[TRTLLM-8507][fix] Fix ray resource cleanup and error handling in LoRA test #8175
Conversation
shuyixiong
changed the title
📝 WalkthroughWalkthroughAdds guards in Ray executor to perform cleanup only when Ray is initialized. Test utilities gain conditional Ray imports and a new context manager to unwrap Ray errors. Test setup/teardown logic is made robust to partial initialization, and a PyTorch test now uses the new error-exposing helper and removes a skip decorator. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant T as Test code
participant CM as try_expose_error_in_ray
participant R as Ray worker/task
Note over CM: MPI disabled path
T->>CM: enter context(error_type=RuntimeError)
T->>R: invoke remote call
R-->>T: RayActorError/RayTaskError (with cause/error_msg)
T->>CM: exception bubbles into context
alt has nested cause of expected type
CM-->>T: raise underlying RuntimeError
else has formatted error_msg matching expected type
CM-->>T: raise RuntimeError with extracted message
else no match
CM-->>T: re-raise original Ray error
end
sequenceDiagram
autonumber
participant E as RayExecutor.shutdown()
participant RI as ray.is_initialized()
participant PG as PlacementGroup
participant Ray as ray
E->>RI: check initialized
alt Ray initialized
E->>PG: remove if present
E->>Ray: ray.shutdown()
else not initialized
Note over E: Skip Ray cleanup
end
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
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. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
tensorrt_llm/executor/ray_executor.py(1 hunks)tests/unittest/llmapi/lora_test_utils.py(2 hunks)tests/unittest/llmapi/test_llm_pytorch.py(2 hunks)tests/unittest/utils/util.py(2 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Use only spaces, no tabs; indent with 4 spaces.
Files:
tensorrt_llm/executor/ray_executor.pytests/unittest/llmapi/lora_test_utils.pytests/unittest/llmapi/test_llm_pytorch.pytests/unittest/utils/util.py
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+.
Indent Python code with 4 spaces; do not use tabs.
Maintain module namespace when importing; prefer 'from package.subpackage import foo' then 'foo.SomeClass()' instead of importing the class directly.
Python filenames should be snake_case (e.g., some_file.py).
Python classes use PascalCase names.
Functions and methods use snake_case names.
Local variables use snake_case; prefix 'k' for variables that start with a number (e.g., k_99th_percentile).
Global variables use upper SNAKE_CASE prefixed with 'G' (e.g., G_MY_GLOBAL).
Constants use upper SNAKE_CASE (e.g., MY_CONSTANT).
Avoid shadowing variables from an outer scope.
Initialize all externally visible members of a class in the constructor.
Prefer docstrings for interfaces that may be used outside a file; comments for in-function or file-local interfaces.
Use Google-style docstrings for classes and functions (Sphinx-parsable).
Document attributes and variables inline so they render under the class/function docstring.
Avoid reflection when a simpler, explicit approach suffices (e.g., avoid dict(**locals()) patterns).
In try/except, catch the most specific exceptions possible.
For duck-typing try/except, keep the try body minimal and use else for the main logic.
Files:
tensorrt_llm/executor/ray_executor.pytests/unittest/llmapi/lora_test_utils.pytests/unittest/llmapi/test_llm_pytorch.pytests/unittest/utils/util.py
**/*.{cpp,cxx,cc,h,hpp,hh,hxx,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Prepend the NVIDIA Apache-2.0 copyright header with current year to the top of all source files (e.g., .cpp, .h, .cu, .py).
Files:
tensorrt_llm/executor/ray_executor.pytests/unittest/llmapi/lora_test_utils.pytests/unittest/llmapi/test_llm_pytorch.pytests/unittest/utils/util.py
🧬 Code graph analysis (2)
tests/unittest/llmapi/lora_test_utils.py (2)
tensorrt_llm/_utils.py (1)
mpi_disabled(518-520)tensorrt_llm/executor/ray_executor.py (1)
shutdown(217-236)
tests/unittest/llmapi/test_llm_pytorch.py (3)
tests/unittest/utils/util.py (1)
try_expose_error_in_ray(462-487)tests/unittest/llmapi/lora_test_utils.py (1)
check_llama_7b_multi_lora_from_request_test_harness(82-133)tensorrt_llm/llmapi/llm_args.py (1)
PeftCacheConfig(973-1039)
🪛 Ruff (0.13.3)
tests/unittest/utils/util.py
472-472: Within an except clause, raise exceptions with raise ... from err or raise ... from None to distinguish them from errors in exception handling
(B904)
484-484: Within an except clause, raise exceptions with raise ... from err or raise ... from None to distinguish them from errors in exception handling
(B904)
487-487: Use raise without specifying exception name
Remove exception name
(TRY201)
⏰ 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 (7)
tensorrt_llm/executor/ray_executor.py (1)
227-236: LGTM! Guards prevent Ray auto-initialization during shutdown.The added
ray.is_initialized()checks correctly prevent triggering Ray's auto-initialization behavior during interpreter exit when cleanup is attempted but Ray is no longer active. This resolves potential crashes or hangs during teardown.tests/unittest/utils/util.py (1)
28-31: LGTM! Guarded import supports environments without Ray.The fallback to
ray_stuballows tests to run in environments where Ray is unavailable, consistent with other files in this PR.tests/unittest/llmapi/test_llm_pytorch.py (2)
31-32: LGTM! Import enables error unwrapping in Ray mode.Importing
try_expose_error_in_rayallows this test file to surface nested errors when running under Ray orchestration.
