWorkflowRunApi(Resource):
             raise NotFound("Conversation Not Exists.")
         except services.errors.conversation.ConversationCompletedError:
             raise ConversationCompletedError()
+        except InvokeRateLimitError as ex:
+            raise InvokeRateLimitHttpError(ex.description)
         except ValueError as e:
             raise e
         except Exception:
@@ -344,15 +352,18 @@ class DraftWorkflowRunApi(Resource):
         parser.add_argument("files", type=list, required=False, location="json")
         args = parser.parse_args()
 
-        response = AppGenerateService.generate(
-            app_model=app_model,
-            user=current_user,
-            args=args,
-            invoke_from=InvokeFrom.DEBUGGER,
-            streaming=True,
-        )
+        try:
+            response = AppGenerateService.generate(
+                app_model=app_model,
+                user=current_user,
+                args=args,
+                invoke_from=InvokeFrom.DEBUGGER,
+                streaming=True,
+            )
 
-        return helper.compact_generate_response(response)
+            return helper.compact_generate_response(response)
+        except InvokeRateLimitError as ex:
+            raise InvokeRateLimitHttpError(ex.description)
 
 
 class WorkflowTaskStopApi(Resource):

api/controllers/console/explore/completion.py

@@ -16,6 +16,7 @@ from controllers.console.app.error import (
 )
 from controllers.console.explore.error import NotChatAppError, NotCompletionAppError
 from controllers.console.explore.wraps import InstalledAppResource
+from controllers.web.error import InvokeRateLimitError as InvokeRateLimitHttpError
 from core.app.apps.base_app_queue_manager import AppQueueManager
 from core.app.entities.app_invoke_entities import InvokeFrom
 from core.errors.error import (
@@ -29,6 +30,7 @@ from libs import helper
 from libs.helper import uuid_value
 from models.model import AppMode
 from services.app_generate_service import AppGenerateService
+from services.errors.llm import InvokeRateLimitError
 
 
 # define completion api for user
@@ -75,7 +77,7 @@ class CompletionApi(InstalledAppResource):
             raise CompletionRequestError(e.description)
         except ValueError as e:
             raise e
-        except Exception as e:
+        except Exception:
             logging.exception("internal server error.")
             raise InternalServerError()
 
@@ -133,9 +135,11 @@ class ChatApi(InstalledAppResource):
             raise ProviderModelCurrentlyNotSupportError()
         except InvokeError as e:
             raise CompletionRequestError(e.description)
+        except InvokeRateLimitError as ex:
+            raise InvokeRateLimitHttpError(ex.description)
         except ValueError as e:
             raise e
-        except Exception as e:
+        except Exception:
             logging.exception("internal server error.")
             raise InternalServerError()
 

api/controllers/console/explore/workflow.py

@@ -11,6 +11,7 @@ from controllers.console.app.error import (
 )
 from controllers.console.explore.error import NotWorkflowAppError
 from controllers.console.explore.wraps import InstalledAppResource
+from controllers.web.error import InvokeRateLimitError as InvokeRateLimitHttpError
 from core.app.apps.base_app_queue_manager import AppQueueManager
 from core.app.entities.app_invoke_entities import InvokeFrom
 from core.errors.error import (
@@ -23,6 +24,7 @@ from libs import helper
 from libs.login import current_user
 from models.model import AppMode, InstalledApp
 from services.app_generate_service import AppGenerateService
+from services.errors.llm import InvokeRateLimitError
 
 logger = logging.getLogger(__name__)
 
@@ -56,9 +58,11 @@ class InstalledAppWorkflowRunApi(InstalledAppResource):
             raise ProviderModelCurrentlyNotSupportError()
         except InvokeError as e:
             raise CompletionRequestError(e.description)
+        except InvokeRateLimitError as ex:
+            raise InvokeRateLimitHttpError(ex.description)
         except ValueError as e:
             raise e
-        except Exception as e:
+        except Exception:
             logging.exception("internal server error.")
             raise InternalServerError()