mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-27 01:22:18 +00:00
* ci: run the unit_selection.sh shard files on every event instead of only fork pull requests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci: rename fork-flag to unit-flag now that it applies on every event * test: move tests/test_litellm root and small trees into tests/unit Pure renames, no content changes. Follow-up commits in this PR fix references, merge the three files that already existed in tests/unit, keep live-provider tests in tests/test_litellm and wire CI. * test: carry tests/test_litellm conftest isolation into tests/unit Callback lists, routing fallbacks, cached HTTP clients, logger state, AWS, proxy-URL and keychain env, and session-end client cleanup now reset for unit tests too. The environment isolation owns its MonkeyPatch so a test's own monkeypatch is undone before the model-cost teardown runs. * test: merge, split and prune the moved root and small-tree tests Merge batches/test_batch_utils.py and the chat_completions and messages dispatch tests into the files that already existed in tests/unit. Keep the live Gemini interactions tests, the async image-fetch format test and the OpenAI embedding scorer test in tests/test_litellm since they need real network or keys. Put test_router.py under tests/unit/test_router so the existing package no longer shadows it. Delete eight tests the audit found superseded by stronger ones kept in this move. * ci: run the moved root and small-tree tests under their legacy flags Add the misc and responses-caching-types flags to unit_selection.sh and CircleCI, extend enterprise-routing and mcp-integration, and point the legacy GHA shards, Makefile, redis-compat workflow, merge smoke manifest and change classifier at the new paths. * test: make the new tests/unit directories packages tests/unit/test_package_layout.py requires every directory to carry an __init__.py, and without one the moved and retained test_litellm_responses_bridge.py modules collide on import. * test: scope the unit socket block to tests/unit in shared sessions The GHA shards collect the legacy test-path and the unit selection in one pytest session. The unit conftest's loopback-only block leaked into legacy modules that reach the network at import. The legacy conftest now lifts the restriction at collect and setup time, and the unit conftest re-applies it when collecting its own modules. * test: move tests/test_litellm/llms into tests/unit/llms Rename-only. Moves the provider tests and the fine-tuning fixtures they load, mirroring the old paths. Follow-up commits merge, split and wire them. * test: merge, split and prune the moved llms tests Merges the Databricks chat transformation tests into the existing unit file, keeps the tests that need real keys or the network in tests/test_litellm, deletes the audited tests a stronger unit test already covers, and points imports at tests.unit.llms. * ci: run the moved llms tests under their legacy flags The Vertex AI and All Other Providers shards keep their legacy test-path for the retained files and add the llm-vertex-ai and llm-other-providers unit selections. CircleCI gets matching unit jobs. * test: make the tests/unit/llms directories packages Adds __init__.py to the moved dirs and drops the legacy ones whose directories no longer hold tests. * test: drop script runners and path hacks the llms split left dangling The __main__ runners in the split openai_like files and the Databricks e2e runner called tests that now live in the other half of the split or were deleted. The retained legacy halves also no longer need sys.path edits. * test: give the shard-script tests their own GITHUB_OUTPUT They only passed where the runner set it. The CircleCI unit job's env allowlist drops it, so the script's redirect failed there. * test: point the router and module-deletion checks at tests/unit router_code_coverage and code_qa_check_tests only searched tests/test_litellm, so the moved router tests no longer counted. The two silent-experiment tests the audit deleted were the only direct callers of those methods; they are replaced with tests that assert the forwarded shadow request and the recursion guard. * test: move tests/test_litellm integrations and secret_managers into tests/unit Rename-only. Mirrors the old paths, including the directory conftests and the prompt and JSON fixtures. Follow-up commits prune and wire them. * test: prune and repoint the moved integrations tests Deletes the 7 audited tests a stronger test in the same tree already covers, imports the TLS sink helpers from their new conftest path, and restores os.environ after