mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-29 06:33:28 +00:00
* fix(router): stop concurrent deployment deletes from crashing deployment lookups delete_deployment and upsert_deployment shrink model_list before they repair the index maps, so a lookup on another thread in that window read a stale index: IndexError, or the neighbouring deployment and its credentials. An add racing a delete could corrupt the index for good. One RLock now covers model_list and the three index maps. Delete, upsert, add and the set_model_list reset change them under it, and every lookup that turns an index into a deployment reads under it. Fixes #45242 * test(router): release the held delete when the read reaches the index lock The race tests released the held write after a fixed 0.2 s, so a reader the runner stalled past it read the repaired list and passed on unfixed code. The read now runs on its own thread and the write stays held until the read returns or reaches the router's index lock; the 10 s bound only fails a broken synchronization. The integration test gains the audit cells: chat, streaming, Responses and Anthropic messages calls (sync and async, by model name and team public name) during a held delete, the same for upsert, add and set_model_list, a request for the deployment being deleted, a provider error, a double delete, and thread and async bursts during upsert churn. * test(router): cover the early return of a delete and the team filter of get_model_ids delete_deployment now returns None from inside the index lock, and get_model_ids by name reads its deployments through _deployments_named; pin both: deleting an unknown id returns None and leaves the list alone, and exclude_team_models drops only the team deployment. * fix(router): read deployments by name under the index lock outside the Router The response polling provider check, the complexity router's adaptive pool and the savings baseline turned the name index into list positions themselves, so a concurrent delete could still hand them a stale slot. They now read through Router.deployments_named, which takes the deployment index lock. * fix(router): type the locked deployment lookups as DeploymentTypedDict deployments_named and _team_deployments_named returned tuple[Any, ...]. They now return DeploymentTypedDict rows, which is what model_list holds. The adaptive pool setup in the Router and the complexity router drop their branches for pydantic Deployment rows, which these lookups never return, and should_include_deployment accepts the typed row. --------- Co-authored-by: yucheng-berri <293926326+yucheng-berri@users.noreply.github.com>
216 lines
15 KiB
Python
216 lines
15 KiB
Python
import ast
|
|
import os
|
|
|
|
|
|
def get_function_names_from_file(file_path):
|
|
"""
|
|
Extracts all function names from a given Python file.
|
|
"""
|
|
with open(file_path, "r", encoding="utf-8") as file:
|
|
tree = ast.parse(file.read())
|
|
|
|
function_names = []
|
|
|
|
for node in tree.body:
|
|
if isinstance(node, (ast.FunctionDef, ast.AsyncFunctionDef)):
|
|
# Top-level functions
|
|
function_names.append(node.name)
|
|
elif isinstance(node, ast.ClassDef):
|
|
# Functions inside classes
|
|
for class_node in node.body:
|
|
if isinstance(class_node, (ast.FunctionDef, ast.AsyncFunctionDef)):
|
|
function_names.append(class_node.name)
|
|
|
|
return function_names
|
|
|
|
|
|
def get_all_functions_called_in_tests(base_dir):
|
|
"""
|
|
Returns a set of function names that are called in test functions
|
|
inside the unit test directories,
|
|
specifically in files containing the word 'router'.
|
|
"""
|
|
called_functions = set()
|
|
test_dirs = ["unit"]
|
|
|
|
for test_dir in test_dirs:
|
|
dir_path = os.path.join(base_dir, test_dir)
|
|
if not os.path.exists(dir_path):
|
|
print(f"Warning: Directory {dir_path} does not exist.")
|
|
continue
|
|
|
|
print("dir_path: ", dir_path)
|
|
for root, _, files in os.walk(dir_path):
|
|
for file in files:
|
|
file_path = os.path.join(root, file)
|
|
if file.endswith(".py") and (
|
|
"router" in os.path.relpath(file_path, dir_path).lower() or test_dir == "unit"
|
|
):
|
|
print("file: ", file)
|
|
with open(file_path, "r", encoding="utf-8") as f:
|
|
try:
|
|
tree = ast.parse(f.read())
|
|
except SyntaxError:
|
|
print(f"Warning: Syntax error in file {file_path}")
|
|
continue
|
|
if file == "test_router_validate_fallbacks.py":
|
|
print(f"tree: {tree}")
|
|
for node in ast.walk(tree):
|
|
if isinstance(node, ast.Call) and isinstance(node.func, ast.Name):
|
|
called_functions.add(node.func.id)
|
|
elif isinstance(node, ast.Call) and isinstance(node.func, ast.Attribute):
|
|
called_functions.add(node.func.attr)
|
|
|
|
return called_functions
|
|
|
|
|
|
def get_functions_from_router(file_path):
|
|
"""
|
|
Extracts all functions defined in router.py.
