mirror of
https://github.com/BerriAI/litellm.git
synced 2025-04-27 03:34:10 +00:00
Merge branch 'main' into litellm_svc_logger
This commit is contained in:
Ishaan Jaff
GitHub
commit
2cf3133669
84 changed files with 3848 additions and 5302 deletions
|
|
@ -1,6 +1,7 @@
|
|||
import json
|
||||
import logging
|
||||
from logging import Formatter
|
||||
import sys
|
||||
|
||||
|
||||
class JsonFormatter(Formatter):
|
||||
|
|
|
|||
|
|
@ -56,8 +56,10 @@ router_settings:
|
|||
|
||||
litellm_settings:
|
||||
success_callback: ["langfuse"]
|
||||
json_logs: true
|
||||
|
||||
general_settings:
|
||||
alerting: ["email"]
|
||||
key_management_system: "aws_kms"
|
||||
key_management_settings:
|
||||
hosted_keys: ["LITELLM_MASTER_KEY"]
|
||||
|
||||
|
|
|
|||
|
|
@ -955,6 +955,7 @@ class KeyManagementSystem(enum.Enum):
|
|||
AZURE_KEY_VAULT = "azure_key_vault"
|
||||
AWS_SECRET_MANAGER = "aws_secret_manager"
|
||||
LOCAL = "local"
|
||||
AWS_KMS = "aws_kms"
|
||||
|
||||
|
||||
class KeyManagementSettings(LiteLLMBase):
|
||||
|
|
|
|||
|
|
@ -106,6 +106,40 @@ def common_checks(
|
|||
raise Exception(
|
||||
f"'user' param not passed in. 'enforce_user_param'={general_settings['enforce_user_param']}"
|
||||
)
|
||||
if general_settings.get("enforced_params", None) is not None:
|
||||
# Enterprise ONLY Feature
|
||||
# we already validate if user is premium_user when reading the config
|
||||
# Add an extra premium_usercheck here too, just incase
|
||||
from litellm.proxy.proxy_server import premium_user, CommonProxyErrors
|
||||
|
||||
if premium_user is not True:
|
||||
raise ValueError(
|
||||
"Trying to use `enforced_params`"
|
||||
+ CommonProxyErrors.not_premium_user.value
|
||||
)
|
||||
|
||||
if route in LiteLLMRoutes.openai_routes.value:
|
||||
# loop through each enforced param
|
||||
# example enforced_params ['user', 'metadata', 'metadata.generation_name']
|
||||
for enforced_param in general_settings["enforced_params"]:
|
||||
_enforced_params = enforced_param.split(".")
|
||||
if len(_enforced_params) == 1:
|
||||
if _enforced_params[0] not in request_body:
|
||||
raise ValueError(
|
||||
f"BadRequest please pass param={_enforced_params[0]} in request body. This is a required param"
|
||||
)
|
||||
elif len(_enforced_params) == 2:
|
||||
# this is a scenario where user requires request['metadata']['generation_name'] to exist
|
||||
if _enforced_params[0] not in request_body:
|
||||
raise ValueError(
|
||||
f"BadRequest please pass param={_enforced_params[0]} in request body. This is a required param"
|
||||
)
|
||||
if _enforced_params[1] not in request_body[_enforced_params[0]]:
|
||||
raise ValueError(
|
||||
f"BadRequest please pass param=[{_enforced_params[0]}][{_enforced_params[1]}] in request body. This is a required param"
|
||||
)
|
||||
|
||||
pass
|
||||
# 7. [OPTIONAL] If 'litellm.max_budget' is set (>0), is proxy under budget
|
||||
if (
|
||||
litellm.max_budget > 0
|
||||
|
|
|
|||
|
|
@ -88,7 +88,7 @@ class _PROXY_AzureContentSafety(
|
|||
verbose_proxy_logger.debug(
|
||||
"Error in Azure Content-Safety: %s", traceback.format_exc()
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
raise
|
||||
|
||||
result = self._compute_result(response)
|
||||
|
|
@ -123,7 +123,12 @@ class _PROXY_AzureContentSafety(
|
|||
except HTTPException as e:
|
||||
raise e
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.hooks.azure_content_safety.py::async_pre_call_hook(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
|
||||
async def async_post_call_success_hook(
|
||||
self,
|
||||
|
|
|
|||
|
|
@ -94,7 +94,12 @@ class _PROXY_BatchRedisRequests(CustomLogger):
|
|||
except HTTPException as e:
|
||||
raise e
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.hooks.batch_redis_get.py::async_pre_call_hook(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
|
||||
async def async_get_cache(self, *args, **kwargs):
|
||||
"""