each integrations test. Some presets write OTEL_EXPORTER_OTLP_HEADERS straight into os.environ, and without the legacy tree's test ordering that header leaked into the AgentOps tests. * ci: run the moved integrations tests under their legacy flag The integrations GHA shard and a new CircleCI job run the integrations unit selection. secret_managers joins the misc selection. * docs: point integrations and secret_managers references at tests/unit * test: make the moved integrations directories packages * test: keep the Databricks manual e2e runner and fix the SageMaker Nova run path The Databricks e2e file is a manual script whose main() calls the tests that were pruned, so pruning them broke the documented run. It is back to its main version. The SageMaker Nova docstring now points at the file's real location in tests/local_testing. * test: move tests/test_litellm core utils, routing, responses, caching and rust_bridge into tests/unit Rename-only. Mirrors the old paths, including fixtures, the stubtest config and the native-route wheel script. Two files that collide with existing unit files are merged in a follow-up commit. * test: merge, prune and repoint the moved core, routing, responses, caching and rust_bridge tests Merges the two files that collided with existing unit files, folding the legacy extra case into test_is_chat_completion_cached_dict, and deletes the 9 audited tests a stronger test in the same file already covers. Keeps what needs the network in tests/test_litellm: test_tokenizers pulls a tokenizer from the Hugging Face hub, and the gpt2 and r50k_base tokenizer cases download their BPE files. The unit core_utils conftest points TIKTOKEN_CACHE_DIR at litellm's bundled encodings so the rest never depend on import order to stay offline, and FakeSecretVault moves to a shared module so both trees can build it. * ci: run the moved core, routing, responses, caching and rust_bridge tests under their flags core_utils gets a core-utils flag and CircleCI job, and its GHA shard keeps the legacy path for the retained network tests. router_utils and router_strategy join enterprise-routing, responses joins responses-caching-types (minus responses/mcp, which mcp-integration owns), caching joins caching-local and rust_bridge joins misc. The redis-compat, test-rust, stubtest and merge-smoke paths follow the move. * docs: point the Rust crate references at tests/unit * test: make the moved core, routing and rust_bridge directories packages * test: keep the no-loop DualCache batch_get_cache regression test It runs the sync path outside any event loop, which the inside-loop test cannot, so a change that picks the Redis client by loop state would only show up there. * test: keep the job's UNIT_FLAG out of the shard-script tests * fix(url_utils): block 192.0.0.0/24 on every Python patch release * test: move the new budget limiter tests into tests/unit/router_strategy * test: move the new sentry scrubbing tests into tests/unit/litellm_core_utils * test: move the new zerobus tests into tests/unit/integrations * test: make tests/unit/integrations/zerobus a package * test: load litellm's own tiktoken cache setup once instead of resetting it per test --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
410 lines
14 KiB
Python
410 lines
14 KiB
Python
"""Regression: a single cluster node's ConnectionError/TimeoutError must reset only that
|
|
node's connections, not tear down the whole cluster client for every other concurrent
|
|
caller. Live confirmation against a real 3-master local cluster (pausing one node with
|
|
CLIENT PAUSE) showed 100% of concurrent commands to the other two, untouched nodes
|
|
stalling for the full pause duration before this fix, and zero after -- these tests pin
|
|
the same behavior at the unit level so it can run without a live Redis Cluster."""
|
|
|
|
import asyncio
|
|
from typing import TYPE_CHECKING
|
|
from unittest.mock import AsyncMock, Mock, patch
|
|
|
|
import pytest
|
|
from redis.exceptions import (
|
|
AskError,
|
|
BusyLoadingError,
|
|
ClusterDownError,
|
|
ClusterError,
|
|
MaxConnectionsError,
|
|
MovedError,
|
|
TryAgainError,
|
|
)
|
|
from redis.exceptions import (
|
|
ConnectionError as RedisConnectionError,
|
|
)
|
|
from redis.exceptions import (
|
|
TimeoutError as RedisTimeoutError,
|
|
)
|
|
|
|
from litellm.caching.redis_cluster_node_isolation import (
|
|
get_litellm_async_redis_cluster_class,
|
|
)
|
|
|
|
if TYPE_CHECKING:
|
|
from redis.asyncio.cluster import RedisCluster as _AsyncRedisClusterType
|
|
|
|
|
|
class _NodeClassWithPerConnectionRecovery:
|
|
def update_active_connections_for_reconnect(self) -> None: ...
