[https://nvbugs/5485593][fix] improve accuracy/test_disaggregated_serving.py #7366
Conversation
Signed-off-by: Lizhi Zhou <[email protected]>
📝 Walkthrough✨ Finishing Touches
🧪 Generate unit tests
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
Status, Documentation and Community
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
tests/integration/defs/accuracy/test_disaggregated_serving.py (2)
1-1: Add the NVIDIA copyright header (required by repo guidelines).Per the coding guidelines, prepend the current-year NVIDIA copyright header to all .py files.
Apply this diff at the top of the file:
+# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.
14-14: Python 3.8 typing incompatibility: built-in generics and subscripting concurrent.futures.Future.
list[...]requires 3.9+, andconcurrent.futures.Future[...]may be evaluated at import time in 3.8. Use typing aliases instead.Apply this diff:
-from typing import Any, Dict, List, Optional +from typing import Any, Dict, List, Optional, Future @@ - self.futures: list[concurrent.futures.Future[RequestOutput]] = [] + self.futures: List[Future[RequestOutput]] = []Also applies to: 59-59
🧹 Nitpick comments (1)
tests/integration/defs/accuracy/test_disaggregated_serving.py (1)
51-53: Good defaults; consider making them CI-tunable via env vars.Allow overrides to reduce flakes across environments.
Apply this diff:
-DEFAULT_TEST_TIMEOUT = 1800 -DEFAULT_SERVER_WAITING_TIMEOUT = 1200 +DEFAULT_TEST_TIMEOUT = int(os.getenv("TRTLLM_DEFAULT_TEST_TIMEOUT", "1800")) +DEFAULT_SERVER_WAITING_TIMEOUT = int(os.getenv("TRTLLM_SERVER_WAIT_TIMEOUT", "1200"))
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
💡 Knowledge Base configuration:
- MCP integration is disabled by default for public repositories
- Jira integration is disabled by default for public repositories
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (1)
tests/integration/defs/accuracy/test_disaggregated_serving.py(10 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{cpp,cc,cxx,cu,py,h,hpp,hh,hxx,cuh}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Use spaces only; indent 4 spaces
Files:
tests/integration/defs/accuracy/test_disaggregated_serving.py
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+
Indent with 4 spaces; do not use tabs
Preserve module namespaces in imports; prefer from package.subpackage import module then module.Symbol
Python file names use snake_case (e.g., some_file.py)
Class names use PascalCase
Function and method names use snake_case
Local variable names use snake_case; if starting with a number, prefix k (e.g., k_99th_percentile)
Global variables use G_ prefix and UPPER_SNAKE_CASE (e.g., G_MY_GLOBAL)
Constants use UPPER_SNAKE_CASE
Avoid shadowing variables from an outer scope
Initialize all externally visible members of a class in init
For interfaces used outside a file, prefer docstrings; reserve comments for internal code or local interfaces
Use Google-style docstrings for classes and functions; document attributes/variables inline as shown
Avoid reflection when simple, explicit code suffices (e.g., prefer def make_complex(x,y) over locals()/dict tricks)
Catch the narrowest exceptions possible in try/except
For duck-typing try/except, keep try body minimal and use else for main logic
Files:
tests/integration/defs/accuracy/test_disaggregated_serving.py
**/*.{cpp,cc,cxx,cu,h,hpp,hh,hxx,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Prepend NVIDIA copyright header (current year) to all source files (.cpp, .h, .cu, .py, etc.)
Files:
tests/integration/defs/accuracy/test_disaggregated_serving.py
🧬 Code graph analysis (1)
tests/integration/defs/accuracy/test_disaggregated_serving.py (1)
tests/integration/defs/trt_test_alternative.py (1)
popen(195-214)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Pre-commit Check
🔇 Additional comments (4)
tests/integration/defs/accuracy/test_disaggregated_serving.py (4)
7-7: Import usage looks good.
itertoolsis used for liveness checks; no issues.
212-214: Context manager ordering LGTM.Server exits first, then workers, then tempdir and thread pool—cleanup order is sensible.
354-354: Unified test timeouts look good.Consistent use of DEFAULT_TEST_TIMEOUT across suites improves predictability.