|
||||
|
|
|
|||
|
|
@ -1,13 +1,13 @@
|
|||
# What this does?
|
||||
## Checks if key is allowed to use the cache controls passed in to the completion() call
|
||||
|
||||
from typing import Optional
|
||||
import litellm
|
||||
from litellm import verbose_logger
|
||||
from litellm.caching import DualCache
|
||||
from litellm.proxy._types import UserAPIKeyAuth
|
||||
from litellm.integrations.custom_logger import CustomLogger
|
||||
from fastapi import HTTPException
|
||||
import json, traceback
|
||||
import traceback
|
||||
|
||||
|
||||
class _PROXY_CacheControlCheck(CustomLogger):
|
||||
|
|
@ -54,4 +54,9 @@ class _PROXY_CacheControlCheck(CustomLogger):
|
|||
except HTTPException as e:
|
||||
raise e
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_logger.error(
|
||||
"litellm.proxy.hooks.cache_control_check.py::async_pre_call_hook(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_logger.debug(traceback.format_exc())
|
||||
|
|
|
|||
|
|
@ -1,10 +1,10 @@
|
|||
from typing import Optional
|
||||
from litellm import verbose_logger
|
||||
import litellm
|
||||
from litellm.caching import DualCache
|
||||
from litellm.proxy._types import UserAPIKeyAuth
|
||||
from litellm.integrations.custom_logger import CustomLogger
|
||||
from fastapi import HTTPException
|
||||
import json, traceback
|
||||
import traceback
|
||||
|
||||
|
||||
class _PROXY_MaxBudgetLimiter(CustomLogger):
|
||||
|
|
@ -44,4 +44,9 @@ class _PROXY_MaxBudgetLimiter(CustomLogger):
|
|||
except HTTPException as e:
|
||||
raise e
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_logger.error(
|
||||
"litellm.proxy.hooks.max_budget_limiter.py::async_pre_call_hook(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_logger.debug(traceback.format_exc())
|
||||
|
|
|
|||
|
|
@ -8,8 +8,8 @@
|
|||
# Tell us how we can improve! - Krrish & Ishaan
|
||||
|
||||
|
||||
from typing import Optional, Literal, Union
|
||||
import litellm, traceback, sys, uuid, json
|
||||
from typing import Optional, Union
|
||||
import litellm, traceback, uuid, json # noqa: E401
|
||||
from litellm.caching import DualCache
|
||||
from litellm.proxy._types import UserAPIKeyAuth
|
||||
from litellm.integrations.custom_logger import CustomLogger
|
||||
|
|
@ -21,8 +21,8 @@ from litellm.utils import (
|
|||
ImageResponse,
|
||||
StreamingChoices,
|
||||
)
|
||||
from datetime import datetime
|
||||
import aiohttp, asyncio
|
||||
import aiohttp
|
||||
import asyncio
|
||||
|
||||
|
||||
class _OPTIONAL_PresidioPIIMasking(CustomLogger):
|
||||
|
|
@ -138,7 +138,12 @@ class _OPTIONAL_PresidioPIIMasking(CustomLogger):
|
|||
else:
|
||||
raise Exception(f"Invalid anonymizer response: {redacted_text}")
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.hooks.presidio_pii_masking.py::async_pre_call_hook(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
raise e
|
||||
|
||||
async def async_pre_call_hook(
|
||||
|
|
|
|||
|
|
@ -204,7 +204,12 @@ class _OPTIONAL_PromptInjectionDetection(CustomLogger):
|
|||
return e.detail["error"]
|
||||
raise e
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.hooks.prompt_injection_detection.py::async_pre_call_hook(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
|
||||
async def async_moderation_hook(
|
||||
self,
|
||||
|
|
|
|||
|
|
@ -21,7 +21,14 @@ model_list:
|
|||
|
||||
general_settings:
|
||||
master_key: sk-1234
|
||||
alerting: ["slack"]
|
||||
|
||||
litellm_settings:
|
||||
callbacks: ["otel"]
|
||||
store_audit_logs: true
|
||||
store_audit_logs: true
|
||||
redact_messages_in_exceptions: True
|
||||
enforced_params:
|
||||
- user
|
||||
- metadata
|
||||
- metadata.generation_name
|
||||
|
||||
|
|
|
|||
|
|
@ -111,6 +111,7 @@ from litellm.proxy.utils import (
|
|||
encrypt_value,
|
||||
decrypt_value,
|
||||
_to_ns,
|
||||
get_error_message_str,
|
||||
)
|
||||
from litellm import (
|
||||
CreateBatchRequest,
|
||||
|
|
@ -120,7 +121,10 @@ from litellm import (
|
|||
CreateFileRequest,
|
||||
)
|
||||
from litellm.proxy.secret_managers.google_kms import load_google_kms
|
||||
from litellm.proxy.secret_managers.aws_secret_manager import load_aws_secret_manager
|
||||
