redis otel tracing + async support for latency routing (#6452)

* docs(exception_mapping.md): add missing exception types

Fixes https://github.com/Aider-AI/aider/issues/2120#issuecomment-2438971183

* fix(main.py): register custom model pricing with specific key

Ensure custom model pricing is registered to the specific model+provider key combination

* test: make testing more robust for custom pricing

* fix(redis_cache.py): instrument otel logging for sync redis calls

ensures complete coverage for all redis cache calls

* refactor: pass parent_otel_span for redis caching calls in router

allows for more observability into what calls are causing latency issues

* test: update tests with new params

* refactor: ensure e2e otel tracing for router

* refactor(router.py): add more otel tracing acrosss router

catch all latency issues for router requests

* fix: fix linting error

* fix(router.py): fix linting error

* fix: fix test

* test: fix tests

* fix(dual_cache.py): pass ttl to redis cache

* fix: fix param

tests/local_testing/test_caching.py

@@ -83,7 +83,9 @@ def test_dual_cache_batch_get_cache():
 
     in_memory_cache.set_cache(key="test_value", value="hello world")
 
-    result = dual_cache.batch_get_cache(keys=["test_value", "test_value_2"])
+    result = dual_cache.batch_get_cache(
+        keys=["test_value", "test_value_2"], parent_otel_span=None
+    )
 
     assert result[0] == "hello world"
     assert result[1] == None

tests/local_testing/test_router.py

@@ -2447,11 +2447,11 @@ async def test_aaarouter_dynamic_cooldown_message_retry_time(sync_mode):
 
         if sync_mode:
             cooldown_deployments = _get_cooldown_deployments(
-                litellm_router_instance=router
+                litellm_router_instance=router, parent_otel_span=None
             )
         else:
             cooldown_deployments = await _async_get_cooldown_deployments(
-                litellm_router_instance=router
+                litellm_router_instance=router, parent_otel_span=None
             )
         print(
             "Cooldown deployments - {}\n{}".format(

tests/local_testing/test_router_cooldowns.py

@@ -242,12 +242,12 @@ async def test_single_deployment_no_cooldowns_test_prod_mock_completion_calls():
             pass
 
     cooldown_list = await _async_get_cooldown_deployments(
-        litellm_router_instance=router
+        litellm_router_instance=router, parent_otel_span=None
     )
     assert len(cooldown_list) == 0
 
     healthy_deployments, _ = await router._async_get_healthy_deployments(
-        model="gpt-3.5-turbo"
+        model="gpt-3.5-turbo", parent_otel_span=None
     )
 
     print("healthy_deployments: ", healthy_deployments)
@@ -351,7 +351,7 @@ async def test_high_traffic_cooldowns_all_healthy_deployments():
     print("model_stats: ", model_stats)
 
     cooldown_list = await _async_get_cooldown_deployments(
-        litellm_router_instance=router
+        litellm_router_instance=router, parent_otel_span=None
     )
     assert len(cooldown_list) == 0
 
@@ -449,7 +449,7 @@ async def test_high_traffic_cooldowns_one_bad_deployment():
     print("model_stats: ", model_stats)
 
     cooldown_list = await _async_get_cooldown_deployments(
-        litellm_router_instance=router
+        litellm_router_instance=router, parent_otel_span=None
     )
     assert len(cooldown_list) == 1
 
@@ -550,7 +550,7 @@ async def test_high_traffic_cooldowns_one_rate_limited_deployment():
     print("model_stats: ", model_stats)
 
     cooldown_list = await _async_get_cooldown_deployments(
-        litellm_router_instance=router
+        litellm_router_instance=router, parent_otel_span=None
     )
     assert len(cooldown_list) == 1
 

tests/router_unit_tests/test_router_helper_utils.py

@@ -440,12 +440,12 @@ def test_update_usage(model_list):
     )
     deployment_id = deployment["model_info"]["id"]
     request_count = router._update_usage(
-        deployment_id=deployment_id,
+        deployment_id=deployment_id, parent_otel_span=None
     )
     assert request_count == 1
 
     request_count = router._update_usage(
-        deployment_id=deployment_id,
+        deployment_id=deployment_id, parent_otel_span=None
     )
 
     assert request_count == 2
@@ -482,7 +482,9 @@ def test_should_raise_content_policy_error(model_list, finish_reason, expected_e
 def test_get_healthy_deployments(model_list):
     """Test if the '_get_healthy_deployments' function is working correctly"""
     router = Router(model_list=model_list)
-    deployments = router._get_healthy_deployments(model="gpt-3.5-turbo")
+    deployments = router._get_healthy_deployments(
+        model="gpt-3.5-turbo", parent_otel_span=None
+    )
     assert len(deployments) > 0
 
 
@@ -756,6 +758,7 @@ def test_track_deployment_metrics(model_list):
             model="gpt-3.5-turbo",
             usage={"total_tokens": 100},
         ),
+        parent_otel_span=None,
     )