|
|
|
|
|
|
class _NodeClassWithoutPerConnectionRecovery:
|
|
pass
|
|
|
|
|
|
class _FakeClusterNode:
|
|
def __init__(self, name: str, raises: Exception | None = None, response: object = None) -> None:
|
|
self.name = name
|
|
|
|
async def execute_command(*args: object, **kwargs: object) -> object:
|
|
await asyncio.sleep(0)
|
|
if raises is not None:
|
|
raise raises
|
|
return response
|
|
|
|
self.execute_command = AsyncMock(side_effect=execute_command)
|
|
self.disconnect = AsyncMock()
|
|
|
|
|
|
class _Fake8xRedisCluster:
|
|
def __init__(self) -> None:
|
|
self._initialize = False
|
|
|
|
async def _execute_command(
|
|
self, target_node: _FakeClusterNode, *args: object, **kwargs: object
|
|
) -> object:
|
|
try:
|
|
return await target_node.execute_command(*args, **kwargs)
|
|
except (RedisConnectionError, RedisTimeoutError):
|
|
self._initialize = True
|
|
await asyncio.sleep(0)
|
|
raise
|
|
|
|
async def aclose(self) -> None:
|
|
self._initialize = True
|
|
|
|
|
|
class _FakeNodesManager:
|
|
def __init__(self, node_to_return: _FakeClusterNode) -> None:
|
|
self._moved_exception: object = None
|
|
self._node_to_return = node_to_return
|
|
|
|
def get_node_from_slot(
|
|
self, slot: int, read_from_replicas: bool, load_balancing_strategy: object
|
|
) -> _FakeClusterNode:
|
|
return self._node_to_return
|
|
|
|
|
|
def _build_cluster_instance() -> "_AsyncRedisClusterType":
|
|
cluster_cls = get_litellm_async_redis_cluster_class(
|
|
cluster_node_class=_NodeClassWithoutPerConnectionRecovery
|
|
)
|
|
instance = cluster_cls.__new__(cluster_cls)
|
|
instance.RedisClusterRequestTTL = 1
|
|
instance.reinitialize_counter = 0
|
|
instance.reinitialize_steps = 5
|
|
instance.read_from_replicas = False
|
|
instance.load_balancing_strategy = None
|
|
instance.aclose = AsyncMock()
|
|
return instance
|
|
|
|
|
|
def _build_8x_cluster_instance() -> _Fake8xRedisCluster:
|
|
cluster_cls = get_litellm_async_redis_cluster_class(
|
|
cluster_node_class=_NodeClassWithPerConnectionRecovery,
|
|
base_cluster_class=_Fake8xRedisCluster,
|
|
)
|
|
return cluster_cls()
|
|
|
|
|
|
def test_unverified_redis_version_logs_warning(caplog: pytest.LogCaptureFixture) -> None:
|
|
import redis
|
|
|
|
with patch.object(redis, "__version__", "8.0.1"):
|
|
get_litellm_async_redis_cluster_class(cluster_node_class=_NodeClassWithoutPerConnectionRecovery)
|
|
|
|
assert "not in the set this cluster-teardown-storm fix was verified against" in caplog.text
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_single_timeout_does_not_request_topology_reinit() -> None:
|
|
error = RedisTimeoutError("timeout")
|
|
target_node = _FakeClusterNode("node-a")
|
|
target_node.execute_command.side_effect = error
|
|
instance = _build_8x_cluster_instance()
|
|
|
|
with pytest.raises(RedisTimeoutError) as exc_info:
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert exc_info.value is error
|
|
assert instance._initialize is False
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_connection_error_preserves_upstream_topology_reinit() -> None:
|
|
error = RedisConnectionError("connection error")
|
|
target_node = _FakeClusterNode("node-a")
|
|
target_node.execute_command.side_effect = error
|
|
instance = _build_8x_cluster_instance()
|
|
|
|
with pytest.raises(RedisConnectionError) as exc_info:
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert exc_info.value is error
|
|
assert instance._initialize is True
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_three_consecutive_timeouts_request_topology_reinit_and_reset_counter() -> None:
|
|
errors = [
|
|
RedisTimeoutError("timeout-1"),
|
|
RedisTimeoutError("timeout-2"),
|
|
RedisTimeoutError("timeout-3"),
|
|
]
|
|
fourth_error = RedisTimeoutError("timeout-4")
|
|
target_node = _FakeClusterNode("node-a")
|
|
target_node.execute_command.side_effect = [*errors, fourth_error]
|
|
instance = _build_8x_cluster_instance()
|
|
|
|
for error in errors:
|
|
with pytest.raises(RedisTimeoutError) as exc_info:
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
assert exc_info.value is error
|
|
|
|
assert instance._initialize is True
|
|
instance._initialize = False
|
|
|
|
with pytest.raises(RedisTimeoutError) as exc_info:
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert exc_info.value is fourth_error
|
|
assert instance._initialize is False
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_success_resets_consecutive_timeout_counter() -> None:
|
|
errors = [RedisTimeoutError("timeout-1"), RedisTimeoutError("timeout-2")]
|
|
final_error = RedisTimeoutError("timeout-3")