from litellm.proxy.secret_managers.aws_secret_manager import (
|
||||
load_aws_secret_manager,
|
||||
load_aws_kms,
|
||||
)
|
||||
import pydantic
|
||||
from litellm.proxy._types import *
|
||||
from litellm.caching import DualCache, RedisCache
|
||||
|
|
@ -133,7 +137,10 @@ from litellm.router import (
|
|||
AssistantsTypedDict,
|
||||
)
|
||||
from litellm.router import ModelInfo as RouterModelInfo
|
||||
from litellm._logging import verbose_router_logger, verbose_proxy_logger
|
||||
from litellm._logging import (
|
||||
verbose_router_logger,
|
||||
verbose_proxy_logger,
|
||||
)
|
||||
from litellm.proxy.auth.handle_jwt import JWTHandler
|
||||
from litellm.proxy.auth.litellm_license import LicenseCheck
|
||||
from litellm.proxy.auth.model_checks import (
|
||||
|
|
@ -1515,7 +1522,12 @@ async def user_api_key_auth(
|
|||
else:
|
||||
raise Exception()
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.user_api_key_auth(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, litellm.BudgetExceededError):
|
||||
raise ProxyException(
|
||||
message=e.message, type="auth_error", param=None, code=400
|
||||
|
|
@ -2782,10 +2794,12 @@ class ProxyConfig:
|
|||
load_google_kms(use_google_kms=True)
|
||||
elif (
|
||||
key_management_system
|
||||
== KeyManagementSystem.AWS_SECRET_MANAGER.value
|
||||
== KeyManagementSystem.AWS_SECRET_MANAGER.value # noqa: F405
|
||||
):
|
||||
### LOAD FROM AWS SECRET MANAGER ###
|
||||
load_aws_secret_manager(use_aws_secret_manager=True)
|
||||
elif key_management_system == KeyManagementSystem.AWS_KMS.value:
|
||||
load_aws_kms(use_aws_kms=True)
|
||||
else:
|
||||
raise ValueError("Invalid Key Management System selected")
|
||||
key_management_settings = general_settings.get(
|
||||
|
|
@ -2819,6 +2833,7 @@ class ProxyConfig:
|
|||
master_key = general_settings.get(
|
||||
"master_key", litellm.get_secret("LITELLM_MASTER_KEY", None)
|
||||
)
|
||||
|
||||
if master_key and master_key.startswith("os.environ/"):
|
||||
master_key = litellm.get_secret(master_key)
|
||||
if not isinstance(master_key, str):
|
||||
|
|
@ -2909,6 +2924,16 @@ class ProxyConfig:
|
|||
)
|
||||
health_check_interval = general_settings.get("health_check_interval", 300)
|
||||
|
||||
## check if user has set a premium feature in general_settings
|
||||
if (
|
||||
general_settings.get("enforced_params") is not None
|
||||
and premium_user is not True
|
||||
):
|
||||
raise ValueError(
|
||||
"Trying to use `enforced_params`"
|
||||
+ CommonProxyErrors.not_premium_user.value
|
||||
)
|
||||
|
||||
router_params: dict = {
|
||||
"cache_responses": litellm.cache
|
||||
!= None, # cache if user passed in cache values
|
||||
|
|
@ -3522,7 +3547,12 @@ async def generate_key_helper_fn(
|
|||
)
|
||||
key_data["token_id"] = getattr(create_key_response, "token", None)
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.generate_key_helper_fn(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise e
|
||||
raise HTTPException(
|
||||
|
|
@ -3561,7 +3591,12 @@ async def delete_verification_token(tokens: List, user_id: Optional[str] = None)
|
|||
else:
|
||||
raise Exception("DB not connected. prisma_client is None")
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.delete_verification_token(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
raise e
|
||||
return deleted_tokens
|
||||
|
||||
|
|
@ -3722,7 +3757,12 @@ async def async_assistants_data_generator(
|
|||
done_message = "[DONE]"
|
||||
yield f"data: {done_message}\n\n"
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.async_assistants_data_generator(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
original_exception=e,
|
||||
|
|
@ -3732,9 +3772,6 @@ async def async_assistants_data_generator(
|
|||
f"\033[1;31mAn error occurred: {e}\n\n Debug this by setting `--debug`, e.g. `litellm --model gpt-3.5-turbo --debug`"
|
||||
)
|
||||
router_model_names = llm_router.model_names if llm_router is not None else []
|
||||
if user_debug:
|
||||
traceback.print_exc()
|
||||
|
||||
if isinstance(e, HTTPException):
|
||||
raise e
|
||||
else:
|
||||
|
|
@ -3774,7 +3811,12 @@ async def async_data_generator(
|
|||
done_message = "[DONE]"
|
||||
