diff --git a/litellm/proxy/hooks/parallel_request_limiter.py b/litellm/proxy/hooks/parallel_request_limiter.py index 30877daf39..5c1893ea51 100644 --- a/litellm/proxy/hooks/parallel_request_limiter.py +++ b/litellm/proxy/hooks/parallel_request_limiter.py @@ -17,7 +17,12 @@ class MaxParallelRequestsHandler(CustomLogger): pass def print_verbose(self, print_statement): - verbose_proxy_logger.debug(print_statement) + try: + verbose_proxy_logger.debug(print_statement) + if litellm.set_verbose: + print(print_statement) # noqa + except: + pass async def async_pre_call_hook( self, diff --git a/litellm/proxy/utils.py b/litellm/proxy/utils.py index 9e46db796c..3ec45203f5 100644 --- a/litellm/proxy/utils.py +++ b/litellm/proxy/utils.py @@ -21,6 +21,7 @@ from datetime import datetime, timedelta def print_verbose(print_statement): + verbose_proxy_logger.debug(print_statement) if litellm.set_verbose: print(f"LiteLLM Proxy: {print_statement}") # noqa @@ -96,6 +97,7 @@ class ProxyLogging: 2. /embeddings 3. /image/generation """ + print_verbose(f"Inside Proxy Logging Pre-call hook!") ### ALERTING ### asyncio.create_task(self.response_taking_too_long(request_data=data)) @@ -1035,7 +1037,7 @@ async def send_email(sender_name, sender_email, receiver_email, subject, html): print_verbose(f"SMTP Connection Init") # Establish a secure connection with the SMTP server with smtplib.SMTP(smtp_host, smtp_port) as server: - if os.getenv("SMTP_TLS", 'True') != "False": + if os.getenv("SMTP_TLS", "True") != "False": server.starttls() # Login to your email account diff --git a/litellm/tests/test_parallel_request_limiter.py b/litellm/tests/test_parallel_request_limiter.py index 384044c9c6..dee909eaa5 100644 --- a/litellm/tests/test_parallel_request_limiter.py +++ b/litellm/tests/test_parallel_request_limiter.py @@ -456,6 +456,7 @@ async def test_streaming_router_call(): @pytest.mark.asyncio async def test_streaming_router_tpm_limit(): + litellm.set_verbose = True model_list = [ { "model_name": "azure-model", @@ -520,7 +521,7 @@ async def test_streaming_router_tpm_limit(): ) async for chunk in response: continue - await asyncio.sleep(1) # success is done in a separate thread + await asyncio.sleep(5) # success is done in a separate thread try: await parallel_request_handler.async_pre_call_hook(