Merge pull request #3622 from msabramo/abramowi/customizable-slack-report-frequency

Add ability to customize slack report frequency

docs/my-website/docs/proxy/alerting.md

@@ -1,6 +1,7 @@
 # 🚨 Alerting 
 
 Get alerts for:
+
 - Hanging LLM api calls
 - Failed LLM api calls
 - Slow LLM api calls
@@ -10,6 +11,12 @@ Get alerts for:
 - Spend Reports - Weekly & Monthly spend per Team, Tag
 - Failed db read/writes
 
+As a bonus, you can also get "daily reports" posted to your slack channel.
+These reports contain key metrics like:
+
+- Top 5 deployments with most failed requests
+- Top 5 slowest deployments
+
 ## Quick Start
 
 Set up a slack alert channel to receive alerts from proxy.
@@ -21,7 +28,8 @@ Get a slack webhook url from https://api.slack.com/messaging/webhooks
 
 ### Step 2: Update config.yaml 
 
-Let's save a bad key to our proxy
+- Set `SLACK_WEBHOOK_URL` in your proxy env to enable Slack alerts.
+- Just for testing purposes, let's save a bad key to our proxy.
 
 ```yaml
 model_list: 
@@ -34,13 +42,11 @@ general_settings:
     alerting: ["slack"]
     alerting_threshold: 300 # sends alerts if requests hang for 5min+ and responses take 5min+ 
 
+environment_variables:
+    SLACK_WEBHOOK_URL: "https://hooks.slack.com/services/<>/<>/<>"
+    SLACK_DAILY_REPORT_FREQUENCY: "86400"  # 24 hours; Optional: defaults to 12 hours
 ```
 
-Set `SLACK_WEBHOOK_URL` in your proxy env
-
-```shell
-SLACK_WEBHOOK_URL: "https://hooks.slack.com/services/<>/<>/<>"
-```
 
 ### Step 3: Start proxy
 

litellm/integrations/slack_alerting.py

@@ -31,7 +31,8 @@ class LiteLLMBase(BaseModel):
 
 
 class SlackAlertingArgs(LiteLLMBase):
-    daily_report_frequency: int = 12 * 60 * 60  # 12 hours
+    default_daily_report_frequency: int = 12 * 60 * 60  # 12 hours
+    daily_report_frequency: int = int(os.getenv("SLACK_DAILY_REPORT_FREQUENCY", default_daily_report_frequency))
     report_check_interval: int = 5 * 60  # 5 minutes