phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 19:24:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix proxy pre call hook - only use if user is using alerting (#7683)

Browse source
This commit is contained in:
Ishaan Jaff 2025-01-10 19:07:05 -08:00 • committed by GitHub
parent 76c81f97d7
commit a49ef48bcb
1 changed files with 19 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

23
litellm/proxy/utils.py
View file

@ -473,10 +473,8 @@ class ProxyLogging:
3. /image/generation 3. /image/generation
""" """
verbose_proxy_logger.debug("Inside Proxy Logging Pre-call hook!") verbose_proxy_logger.debug("Inside Proxy Logging Pre-call hook!")
## ALERTING ###
asyncio.create_task( self._init_response_taking_too_long_task(data=data)
self.slack_alerting_instance.response_taking_too_long(request_data=data)
)
if data is None: if data is None:
return None return None
@ -1012,6 +1010,23 @@ class ProxyLogging:
raise e raise e
return new_response return new_response
def _init_response_taking_too_long_task(self, data: Optional[dict] = None):
"""
Initialize the response taking too long task if user is using slack alerting
Only run task if user is using slack alerting
This handles checking for if a request is hanging for too long
"""
## ALERTING ###
if (
self.slack_alerting_instance
and self.slack_alerting_instance.alerting is not None
):
asyncio.create_task(
self.slack_alerting_instance.response_taking_too_long(request_data=data)
)
### DB CONNECTOR ### ### DB CONNECTOR ###
# Define the retry decorator with backoff strategy # Define the retry decorator with backoff strategy