|
|
target_node = _FakeClusterNode("node-a")
|
|
target_node.execute_command.side_effect = [*errors, b"value", final_error]
|
|
instance = _build_8x_cluster_instance()
|
|
|
|
for error in errors:
|
|
with pytest.raises(RedisTimeoutError) as exc_info:
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
assert exc_info.value is error
|
|
assert instance._initialize is False
|
|
|
|
result = await instance._execute_command(target_node, "GET", "k")
|
|
assert result == b"value"
|
|
assert instance._initialize is False
|
|
|
|
with pytest.raises(RedisTimeoutError) as exc_info:
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert exc_info.value is final_error
|
|
assert instance._initialize is False
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_timeout_counters_are_per_node() -> None:
|
|
node_a_errors = [RedisTimeoutError("node-a-1"), RedisTimeoutError("node-a-2")]
|
|
node_b_error = RedisTimeoutError("node-b-1")
|
|
node_a = _FakeClusterNode("node-a")
|
|
node_b = _FakeClusterNode("node-b")
|
|
node_a.execute_command.side_effect = node_a_errors
|
|
node_b.execute_command.side_effect = node_b_error
|
|
instance = _build_8x_cluster_instance()
|
|
|
|
for target_node, error in (
|
|
(node_a, node_a_errors[0]),
|
|
(node_b, node_b_error),
|
|
(node_a, node_a_errors[1]),
|
|
):
|
|
with pytest.raises(RedisTimeoutError) as exc_info:
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
assert exc_info.value is error
|
|
|
|
assert instance._initialize is False
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_timeout_does_not_clear_concurrent_topology_reinit_request() -> None:
|
|
error = RedisTimeoutError("timeout")
|
|
instance = _build_8x_cluster_instance()
|
|
|
|
async def request_reinit(*args: object, **kwargs: object) -> object:
|
|
await instance.aclose()
|
|
raise error
|
|
|
|
target_node = _FakeClusterNode("node-a")
|
|
target_node.execute_command.side_effect = request_reinit
|
|
|
|
with pytest.raises(RedisTimeoutError) as exc_info:
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert exc_info.value is error
|
|
assert instance._initialize is True
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_tolerated_timeout_does_not_erase_concurrent_connection_error_reinit() -> None:
|
|
instance = _build_8x_cluster_instance()
|
|
failing_node = _FakeClusterNode("node-a", raises=RedisConnectionError("gone"))
|
|
slow_node = _FakeClusterNode("node-b", raises=RedisTimeoutError("slow"))
|
|
|
|
results = await asyncio.gather(
|
|
instance._execute_command(failing_node, "GET", "a"),
|
|
instance._execute_command(slow_node, "GET", "b"),
|
|
return_exceptions=True,
|
|
)
|
|
|
|
assert isinstance(results[0], RedisConnectionError)
|
|
assert isinstance(results[1], RedisTimeoutError)
|
|
assert instance._initialize is True
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_overlapping_tolerated_timeouts_do_not_request_topology_reinit() -> None:
|
|
instance = _build_8x_cluster_instance()
|
|
node_a = _FakeClusterNode("node-a", raises=RedisTimeoutError("slow-a"))
|
|
node_b = _FakeClusterNode("node-b", raises=RedisTimeoutError("slow-b"))
|
|
|
|
results = await asyncio.gather(
|
|
instance._execute_command(node_a, "GET", "a"),
|
|
instance._execute_command(node_b, "GET", "b"),
|
|
return_exceptions=True,
|
|
)
|
|
|
|
assert all(isinstance(result, RedisTimeoutError) for result in results)
|
|
assert instance._initialize is False
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_tolerated_timeout_does_not_clear_pending_reinit() -> None:
|
|
instance = _build_8x_cluster_instance()
|
|
instance._initialize = True
|
|
target_node = _FakeClusterNode("node-a", raises=RedisTimeoutError("slow"))
|
|
|
|
with pytest.raises(RedisTimeoutError):
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert instance._initialize is True
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_success_returns_value_without_topology_reinit() -> None:
|
|
target_node = _FakeClusterNode("node-a", response=b"value")
|
|
instance = _build_8x_cluster_instance()
|
|
|
|
result = await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert result == b"value"
|
|
assert instance._initialize is False
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
@pytest.mark.parametrize("error_cls", [RedisConnectionError, RedisTimeoutError])
|
|
async def test_node_level_error_resets_only_that_node_not_the_whole_client(error_cls: type[Exception]) -> None:
|
|
"""The fix: a ConnectionError/TimeoutError must disconnect only the failing node
|
|
and must NOT call the client-wide aclose() that tears down every node."""
