Feat add retention config (#10815)

* add function to check config flag

* added unit tests

* convert to seconds support

* added in settings.md

* Updated config_settings.md

* remove extra point

* change config var

* resolve conflict

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

@@ -1,6 +1,5 @@
 # All settings
 
-
 ```yaml
 environment_variables: {}
 
@@ -95,6 +94,7 @@ general_settings:
   allowed_routes: ["route1", "route2"]  # list of allowed proxy API routes - a user can access. (currently JWT-Auth only)
   key_management_system: google_kms  # either google_kms or azure_kms
   master_key: string
+  maximum_spend_logs_retention_period: 30d # The maximum time to retain spend logs before deletion.
 
   # Database Settings
   database_url: string
@@ -211,6 +211,7 @@ general_settings:
 | enable_oauth2_proxy_auth | boolean | (Enterprise Feature) If true, enables oauth2.0 authentication |
 | forward_openai_org_id | boolean | If true, forwards the OpenAI Organization ID to the backend LLM call (if it's OpenAI). |
 | forward_client_headers_to_llm_api | boolean | If true, forwards the client headers (any `x-` headers) to the backend LLM call |
+| maximum_spend_logs_retention_period               | str                   | Used to set the max retention time for spend logs in the db, after which they will be auto-purged                                                                                                                                                                                                                             |
 
 ### router_settings - Reference
 

litellm/proxy/db/db_transaction_queue/spend_log_cleanup.py

@@ -0,0 +1,39 @@
+"""
+Handles checking if spend logs should be deleted based on maximum retention period
+"""
+
+from typing import Optional, Union
+
+from litellm._logging import verbose_proxy_logger
+from litellm.litellm_core_utils.duration_parser import duration_in_seconds
+from litellm.proxy.proxy_server import general_settings
+
+
+def _should_delete_spend_logs() -> bool:
+    """
+    Checks if the Pod should delete spend logs based on maximum retention period
+
+    This setting enables automatic deletion of old spend logs to manage database size.
+    The maximum_spend_logs_retention_period can be specified in:
+    - Days (e.g., "30d")
+    - Hours (e.g., "24h")
+    - Minutes (e.g., "60m")
+    - Seconds (e.g., "3600s" or just "3600")
+    """
+    _maximum_spend_logs_retention_period: Optional[Union[int, str]] = general_settings.get(
+        "maximum_spend_logs_retention_period", None
+    )
+    
+    if _maximum_spend_logs_retention_period is None:
+        return False
+
+    try:
+        if isinstance(_maximum_spend_logs_retention_period, int):
+            _maximum_spend_logs_retention_period = str(_maximum_spend_logs_retention_period)
+        duration_in_seconds(_maximum_spend_logs_retention_period)
+        return True
+    except ValueError as e:
+        verbose_proxy_logger.error(
+            f"Invalid maximum_spend_logs_retention_period value: {_maximum_spend_logs_retention_period}, error: {str(e)}"
+        )
+        return False 

tests/proxy_unit_tests/test_spend_log_cleanup.py

@@ -0,0 +1,56 @@
+"""
+Test cases for spend log cleanup functionality
+"""
+
+import pytest
+from litellm.proxy.db.db_transaction_queue.spend_log_cleanup import _should_delete_spend_logs
+from litellm.proxy.proxy_server import general_settings
+
+
+@pytest.mark.asyncio
+async def test_should_delete_spend_logs():
+    """
+    Test the _should_delete_spend_logs function with various scenarios
+    """
+    # Test case 1: No maximum_spend_logs_retention_period set
+    general_settings.clear()
+    assert _should_delete_spend_logs() is False
+
+    # Test case 2: Valid integer maximum_spend_logs_retention_period (in seconds)
+    general_settings["maximum_spend_logs_retention_period"] = 3600
+    assert _should_delete_spend_logs() is True
+
+    # Test case 3: Valid duration string - days
+    general_settings["maximum_spend_logs_retention_period"] = "30d"
+    assert _should_delete_spend_logs() is True
+
+    # Test case 4: Valid duration string - hours
+    general_settings["maximum_spend_logs_retention_period"] = "24h"
+    assert _should_delete_spend_logs() is True
+
+    # Test case 5: Valid duration string - minutes
+    general_settings["maximum_spend_logs_retention_period"] = "60m"
+    assert _should_delete_spend_logs() is True
+
+    # Test case 6: Valid duration string - seconds
+    general_settings["maximum_spend_logs_retention_period"] = "3600s"
+    assert _should_delete_spend_logs() is True
+
+    # Test case 7: Valid duration string - weeks
+    general_settings["maximum_spend_logs_retention_period"] = "1w"
+    assert _should_delete_spend_logs() is True
+
+    # Test case 8: Valid duration string - months
+    general_settings["maximum_spend_logs_retention_period"] = "1mo"
+    assert _should_delete_spend_logs() is True
+
+    # Test case 9: Invalid duration string
+    general_settings["maximum_spend_logs_retention_period"] = "invalid"
+    assert _should_delete_spend_logs() is False
+
+    # Test case 10: None value
+    general_settings["maximum_spend_logs_retention_period"] = None
+    assert _should_delete_spend_logs() is False
+
+    # Clean up
+    general_settings.clear()