yield f"data: {done_message}\n\n"
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.async_data_generator(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
original_exception=e,
|
||||
|
|
@ -3784,8 +3826,6 @@ async def async_data_generator(
|
|||
f"\033[1;31mAn error occurred: {e}\n\n Debug this by setting `--debug`, e.g. `litellm --model gpt-3.5-turbo --debug`"
|
||||
)
|
||||
router_model_names = llm_router.model_names if llm_router is not None else []
|
||||
if user_debug:
|
||||
traceback.print_exc()
|
||||
|
||||
if isinstance(e, HTTPException):
|
||||
raise e
|
||||
|
|
@ -3846,6 +3886,18 @@ def on_backoff(details):
|
|||
verbose_proxy_logger.debug("Backing off... this was attempt # %s", details["tries"])
|
||||
|
||||
|
||||
def giveup(e):
|
||||
result = not (
|
||||
isinstance(e, ProxyException)
|
||||
and getattr(e, "message", None) is not None
|
||||
and isinstance(e.message, str)
|
||||
and "Max parallel request limit reached" in e.message
|
||||
)
|
||||
if result:
|
||||
verbose_proxy_logger.info(json.dumps({"event": "giveup", "exception": str(e)}))
|
||||
return result
|
||||
|
||||
|
||||
@router.on_event("startup")
|
||||
async def startup_event():
|
||||
global prisma_client, master_key, use_background_health_checks, llm_router, llm_model_list, general_settings, proxy_budget_rescheduler_min_time, proxy_budget_rescheduler_max_time, litellm_proxy_admin_name, db_writer_client, store_model_in_db
|
||||
|
|
@ -4130,12 +4182,8 @@ def model_list(
|
|||
max_tries=litellm.num_retries or 3, # maximum number of retries
|
||||
max_time=litellm.request_timeout or 60, # maximum total time to retry for
|
||||
on_backoff=on_backoff, # specifying the function to call on backoff
|
||||
giveup=lambda e: not (
|
||||
isinstance(e, ProxyException)
|
||||
and getattr(e, "message", None) is not None
|
||||
and isinstance(e.message, str)
|
||||
and "Max parallel request limit reached" in e.message
|
||||
), # the result of the logical expression is on the second position
|
||||
giveup=giveup,
|
||||
logger=verbose_proxy_logger,
|
||||
)
|
||||
async def chat_completion(
|
||||
request: Request,
|
||||
|
|
@ -4144,6 +4192,7 @@ async def chat_completion(
|
|||
user_api_key_dict: UserAPIKeyAuth = Depends(user_api_key_auth),
|
||||
):
|
||||
global general_settings, user_debug, proxy_logging_obj, llm_model_list
|
||||
|
||||
data = {}
|
||||
try:
|
||||
body = await request.body()
|
||||
|
|
@ -4434,7 +4483,12 @@ async def chat_completion(
|
|||
return _chat_response
|
||||
except Exception as e:
|
||||
data["litellm_status"] = "fail" # used for alerting
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.chat_completion(): Exception occured - {}".format(
|
||||
get_error_message_str(e=e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
|
|
@ -4445,8 +4499,6 @@ async def chat_completion(
|
|||
litellm_debug_info,
|
||||
)
|
||||
router_model_names = llm_router.model_names if llm_router is not None else []
|
||||
if user_debug:
|
||||
traceback.print_exc()
|
||||
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
|
|
@ -4678,15 +4730,12 @@ async def completion(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
verbose_proxy_logger.debug("EXCEPTION RAISED IN PROXY MAIN.PY")
|
||||
litellm_debug_info = getattr(e, "litellm_debug_info", "")
|
||||
verbose_proxy_logger.debug(
|
||||
"\033[1;31mAn error occurred: %s %s\n\n Debug this by setting `--debug`, e.g. `litellm --model gpt-3.5-turbo --debug`",
|
||||
e,
|
||||
litellm_debug_info,
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.completion(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
traceback.print_exc()
|
||||
error_traceback = traceback.format_exc()
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -4896,7 +4945,12 @@ async def embeddings(
|
|||
e,
|
||||
litellm_debug_info,
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.embeddings(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e)),
|
||||
|
|
@ -5075,7 +5129,12 @@ async def image_generation(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.image_generation(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e)),
|
||||
|
|