|
|
"""
|
|
return get_function_names_from_file(file_path)
|
|
|
|
|
|
ignored_function_names = [
|
|
"_acancel_batch",
|
|
"_acreate_batch",
|
|
"_acreate_file",
|
|
"__init__",
|
|
"avector_store_create", # Tested via proxy vector_store_endpoints (files lack "router" in name)
|
|
"_override_vector_store_methods_for_router", # No-op placeholder, called during Router init
|
|
"_merge_tools_from_deployment", # Tested indirectly via _update_kwargs_with_deployment (test files lack "router" in name)
|
|
"_invalidate_access_groups_cache", # Tested indirectly via set_model_list, upsert_model etc. (test files lack "router" in name)
|
|
"has_buffered_provider_output", # Property, so its reads in test_router.py are never an ast.Call
|
|
"chunks", # Property on FallbackAwareAnthropicMessagesStream, so its reads in tests are never an ast.Call
|
|
"messages", # Property on FallbackAwareAnthropicMessagesStream, so its reads in tests are never an ast.Call
|
|
"model", # Property on FallbackAwareAnthropicMessagesStream, so its reads in tests are never an ast.Call
|
|
"_routing_groups", # Property getter and setter reads are never ast.Call nodes
|
|
"_request_header", # Tested through Claude Code session routing in test_router.py
|
|
"_claude_code_session_router_cache_key", # Tested through Claude Code session routing in test_router.py
|
|
"_delete_claude_code_session_router_binding", # Tested through Redis cleanup failure in test_router.py
|
|
"_resolve_claude_code_session_router", # Tested through Claude Code session routing in test_router.py
|
|
"_get_claude_code_session_router_binding", # Tested through the two-worker session routing test in test_router.py
|
|
"_apply_updated_routing_strategy_args", # Tested via update_settings in test_lowest_latency.py (file lacks "router" in name)
|
|
"_wait_for_scheduler_turn", # Tested through prioritized acompletion and atext_completion in test_router.py
|
|
"arm_routing_read_prefetch", # Tested in tests/unit/caching/test_request_redis_batch_pre_call.py (file lacks "router" in name)
|
|
"_deployment_params_base_model", # Tested through log_retry in test_router_helper_utils.py
|
|
"_configured_model_info", # Tested through get_configured_service_tiers in test_router.py
|
|
"_routable_deployments", # Tested through get_configured_service_tiers and get_routable_upstream_model in test_router.py
|
|
"_async_get_available_deployment", # Body of the `route {model}` phase wrapper, exercised through async_get_available_deployment in test_router.py
|
|
"_async_get_available_deployment_for_pass_through", # Same, through async_get_available_deployment_for_pass_through in test_router.py
|
|
"_embedding",
|
|
"_aembedding",
|
|
"_stream_pre_content_error", # Tested through the non-retriable retry error tests in test_router.py
|
|
"_narrow_to_video_deployment", # Tested through video_status and avideo_status polls in test_router_provider_endpoints.py
|
|
"_video_call_pin", # Tested through the video_status pin tests and the chat video_id test in test_router_provider_endpoints.py
|
|
"_deployment_num_retries", # Tested through the deployment num_retries mid-stream budget test in test_router.py
|
|
"_request_fallback_list", # Tested through every mid-stream retry test in test_router.py
|
|
"_request_model_group", # Tested through test_anthropic_messages_retry_budget_precedence_direct_call
|
|
"_mid_stream_retry_trigger", # Tested through the retry policy mid-stream budget test in test_router.py
|
|
"_anthropic_messages_group_retry_policy", # Tested through the retry budget precedence test in test_router.py
|
|
"_anthropic_messages_resolved_retry_policy", # Tested through the malformed retry policy tests in test_router.py
|
|
"_anthropic_messages_plain_retry_budget", # Tested through the retry budget precedence test in test_router.py
|
|
"_stream_should_retry", # Tested through every mid-stream retry test in test_router.py
|
|
"_astream_retry_same_group", # Tested through the dropped-before-content retry tests in test_router.py
|
|
"_aanthropic_messages_yield_recovered", # Tested through every mid-stream retry and fallback test in test_router.py
|
|