Also applies to: 611-611, 650-650, 734-734, 789-789, 904-904
206-210: Use the underscore form—no change needed.The CLI in
tensorrt_llm/commands/serve.pyregisters only--server_start_timeout(at line 446); there is no--server-start-timeoutalias. Tests are already using the correct flag.
| def launch_disaggregated_llm( | ||
| disaggregated_server_config: Dict[str, Any], | ||
| ctx_server_config: Dict[str, Any], | ||
| gen_server_config: Dict[str, Any], | ||
| model_name: str, | ||
| tensor_parallel_size: int = 1, | ||
| ctx_model: str = None, | ||
| gen_model: str = None, | ||
| server_waiting_timeout: int = DEFAULT_SERVER_WAITING_TIMEOUT): | ||
| temp_dir = tempfile.TemporaryDirectory() |
There was a problem hiding this comment.
🛠️ Refactor suggestion
Tighten type hints and add a short docstring for the new parameter.
Use Optional[str] for nullable params and document the new timeout.
Apply this diff:
@contextlib.contextmanager
def launch_disaggregated_llm(
disaggregated_server_config: Dict[str, Any],
ctx_server_config: Dict[str, Any],
gen_server_config: Dict[str, Any],
model_name: str,
tensor_parallel_size: int = 1,
- ctx_model: str = None,
- gen_model: str = None,
+ ctx_model: Optional[str] = None,
+ gen_model: Optional[str] = None,
server_waiting_timeout: int = DEFAULT_SERVER_WAITING_TIMEOUT):
+ """
+ Launch disaggregated serving (context workers + generation workers + router) for accuracy tests.
+ Args:
+ disaggregated_server_config: Router config with context/generation URLs.
+ ctx_server_config: Per-context-server options.
+ gen_server_config: Per-generation-server options.
+ model_name: HF model id or local path used by client.
+ tensor_parallel_size: TP passed to trtllm-serve nodes when not set in server configs.
+ ctx_model: Optional model for context servers; defaults to model_name.
+ gen_model: Optional model for generation servers; defaults to model_name.
+ server_waiting_timeout: Seconds to wait for server readiness; also passed to the router.
+ Yields:
+ DuckLLM(args, tokenizer, generate_async).
+ Raises:
+ TimeoutError if the router health never reaches 200 within the timeout.
+ Exception if any subprocess exits early.
+ """📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| def launch_disaggregated_llm( | |
| disaggregated_server_config: Dict[str, Any], | |
| ctx_server_config: Dict[str, Any], | |
| gen_server_config: Dict[str, Any], | |
| model_name: str, | |
| tensor_parallel_size: int = 1, | |
| ctx_model: str = None, | |
| gen_model: str = None, | |
| server_waiting_timeout: int = DEFAULT_SERVER_WAITING_TIMEOUT): | |
| temp_dir = tempfile.TemporaryDirectory() | |
| @contextlib.contextmanager | |
| def launch_disaggregated_llm( | |
| disaggregated_server_config: Dict[str, Any], | |
| ctx_server_config: Dict[str, Any], | |
| gen_server_config: Dict[str, Any], | |
| model_name: str, | |
| tensor_parallel_size: int = 1, | |
| ctx_model: Optional[str] = None, | |
| gen_model: Optional[str] = None, | |
| server_waiting_timeout: int = DEFAULT_SERVER_WAITING_TIMEOUT): | |
| """ | |
| Launch disaggregated serving (context workers + generation workers + router) for accuracy tests. | |
| Args: | |
| disaggregated_server_config: Router config with context/generation URLs. | |
| ctx_server_config: Per-context-server options. | |
| gen_server_config: Per-generation-server options. | |
| model_name: HF model id or local path used by client. | |
| tensor_parallel_size: TP passed to trtllm-serve nodes when not set in server configs. | |
| ctx_model: Optional model for context servers; defaults to model_name. | |
| gen_model: Optional model for generation servers; defaults to model_name. | |
| server_waiting_timeout: Seconds to wait for server readiness; also passed to the router. | |
| Yields: | |
| DuckLLM(args, tokenizer, generate_async). | |
| Raises: | |
| TimeoutError: If the router health never reaches HTTP 200 within the timeout. | |
| Exception: If any subprocess exits early. | |
| """ | |
| temp_dir = tempfile.TemporaryDirectory() | |
| # … rest of implementation … |
🤖 Prompt for AI Agents
In tests/integration/defs/accuracy/test_disaggregated_serving.py around lines 74
to 83, the function signature uses nullable string params without Optional and
lacks a docstring for the new server_waiting_timeout parameter; change the type
hints for ctx_model and gen_model to Optional[str], import Optional from typing
if not present, and add a short docstring above the function describing each
parameter including server_waiting_timeout (what it controls and units) so
callers understand the timeout behavior.
| while time.time() - start_time < server_waiting_timeout: | ||
| time.sleep(5) | ||
| for process in itertools.chain(worker_processes, [server_process]): | ||
| if process.poll() is not None: | ||
| raise Exception( | ||
| f"process {process.pid} exited with code {process.returncode}" | ||
| ) |
There was a problem hiding this comment.