@ -5253,7 +5312,12 @@ async def audio_speech(
|
|||
)
|
||||
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.audio_speech(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
raise e
|
||||
|
||||
|
||||
|
|
@ -5442,7 +5506,12 @@ async def audio_transcriptions(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.audio_transcription(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -5451,7 +5520,6 @@ async def audio_transcriptions(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -5579,7 +5647,12 @@ async def get_assistants(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.get_assistants(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -5588,7 +5661,6 @@ async def get_assistants(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -5708,7 +5780,12 @@ async def create_threads(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.create_threads(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -5717,7 +5794,6 @@ async def create_threads(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -5836,7 +5912,12 @@ async def get_thread(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.get_thread(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -5845,7 +5926,6 @@ async def get_thread(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -5967,7 +6047,12 @@ async def add_messages(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.add_messages(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -5976,7 +6061,6 @@ async def add_messages(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -6094,7 +6178,12 @@ async def get_messages(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.get_messages(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -6103,7 +6192,6 @@ async def get_messages(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -6235,7 +6323,12 @@ async def run_thread(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.run_thread(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -6244,7 +6337,6 @@ async def run_thread(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -6383,7 +6475,12 @@ async def create_batch(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.create_batch(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -6392,7 +6489,6 @@ async def create_batch(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -6526,7 +6622,12 @@ async def retrieve_batch(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.retrieve_batch(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -6679,7 +6780,12 @@ async def create_file(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.create_file(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e.detail)),
|
||||
|
|
@ -6688,7 +6794,6 @@ async def create_file(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -6864,7 +6969,12 @@ async def moderations(
|
|||
await proxy_logging_obj.post_call_failure_hook(
|
||||
user_api_key_dict=user_api_key_dict, original_exception=e, request_data=data
|
||||
)
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.moderations(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", str(e)),
|
||||
|
|
@ -6873,7 +6983,6 @@ async def moderations(
|
|||
code=getattr(e, "status_code", status.HTTP_400_BAD_REQUEST),
|
||||
)
|
||||
else:
|
||||
error_traceback = traceback.format_exc()
|
||||
error_msg = f"{str(e)}"
|
||||
raise ProxyException(
|
||||
message=getattr(e, "message", error_msg),
|
||||
|
|
@ -7184,7 +7293,12 @@ async def generate_key_fn(
|
|||
|
||||
return GenerateKeyResponse(**response)
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.generate_key_fn(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -9639,7 +9753,12 @@ async def user_info(
|
|||
}
|
||||
return response_data
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.user_info(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -9734,7 +9853,12 @@ async def user_update(data: UpdateUserRequest):
|
|||
return response
|
||||
# update based on remaining passed in values
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.user_update(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -9787,7 +9911,12 @@ async def user_request_model(request: Request):
|
|||
return {"status": "success"}
|
||||
