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

fix return num retries in exceptions

Browse source
This commit is contained in:
Ishaan Jaff 2024-05-04 19:09:34 -07:00
parent 5be8c95c6e
commit 6b59aeb603
1 changed files with 7 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

10
litellm/router.py
View file

@ -1544,6 +1544,10 @@ class Router:
num_retries=num_retries, num_retries=num_retries,
) )
await asyncio.sleep(_timeout) await asyncio.sleep(_timeout)
try:
original_exception.message += f"\nNumber Retries = {current_attempt}"
except:
pass
raise original_exception raise original_exception
def function_with_fallbacks(self, *args, **kwargs): def function_with_fallbacks(self, *args, **kwargs):
@ -1703,7 +1707,7 @@ class Router:
response = original_function(*args, **kwargs) response = original_function(*args, **kwargs)
return response return response
except Exception as e: except Exception as most_recent_exception:
## LOGGING ## LOGGING
kwargs = self.log_retry(kwargs=kwargs, e=e) kwargs = self.log_retry(kwargs=kwargs, e=e)
remaining_retries = num_retries - current_attempt remaining_retries = num_retries - current_attempt
@ -1713,7 +1717,7 @@ class Router:
num_retries=num_retries, num_retries=num_retries,
) )
time.sleep(_timeout) time.sleep(_timeout)
raise original_exception raise most_recent_exception
### HELPER FUNCTIONS ### HELPER FUNCTIONS
@ -1848,7 +1852,7 @@ class Router:
exception_status = 500 exception_status = 500
_should_retry = litellm._should_retry(status_code=exception_status) _should_retry = litellm._should_retry(status_code=exception_status)
if updated_fails > self.allowed_fails or _should_retry == False: if updated_fails > self.allowed_fails:
# get the current cooldown list for that minute # get the current cooldown list for that minute
cooldown_key = f"{current_minute}:cooldown_models" # group cooldown models by minute to reduce number of redis calls cooldown_key = f"{current_minute}:cooldown_models" # group cooldown models by minute to reduce number of redis calls
cached_value = self.cache.get_cache(key=cooldown_key) cached_value = self.cache.get_cache(key=cooldown_key)