Timeout loop never fails closed; add explicit failure and HTTP timeouts.
If health never returns 200 within the window, code proceeds instead of failing fast. Also add a short HTTP timeout to avoid hangs on half-open connections.
Apply this diff:
- start_time = time.time()
- while time.time() - start_time < server_waiting_timeout:
+ start_time = time.time()
+ ready = False
+ while time.time() - start_time < server_waiting_timeout:
time.sleep(5)
for process in itertools.chain(worker_processes, [server_process]):
if process.poll() is not None:
raise Exception(
f"process {process.pid} exited with code {process.returncode}"
)
try:
print("Checking health endpoint")
- response = requests.get("http://localhost:8000/health")
+ response = requests.get("http://localhost:8000/health", timeout=5)
if response.status_code == 200:
- break
- except requests.exceptions.ConnectionError:
+ ready = True
+ break
+ except (requests.exceptions.ConnectionError, requests.exceptions.Timeout):
continue
+ if not ready:
+ raise TimeoutError(
+ f"Server did not become ready within {server_waiting_timeout} seconds."
+ )📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| while time.time() - start_time < server_waiting_timeout: | |
| time.sleep(5) | |
| for process in itertools.chain(worker_processes, [server_process]): | |
| if process.poll() is not None: | |
| raise Exception( | |
| f"process {process.pid} exited with code {process.returncode}" | |
| ) | |
| start_time = time.time() | |
| ready = False | |
| while time.time() - start_time < server_waiting_timeout: | |
| time.sleep(5) | |
| for process in itertools.chain(worker_processes, [server_process]): | |
| if process.poll() is not None: | |
| raise Exception( | |
| f"process {process.pid} exited with code {process.returncode}" | |
| ) | |
| try: | |
| print("Checking health endpoint") | |
| response = requests.get("http://localhost:8000/health", timeout=5) | |
| if response.status_code == 200: | |
| ready = True | |
| break | |
| except (requests.exceptions.ConnectionError, requests.exceptions.Timeout): | |
| continue | |
| if not ready: | |
| raise TimeoutError( | |
| f"Server did not become ready within {server_waiting_timeout} seconds." | |
| ) |
Detect subprocesses' status and exit early on failure.
Summary by CodeRabbit
New Features
Bug Fixes
Tests
Description
Test Coverage
PR Checklist
Please review the following before submitting your PR:
PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.
PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.
Test cases are provided for new code paths (see test instructions)
Any new dependencies have been scanned for license and vulnerabilities
CODEOWNERS updated if ownership changes
Documentation updated as needed
The reviewers assigned automatically/manually are appropriate for the PR.
Please check this after reviewing the above items as appropriate for this PR.
GitHub Bot Help
/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...Provide a user friendly way for developers to interact with a Jenkins server.
Run
/bot [-h|--help]to print this help message.See details below for each supported subcommand.
run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]Launch build/test pipelines. All previously running jobs will be killed.
--reuse-test (optional)pipeline-id(OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.--disable-reuse-test(OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.--disable-fail-fast(OPTIONAL) : Disable fail fast on build/tests/infra failures.--skip-test(OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.--stage-list "A10-PyTorch-1, xxx"(OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.--gpu-type "A30, H100_PCIe"(OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.--test-backend "pytorch, cpp"(OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.--only-multi-gpu-test(OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.--disable-multi-gpu-test(OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.--add-multi-gpu-test(OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.--post-merge(OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx"(OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".--detailed-log(OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.--debug(OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in thestage-listparameter to access the appropriate container environment. Note: Does NOT update GitHub check status.For guidance on mapping tests to stage names, see
docs/source/reference/ci-overview.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip testing for latest commit on pull request.
--comment "Reason for skipping build/test"is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.reuse-pipeline
reuse-pipelineReuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.