"_anthropic_messages_policy_retries", # Tested through the retry budget precedence test in test_router.py
|
|
"_get_wildcard_deployments", # Tested through the get_model_list_of_routed_group wildcard test in test_router.py
|
|
"_prompt_management_acompletion", # Tested through the scheduled prompt-management model test in test_router.py
|
|
"_request_deployment_params", # Tested through the team-route and deployment-id drop_params priority tests in test_router.py
|
|
"split_request_priority", # Tested through the acompletion, atext_completion and queue route priority tests
|
|
"_pinned_deployment", # Tested through the deployment-id scheduler test in test_router.py
|
|
"_is_fallback_hop", # Tested through the order fallback hop tests in test_router_order_fallback.py
|
|
"_deployment_that_just_failed", # Tested through the same-boundary hop test in test_router_order_fallback.py
|
|
"_merged_tools", # Tested through the _update_kwargs_with_deployment tools merge tests in test_router.py
|
|
"_drops_malformed_request_tools", # Tested through the _update_kwargs_with_deployment tools merge tests in test_router.py
|
|
"_collect_ptu_share_teams", # Cached behind team_ptu_ceiling, tested through the live model list test in test_router.py
|
|
"_resolve_team_ptu_ceiling", # Cached behind team_ptu_ceiling, tested through the live model list test in test_router.py
|
|
"_register_continuation_prefill_check", # Tested through the continuation tests in test_router.py
|
|
"_set_mid_stream_fallback_continuation", # Tested through the update_settings continuation test in test_router.py
|
|
"_is_flag_enabled", # Tested through the string-flag continuation tests in test_router.py
|
|
"_params_sent_to_deployment", # Tested through the configured-constraint continuation tests in test_router.py
|
|
"_names_deployment_that_cannot_continue", # Tested through the deployment-id continuation tests in test_router.py
|
|
"_set_retry_policy", # Tested through update_settings in test_router_retry_policy_update.py
|
|
"_resolve_pre_call_check_input_tokens", # Tested through the token-count failure pre-call check tests in test_router.py
|
|
"_params_neither_dropped_nor_allowed", # Tested through the allowed-or-dropped param pre-call check tests in test_router.py
|
|
"_deployment_rejects_params", # Tested through the unsupported-param pre-call check tests in test_router.py
|
|
"_deployment_lacks_params", # Tested through the unsupported-param pre-call check tests in test_router.py
|
|
"_pre_call_check_requested_chat_params", # Tested through the unsupported-param pre-call check tests in test_router.py
|
|
"_deployment_supports_param", # Tested through the deployment-capability pre-call check tests in test_router.py
|
|
"_capability_model_and_provider", # Tested through the authenticating-provider pre-call check test in test_router.py
|
|
"_route_capability_pin", # Exercised through Decisions routing tests in test_router.py
|
|
"_deployment_serves_route", # Exercised through mixed-group Decisions tests in test_router.py
|
|
"_without_deployments_that_cannot_serve_route", # Exercised through Decisions routing tests in test_router.py
|
|
"_exact_base_model_max_input_tokens", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
"_base_model_widened_window", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
"_provider_model_id", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
"_names_route_model", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
"_declared_base_model", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
"async_admit_deployment_turn", # Bound as the WebSocket turn_admission partial, tested through it in test_router_helper_utils.py
|
|
"_deployment_as_configured_now", # Read by async_admit_deployment_turn, tested through turn_admission in test_router_helper_utils.py
|
|
"_admit_turn_through_routing_strategy", # Run by async_admit_deployment_turn, tested through turn_admission in test_router_helper_utils.py
|
|
"_callback_selectors", # Run by discard, tested through the discard tests in test_router_helper_utils.py
|
|