# update based on remaining passed in values
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.user_request_model(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -9829,7 +9958,12 @@ async def user_get_requests():
|
|||
return {"requests": response}
|
||||
# update based on remaining passed in values
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.user_get_requests(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -10219,7 +10353,12 @@ async def update_end_user(
|
|||
|
||||
# update based on remaining passed in values
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.update_end_user(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Internal Server Error({str(e)})"),
|
||||
|
|
@ -10303,7 +10442,12 @@ async def delete_end_user(
|
|||
|
||||
# update based on remaining passed in values
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.delete_end_user(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Internal Server Error({str(e)})"),
|
||||
|
|
@ -11606,7 +11750,12 @@ async def add_new_model(
|
|||
return model_response
|
||||
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.add_new_model(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -11720,7 +11869,12 @@ async def update_model(
|
|||
|
||||
return model_response
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.update_model(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -13954,7 +14108,12 @@ async def update_config(config_info: ConfigYAML):
|
|||
|
||||
return {"message": "Config updated successfully"}
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.update_config(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -14427,7 +14586,12 @@ async def get_config():
|
|||
"available_callbacks": all_available_callbacks,
|
||||
}
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.get_config(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -14678,7 +14842,12 @@ async def health_services_endpoint(
|
|||
}
|
||||
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.health_services_endpoint(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
if isinstance(e, HTTPException):
|
||||
raise ProxyException(
|
||||
message=getattr(e, "detail", f"Authentication Error({str(e)})"),
|
||||
|
|
@ -14757,7 +14926,12 @@ async def health_endpoint(
|
|||
"unhealthy_count": len(unhealthy_endpoints),
|
||||
}
|
||||
except Exception as e:
|
||||
traceback.print_exc()
|
||||
verbose_proxy_logger.error(
|
||||
"litellm.proxy.proxy_server.py::health_endpoint(): Exception occured - {}".format(
|
||||
str(e)
|
||||
)
|
||||
)
|
||||
verbose_proxy_logger.debug(traceback.format_exc())
|
||||
raise e
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -8,7 +8,8 @@ Requires:
|
|||
* `pip install boto3>=1.28.57`
|
||||
"""
|
||||
|
||||
import litellm, os
|
||||
import litellm
|
||||
import os
|
||||
from typing import Optional
|
||||
from litellm.proxy._types import KeyManagementSystem
|
||||
|
||||
|
|
@ -38,3 +39,21 @@ def load_aws_secret_manager(use_aws_secret_manager: Optional[bool]):
|
|||
|
||||
except Exception as e:
|
||||
raise e
|
||||
|
||||
|
||||
def load_aws_kms(use_aws_kms: Optional[bool]):
|
||||
if use_aws_kms is None or use_aws_kms is False:
|
||||
return
|
||||
try:
|
||||
import boto3
|
||||
|
||||
validate_environment()
|
||||
|
||||
# Create a Secrets Manager client
|
||||
kms_client = boto3.client("kms", region_name=os.getenv("AWS_REGION_NAME"))
|
||||
|
||||
litellm.secret_manager_client = kms_client
|
||||
litellm._key_management_system = KeyManagementSystem.AWS_KMS
|
||||
|
||||
except Exception as e:
|
||||
raise e
|
||||
|
|
|
|||
|
|
@ -2889,3 +2889,16 @@ missing_keys_html_form = """
|
|||
|
||||
def _to_ns(dt):
|
||||
return int(dt.timestamp() * 1e9)
|
||||
|
||||
def get_error_message_str(e: Exception) -> str:
|
||||
error_message = ""
|
||||
if isinstance(e, HTTPException):
|
||||
if isinstance(e.detail, str):
|
||||
error_message = e.detail
|
||||
elif isinstance(e.detail, dict):
|
||||
error_message = json.dumps(e.detail)
|
||||
else:
|
||||
error_message = str(e)
|
||||
else:
|
||||
error_message = str(e)
|
||||
return error_message
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue