Skip to content

test(litellm): Replace mocks with httpx types in rate-limit test #5975

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
alexander-alderman-webb wants to merge 6 commits into
base: webb/litellm/embedding-tests
Choose a base branch
from
Open

test(litellm): Replace mocks with httpx types in rate-limit test #5975

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
a7dfb23
.
alexander-alderman-webb Apr 10, 2026
99de614
Merge branch 'webb/litellm/embedding-tests' into webb/litellm/rate-li…
alexander-alderman-webb Apr 10, 2026
c8d86a8
add fixture
alexander-alderman-webb Apr 10, 2026
4226d2c
Merge branch 'webb/litellm/embedding-tests' into webb/litellm/rate-li…
alexander-alderman-webb Apr 10, 2026
54925ab
patch completions client instead of embeddings
alexander-alderman-webb Apr 10, 2026
179f14b
Merge branch 'webb/litellm/embedding-tests' into webb/litellm/rate-li…
alexander-alderman-webb Apr 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 22 additions & 0 deletions tests/conftest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1080,6 +1080,28 @@ def inner(response_content, serialize_pydantic=False, request_headers=None):
return inner


@pytest.fixture
def get_rate_limit_model_response():
def inner(request_headers=None):
if request_headers is None:
request_headers = {}

model_request = HttpxRequest(
"POST",
"/responses",
headers=request_headers,
)

response = HttpxResponse(
429,
request=model_request,
)

return response

return inner


@pytest.fixture
def streaming_chat_completions_model_response():
return [
Expand Down
31 changes: 18 additions & 13 deletions tests/integrations/litellm/test_litellm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -465,7 +465,9 @@ def test_embeddings_no_pii(
assert SPANDATA.GEN_AI_EMBEDDINGS_INPUT not in span["data"]


def test_exception_handling(sentry_init, capture_events):
def test_exception_handling(
reset_litellm_executor, sentry_init, capture_events, get_rate_limit_model_response
):
sentry_init(
integrations=[LiteLLMIntegration()],
traces_sample_rate=1.0,
Expand All @@ -474,19 +476,22 @@ def test_exception_handling(sentry_init, capture_events):

messages = [{"role": "user", "content": "Hello!"}]

with start_transaction(name="litellm test"):
kwargs = {
"model": "gpt-3.5-turbo",
"messages": messages,
}
client = OpenAI(api_key="z")
Copy link
Copy Markdown

@cursor cursor bot Apr 10, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

OpenAI client retries 429 causing unnecessary test delay

Low Severity

OpenAI(api_key="z") uses the default max_retries=2, so when the mock returns a 429 response, the OpenAI client will automatically retry twice with exponential backoff (~1.5 seconds of sleeping) before finally raising RateLimitError. The old test called _failure_callback directly and had no such delay. Creating the client with max_retries=0 would avoid unnecessary retries and keep the test fast.

Fix in Cursor Fix in Web

Reviewed by Cursor Bugbot for commit 179f14b. Configure here.


_input_callback(kwargs)
_failure_callback(
kwargs,
Exception("API rate limit reached"),
datetime.now(),
datetime.now(),
)
model_response = get_rate_limit_model_response()

with mock.patch.object(
client.completions._client._client,
"send",
return_value=model_response,
):
with start_transaction(name="litellm test"):
with pytest.raises(litellm.RateLimitError):
litellm.completion(
model="gpt-3.5-turbo",
messages=messages,
client=client,
)

# Should have error event and transaction
assert len(events) >= 1
Expand Down
Loading