litellm fires both failure_handler (sync) and async_failure_handler for
every call. deployment_callback_on_failure was doing a blocking Redis SET
via _set_cooldown_deployments() on every failure, including from async
request paths — stalling the event loop.
Changes:
- CooldownCache.async_add_deployment_to_cooldown(): new method using
await self.cache.async_set_cache()
- async_set_cooldown_deployments(): async counterpart to
_set_cooldown_deployments(), calls the new async cache method
- async_deployment_callback_on_failure(): now also handles cooldown
writes via await async_set_cooldown_deployments()
- deployment_callback_on_failure(): skips the cooldown write for async
callers (detected via CallTypes flags in litellm_params) to avoid the
double Redis write; sync callers keep the existing sync path
- async_routing_strategy_pre_call_checks(): swapped to
await async_set_cooldown_deployments()
Set prometheus_emit_stream_label: true in litellm_settings to emit a
stream label (True/False/None) on litellm_proxy_total_requests_metric.
Opt-in to avoid breaking cardinality on existing deployments.
AsyncHTTPHandler.__del__ was closing httpx clients still in use by
AsyncOpenAI/AsyncAzureOpenAI due to independent cache lifecycles.
Restores standalone httpx client creation for OpenAI/Azure providers.
- Check for both litellm_proxy_failed_requests_metric_total and the deprecated litellm_llm_api_failed_requests_metric_total
- The proxy-level failure hook may not always be called depending on where the exception occurs
- Simplify total_requests check to only verify key fields
Co-authored-by: Cursor <cursoragent@cursor.com>
* litellm_fix_mapped_tests_core: fix test isolation and mock injection issues
## Problem
Four tests in litellm_mapped_tests_core were failing:
1. test_register_model_with_scientific_notation - KeyError due to test isolation issues
2. test_search_uses_registry_credentials - Mock not being called due to incorrect patch path
3. test_send_email_missing_api_key - Real API calls despite mocking
4. test_stream_transformation_error_sync - Mock not effective, real API called
## Solution
### test_register_model_with_scientific_notation
- Use unique model name to avoid conflicts with other tests
- Clear LRU caches before test to prevent stale data
- Clean up model_cost entry after test
### test_search_uses_registry_credentials
- Use patch.object() on the actual base_llm_http_handler instance
- String-based patching for instance methods can fail; direct object patching is more reliable
### test_send_email_missing_api_key
- Directly inject mock HTTP client into logger instance
- This bypasses any caching issues that could cause the fixture mock to be ineffective
### test_stream_transformation_error_sync
- Patch litellm.completion directly instead of the handler module's litellm reference
- This ensures the mock is effective regardless of import order
## Regression
These tests were affected by LRU caching added in #19606 and HTTP client caching.
* fix(test): use patch.object for container API tests to fix mock injection
## Problem
test_retrieve_container_basic tests were failing because mocks weren't
being applied correctly. The tests used string-based patching:
patch('litellm.containers.main.base_llm_http_handler')
But base_llm_http_handler is imported at module level, so the mock wasn't
intercepting the actual handler calls, resulting in real HTTP requests
to OpenAI API.
## Solution
Use patch.object() to directly mock methods on the imported handler
instance. Import base_llm_http_handler in the test file and patch like:
patch.object(base_llm_http_handler, 'container_retrieve_handler', ...)
This ensures the mock is applied to the actual object being used,
regardless of import order or caching.
* fix(test): add missing Prometheus metric labels to test_proxy_failure_metrics
Add client_ip, user_agent, model_id labels to expected metric patterns.
These labels were added in PRs #19717 and #19678 but test wasn't updated.
* fix(test_resend_email): use direct mock injection for all email tests
Extend the mock injection pattern used in test_send_email_missing_api_key
to all other tests in the file:
- test_send_email_success
- test_send_email_multiple_recipients
Instead of relying on fixture-based patching and respx mocks which can
fail due to import order and caching issues, directly inject the mock
HTTP client into the logger instance. This ensures mocks are always used
regardless of test execution order.
* fix(test): use patch.object for image_edit and vector_store tests
- test_image_edit_merges_headers_and_extra_headers: import base_llm_http_handler
and use patch.object instead of string path patching
- test_search_uses_registry_credentials: import module and patch via
module.base_llm_http_handler to ensure we patch the right instance
---------
Co-authored-by: Ishaan Jaff <ishaanjaffer0324@gmail.com>
The user_id field 'default_user_id' is being masked to '*******_user_id'
in prometheus metrics for privacy. Updated test expectations to match
the actual behavior.
