From 9647c6158ee90a7db621eed1abec302dc4e400d7 Mon Sep 17 00:00:00 2001
From: Ishaan Jaff <ishaanjaffer0324@gmail.com>
Date: Thu, 31 Oct 2024 18:15:35 +0530
Subject: [PATCH] fix use failing_model as cache key for failed_tracking_alert

---
 litellm/integrations/SlackAlerting/slack_alerting.py | 12 +++++++++---
 litellm/proxy/proxy_config.yaml                      | 10 +++++-----
 litellm/proxy/proxy_server.py                        |  1 +
 litellm/proxy/utils.py                               |  9 +++++++--
 4 files changed, 22 insertions(+), 10 deletions(-)

diff --git a/litellm/integrations/SlackAlerting/slack_alerting.py b/litellm/integrations/SlackAlerting/slack_alerting.py
index dbe9e4161da..85d54a337d1 100644
--- a/litellm/integrations/SlackAlerting/slack_alerting.py
+++ b/litellm/integrations/SlackAlerting/slack_alerting.py
@@ -550,8 +550,14 @@ class SlackAlerting(CustomBatchLogger):
                     alerting_metadata=alerting_metadata,
                 )
 
-    async def failed_tracking_alert(self, error_message: str):
-        """Raise alert when tracking failed for specific model"""
+    async def failed_tracking_alert(self, error_message: str, failing_model: str):
+        """
+        Raise alert when tracking failed for specific model
+
+        Args:
+            error_message (str): Error message
+            failing_model (str): Model that failed tracking
+        """
         if self.alerting is None or self.alert_types is None:
             # do nothing if alerting is not switched on
             return
@@ -560,7 +566,7 @@ class SlackAlerting(CustomBatchLogger):
 
         _cache: DualCache = self.internal_usage_cache
         message = "Failed Tracking Cost for " + error_message
-        _cache_key = "budget_alerts:failed_tracking:{}".format(message)
+        _cache_key = "budget_alerts:failed_tracking:{}".format(failing_model)
         result = await _cache.async_get_cache(key=_cache_key)
         if result is None:
             await self.send_alert(
diff --git a/litellm/proxy/proxy_config.yaml b/litellm/proxy/proxy_config.yaml
index 95eff095c22..23834f75970 100644
--- a/litellm/proxy/proxy_config.yaml
+++ b/litellm/proxy/proxy_config.yaml
@@ -1,11 +1,11 @@
 model_list:
   - model_name: gpt-4o
     litellm_params:
-      model: gpt-4o
+      model: openai/gpt-5
       api_key: os.environ/OPENAI_API_KEY
       api_base: https://exampleopenaiendpoint-production.up.railway.app/
 
-litellm_settings:
-  callbacks: ["prometheus"]
-  service_callback: ["prometheus_system"]
-  cache: true
+
+general_settings: 
+    alerting: ["slack"]
+    alerting_threshold: 0.001
diff --git a/litellm/proxy/proxy_server.py b/litellm/proxy/proxy_server.py
index 1f0271f3451..402063177a6 100644
--- a/litellm/proxy/proxy_server.py
+++ b/litellm/proxy/proxy_server.py
@@ -845,6 +845,7 @@ async def _PROXY_track_cost_callback(
         asyncio.create_task(
             proxy_logging_obj.failed_tracking_alert(
                 error_message=error_msg,
+                failing_model=model,
             )
         )
         verbose_proxy_logger.debug("error in tracking cost callback - %s", e)
diff --git a/litellm/proxy/utils.py b/litellm/proxy/utils.py
index 656e97a6c5c..8919da97861 100644
--- a/litellm/proxy/utils.py
+++ b/litellm/proxy/utils.py
@@ -667,13 +667,18 @@ class ProxyLogging:
                 raise e
         return data
 
-    async def failed_tracking_alert(self, error_message: str):
+    async def failed_tracking_alert(
+        self,
+        error_message: str,
+        failing_model: str,
+    ):
         if self.alerting is None:
             return
 
         if self.slack_alerting_instance:
             await self.slack_alerting_instance.failed_tracking_alert(
-                error_message=error_message
+                error_message=error_message,
+                failing_model=failing_model,
             )
 
     async def budget_alerts(