phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

fix(caching.py): add logging module support for caching

Browse source
This commit is contained in:
Krrish Dholakia 2024-01-20 17:34:16 -08:00
parent d60af6fff5
commit 3e5b743b89
3 changed files with 23 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

1
litellm/_logging.py
View file

@ -28,3 +28,4 @@ verbose_logger = logging.getLogger("LiteLLM")
# Add the handler to the logger # Add the handler to the logger
verbose_router_logger.addHandler(handler) verbose_router_logger.addHandler(handler)
verbose_proxy_logger.addHandler(handler) verbose_proxy_logger.addHandler(handler)
verbose_logger.addHandler(handler)

6
litellm/caching.py
View file

@ -12,10 +12,12 @@ import time, logging
import json, traceback, ast, hashlib import json, traceback, ast, hashlib
from typing import Optional, Literal, List, Union, Any from typing import Optional, Literal, List, Union, Any
from openai._models import BaseModel as OpenAIObject from openai._models import BaseModel as OpenAIObject
from litellm._logging import verbose_logger
def print_verbose(print_statement): def print_verbose(print_statement):
try: try:
verbose_logger.debug(print_statement)
if litellm.set_verbose: if litellm.set_verbose:
print(print_statement) # noqa print(print_statement) # noqa
except: except:
@ -175,7 +177,7 @@ class S3Cache(BaseCache):
CacheControl=cache_control, CacheControl=cache_control,
ContentType="application/json", ContentType="application/json",
ContentLanguage="en", ContentLanguage="en",
ContentDisposition=f"inline; filename=\"{key}.json\"" ContentDisposition=f'inline; filename="{key}.json"',
) )
else: else:
cache_control = "immutable, max-age=31536000, s-maxage=31536000" cache_control = "immutable, max-age=31536000, s-maxage=31536000"
@ -187,7 +189,7 @@ class S3Cache(BaseCache):
CacheControl=cache_control, CacheControl=cache_control,
ContentType="application/json", ContentType="application/json",
ContentLanguage="en", ContentLanguage="en",
ContentDisposition=f"inline; filename=\"{key}.json\"" ContentDisposition=f'inline; filename="{key}.json"',
) )
except Exception as e: except Exception as e:
# NON blocking - notify users S3 is throwing an exception # NON blocking - notify users S3 is throwing an exception

20
litellm/proxy/proxy_server.py
View file

@ -2879,11 +2879,27 @@ async def health_readiness():
Unprotected endpoint for checking if worker can receive requests Unprotected endpoint for checking if worker can receive requests
""" """
global prisma_client global prisma_client
cache_type = None
if litellm.cache is not None:
cache_type = litellm.cache.type
if prisma_client is not None: # if db passed in, check if it's connected if prisma_client is not None: # if db passed in, check if it's connected
if prisma_client.db.is_connected() == True: if prisma_client.db.is_connected() == True:
return {"status": "healthy", "db": "connected"} response_object = {"db": "connected"}
return {
"status": "healthy",
"db": "connected",
"cache": cache_type,
"success_callbacks": litellm.success_callback,
}
else: else:
return {"status": "healthy", "db": "Not connected"} return {
"status": "healthy",
"db": "Not connected",
"cache": cache_type,
"success_callbacks": litellm.success_callback,
}
raise HTTPException(status_code=503, detail="Service Unhealthy") raise HTTPException(status_code=503, detail="Service Unhealthy")