Co-authored-by: Cursor <cursoragent@cursor.com>
## Problem
Tests using mocked HTTP clients were hitting real APIs because:
1. HTTP client cache was returning previously cached real clients
2. isinstance checks failed due to module identity issues from sys.path
### Tests affected:
- test_send_email_missing_api_key
- test_send_email_multiple_recipients (resend & sendgrid)
- test_search_uses_registry_credentials
- test_vector_store_create_with_simple_provider_name
- test_vector_store_create_with_provider_api_type
- test_vector_store_create_with_ragflow_provider
- test_image_edit_merges_headers_and_extra_headers
- test_retrieve_container_basic (container API tests)
## Solution
1. Add clear_client_cache fixture (autouse=True) to clear
litellm.in_memory_llm_clients_cache before each test
2. Fix isinstance checks to use type name comparison
(avoids module identity issues from sys.path.insert)
## Why not disable_aiohttp_transport
The default transport is aiohttp, so tests should work with it.
Clearing the cache ensures mocks are used instead of cached real clients.
## Regression
PR #19829 (commit f95572e3ed) added @respx.mock but cached clients
from earlier tests were being reused, bypassing the mocks.
Co-authored-by: shin-bot-litellm <shin-bot-litellm@users.noreply.github.com>
The test had prompt_tokens=1000 but the sum of token details was 1150
(text=700 + audio=100 + cached=200 + cache_creation=150).
This triggered the double-counting detection logic which recalculated
text_tokens to 550, causing the assertion to fail.
Fixed by setting prompt_tokens=1150 to match the sum of details.
The test uses an invalid API key to verify that batch_completion returns
exceptions rather than raising them. However, depending on network conditions,
the error may be:
- AuthenticationError: API properly rejected the invalid key
- InternalServerError: Connection error occurred before API could respond
Both are valid outcomes for this test case.
Co-authored-by: shin-bot-litellm <shin-bot-litellm@users.noreply.github.com>
- batch_endpoints.py: Pop batch_id from data before creating CancelBatchRequest
to avoid duplicate batch_id when data already contains it from earlier cast
- files_endpoints.py: Pop file_id from data before calling afile_retrieve
to avoid duplicate file_id when data was initialized with {"file_id": file_id}
- test_claude_agent_sdk.py: Disable bedrock-nova-premier test as it requires
an inference profile for on-demand throughput (AWS limitation)
Fixes: e2e_openai_endpoints tests (test_batches_operations, test_file_operations)
Fixes: proxy_e2e_anthropic_messages_tests (nova-premier model skip)
The proxy now adds metadata to the request body during processing.
Updated test to compare fields individually and strip metadata from
body comparison.
Fixes litellm_proxy_unit_testing_part2 CI failure.
- Add test_router_acancel_batch.py with mock test for router.acancel_batch()
- Add _acancel_batch to ignored list (internal helper tested via public API)
Fixes CI failure in check_code_and_doc_quality job
hosted_vllm no longer uses the OpenAI client, so these tests
that mock the OpenAI client are not applicable to hosted_vllm.
Removes hosted_vllm from:
- test_openai_compatible_custom_api_base
- test_openai_compatible_custom_api_video
This fixes the failing litellm_mapped_enterprise_tests (metrics/logging) job.
Recent commits added new labels to several Prometheus metrics (model_id, client_ip, user_agent)
but the test assertions weren't fully updated to expect these new labels.
Tests fixed:
- test_async_post_call_failure_hook
- test_async_log_failure_event
- test_increment_token_metrics
- test_log_failure_fallback_event
- test_set_latency_metrics
- test_set_llm_deployment_success_metrics
Labels added to test assertions:
- model_id for token metrics (litellm_tokens_metric, litellm_input_tokens_metric, litellm_output_tokens_metric)
- model_id for latency metrics (litellm_llm_api_latency_metric)
- model_id for remaining requests/tokens metrics
- model_id for fallback metrics
- model_id for overhead latency metric
- client_ip and user_agent for deployment failure/total/success responses
- client_ip and user_agent for proxy failed/total requests metrics