|
|
target_node = _FakeClusterNode("node-a", raises=error_cls("boom"))
|
|
instance = _build_cluster_instance()
|
|
|
|
with pytest.raises(error_cls):
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
target_node.disconnect.assert_awaited_once()
|
|
instance.aclose.assert_not_awaited()
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_moved_error_retries_without_full_reinit_before_threshold() -> None:
|
|
moved_error = MovedError("1 127.0.0.1:7001")
|
|
target_node = _FakeClusterNode("node-a")
|
|
target_node.execute_command = AsyncMock(side_effect=[moved_error, b"value"])
|
|
instance = _build_cluster_instance()
|
|
instance.RedisClusterRequestTTL = 2
|
|
instance.nodes_manager = _FakeNodesManager(node_to_return=target_node)
|
|
instance._determine_slot = AsyncMock(return_value=0)
|
|
|
|
result = await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert result == b"value"
|
|
assert instance.nodes_manager._moved_exception is moved_error
|
|
instance.aclose.assert_not_awaited()
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_ask_error_sends_asking_and_retries_on_redirected_node() -> None:
|
|
ask_error = AskError("0 127.0.0.1:7001")
|
|
target_node = _FakeClusterNode("node-a")
|
|
target_node.execute_command = AsyncMock(side_effect=[ask_error, None, b"value"])
|
|
instance = _build_cluster_instance()
|
|
instance.RedisClusterRequestTTL = 2
|
|
instance.get_node = Mock(return_value=target_node)
|
|
|
|
result = await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert result == b"value"
|
|
instance.get_node.assert_called_once_with(node_name="127.0.0.1:7001")
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_try_again_error_exhausts_ttl() -> None:
|
|
target_node = _FakeClusterNode("node-a", raises=TryAgainError("try again"))
|
|
instance = _build_cluster_instance()
|
|
instance.RedisClusterRequestTTL = 2
|
|
|
|
with pytest.raises(ClusterError):
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_successful_command_touches_neither_disconnect_nor_aclose() -> None:
|
|
target_node = _FakeClusterNode("node-a", response=b"v")
|
|
instance = _build_cluster_instance()
|
|
|
|
result = await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert result == b"v"
|
|
target_node.disconnect.assert_not_awaited()
|
|
instance.aclose.assert_not_awaited()
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
@pytest.mark.parametrize("error_cls", [BusyLoadingError, MaxConnectionsError])
|
|
async def test_busy_loading_and_max_connections_reraise_without_any_reset(error_cls: type[Exception]) -> None:
|
|
"""Unchanged from upstream: these say nothing about node health, so neither the
|
|
node nor the client should be reset."""
|
|
target_node = _FakeClusterNode("node-a", raises=error_cls("boom"))
|
|
instance = _build_cluster_instance()
|
|
|
|
with pytest.raises(error_cls):
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
target_node.disconnect.assert_not_awaited()
|
|
instance.aclose.assert_not_awaited()
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_cluster_down_error_still_triggers_a_full_reinit() -> None:
|
|
"""Unchanged from upstream: ClusterDownError is real evidence the topology
|
|
changed, so a full-client reinit (unlike a plain timeout) is still correct here."""
|
|
target_node = _FakeClusterNode("node-a", raises=ClusterDownError("boom"))
|
|
instance = _build_cluster_instance()
|
|
|
|
with pytest.raises(ClusterDownError):
|
|
await instance._execute_command(target_node, "GET", "k")
|
|
|
|
instance.aclose.assert_awaited_once()
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_moved_error_still_triggers_reinit_after_reinitialize_steps() -> None:
|
|
"""Unchanged from upstream: repeated MOVED responses are real evidence of a
|
|
slot migration, so they should still force a full reinit every `reinitialize_steps`."""
|
|
target_node = _FakeClusterNode("node-a", raises=MovedError("1 127.0.0.1:7001"))
|
|
instance = _build_cluster_instance()
|
|
instance.reinitialize_steps = 1
|
|
instance.RedisClusterRequestTTL = 2
|
|
instance.nodes_manager = _FakeNodesManager(node_to_return=target_node)
|
|
instance._determine_slot = AsyncMock(return_value=0)
|
|
|
|
target_node.execute_command = AsyncMock(side_effect=[MovedError("1 127.0.0.1:7001"), b"v"])
|
|
|
|
result = await instance._execute_command(target_node, "GET", "k")
|
|
|
|
assert result == b"v"
|
|
instance.aclose.assert_awaited_once()
|
|
assert instance.reinitialize_counter == 0
|