"bind_to_request_logging", # Run by the proxy user_config route, tested through the user_config tests in test_route_llm_request.py
|
|
"serve_through_request_logging", # Run by the proxy user_config route, tested through the user_config tests in test_route_llm_request.py
|
|
"_request_adaptive_router_hooks", # Run by serve_through_request_logging, tested through the user_config adaptive test
|
|
"_release_global_callbacks", # Run by discard, tested through the discard tests in test_router_helper_utils.py
|
|
"_routing_callbacks", # Read by the pre-call checks, tested through the user_config limit tests in test_route_llm_request.py
|
|
"_request_policies_not_set_globally", # Read by bind_to_request_logging, tested through the user_config budget test in test_route_llm_request.py
|
|
"_optional_loggers", # Read by discard and serve_through_request_logging, tested through the budget limiter discard test
|
|
"retire_routing_state", # Run by discard, tested through the budget limiter discard test
|
|
"_slack_alerting_loggers", # Run by discard, tested through the Slack alerting discard test in test_router_helper_utils.py
|
|
"_adaptive_router_hooks", # Run by discard, tested through the adaptive router discard tests in test_router_helper_utils.py
|
|
"_global_callback_candidates", # Run by discard, tested through the survivor discard tests in test_router_helper_utils.py
|
|
"without_deployments_that_cannot_batch", # Tested through the Azure and OpenAI mixed-group acreate_batch tests in test_router.py
|
|
"batch_create_routing", # Tested through the mixed-group acreate_batch tests in test_router.py
|
|
"is_batch_create_routing", # Tested through the chat-after-batch-create test in test_router.py
|
|
"is_batch_capable", # Tested through the mixed-group acreate_batch and model group info tests in test_router.py
|
|
"deployments_the_caller_may_use", # Tested through the usage-based-routing team batch create test in test_router.py
|
|
"deployment_supported_endpoints", # Tested through the model group info batch endpoint tests in test_router.py
|
|
"_endpoint_paths", # Read by deployment_supported_endpoints, tested through the malformed endpoint list group info test in test_router.py
|
|
"_deployment_field", # Read by is_batch_capable, tested through the mixed-group tests in test_router.py
|
|
"_deployment_provider", # Read by without_deployments_that_cannot_batch, tested through the mixed-group tests in test_router.py
|
|
"_cannot_run_batches", # Tested through the Azure and OpenAI mixed-group acreate_batch tests in test_router.py
|
|
"_team_deployments_named", # Tested through the team lookups in the concurrent delete tests in test_router.py
|
|
]
|
|
|
|
|
|
def main():
|
|
router_file = [
|
|
"./litellm/router.py",
|
|
"./litellm/router_utils/batch_utils.py",
|
|
"./litellm/router_utils/pattern_match_deployments.py",
|
|
]
|
|
# router_file = [
|
|
# "../../litellm/router.py",
|
|
# "../../litellm/router_utils/pattern_match_deployments.py",
|
|
# "../../litellm/router_utils/batch_utils.py",
|
|
# ] ## LOCAL TESTING
|
|
tests_dir = "./tests/" # Update this path if your tests directory is located elsewhere
|
|
# tests_dir = "../../tests/" # LOCAL TESTING
|
|
|
|
router_functions = []
|
|
for file in router_file:
|
|
router_functions.extend(get_functions_from_router(file))
|
|
print("router_functions: ", router_functions)
|
|
called_functions_in_tests = get_all_functions_called_in_tests(tests_dir)
|
|
untested_functions = [fn for fn in router_functions if fn not in called_functions_in_tests]
|
|
|
|
if untested_functions:
|
|
all_untested_functions = []
|
|
for func in untested_functions:
|
|
if func not in ignored_function_names:
|
|
all_untested_functions.append(func)
|
|
untested_perc = (len(all_untested_functions)) / len(router_functions)
|
|
print("untested_perc: ", untested_perc)
|
|
if untested_perc > 0:
|
|
print("The following functions in router.py are not tested:")
|
|
raise Exception(
|
|
f"{untested_perc * 100:.2f}% of functions in router.py are not tested: {all_untested_functions}"
|
|
)
|
|
else:
|
|
print("All functions in router.py are covered by tests.")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|