416-451: Test now runs in Ray mode with proper error surfacing.The removal of
@skip_rayand addition oftry_expose_error_in_raycontext managers correctly enable this test to run under Ray orchestration while still catching the expectedRuntimeError. The helper unwraps Ray's error wrapping to expose the underlying exception for proper assertion.tests/unittest/llmapi/lora_test_utils.py (3)
12-12: LGTM! Import enables conditional Ray cleanup.The
mpi_disabledimport is used correctly in the finally block to determine when manual Ray cleanup is needed.
16-19: LGTM! Guarded import supports environments without Ray.Consistent with the pattern in
tests/unittest/utils/util.py, this fallback allows tests to function when Ray is unavailable.
115-130: LGTM! Robust cleanup handles partial initialization.The pattern correctly handles cleanup in both success and failure cases:
- If
llminitialization succeeds, itsshutdown()method handles Ray cleanup- If initialization fails (llm remains None), manual
ray.shutdown()prevents resource leaks that would trigger pytest-threadleak detection
|
/bot run |
|
PR_Github #20781 [ run ] triggered by Bot |
shuyixiong
changed the title
|
PR_Github #20781 [ run ] completed with state |
|
/bot run |
|
PR_Github #20800 [ run ] triggered by Bot |
|
PR_Github #20800 [ run ] completed with state |
|
/bot run --stage-list "H100_PCIe-PyTorch-1" --disable-fail-fast |
|
PR_Github #20828 [ run ] triggered by Bot |
|
/bot kill |
|
/bot help |
GitHub Bot Help
Provide a user friendly way for developers to interact with a Jenkins server. Run See details below for each supported subcommand.
Launch build/test pipelines. All previously running jobs will be killed.
kill
Kill all running builds associated with pull request. skip
Skip testing for latest commit on pull request. reuse-pipeline
Reuse 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. |
|
/bot run --reuse-test 15725 |
|
PR_Github #20834 [ run ] triggered by Bot |
|
PR_Github #20828 [ run ] completed with state |
There was a problem hiding this comment.
Let's refine the ray_stub.py a bit:
Subject: [PATCH] refine stub
---
Index: tensorrt_llm/ray_stub.py
IDEA additional info:
Subsystem: com.intellij.openapi.diff.impl.patch.CharsetEP
<+>UTF-8
===================================================================
diff --git a/tensorrt_llm/ray_stub.py b/tensorrt_llm/ray_stub.py
--- a/tensorrt_llm/ray_stub.py (revision 3a4ffaed2a588d4d4bb2588c490df0348103b5d0)
+++ b/tensorrt_llm/ray_stub.py (revision 44d40b337f14acb492f28efd2362a1669abd1526)
@@ -12,11 +12,11 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
-import functools
+from functools import wraps as _wraps
-from tensorrt_llm._utils import mpi_disabled
+from tensorrt_llm._utils import mpi_disabled as _mpi_disabled
-if mpi_disabled():
+if _mpi_disabled():
raise RuntimeError(
"Ray requested (TLLM_DISABLE_MPI=1), but not installed. Please install Ray."
)
@@ -27,10 +27,11 @@
def decorator(func):
# Returns a function that always raises.
# Decorated class depends on ray, but ray is not installed.
- @functools.wraps(func)
+ @_wraps(func)
def stub_checker(*_, **__):
raise RuntimeError(
- "Ray not installed, cannot use Ray based feature.")
+ f'Ray not installed, so the remote function / actor "{func.__name__}" is not available.'
+ )
return stub_checker
@@ -38,3 +39,9 @@
return decorator(args[0])
return decorator
+
+
+def __getattr__(name):
+ raise RuntimeError(
+ f'Ray not installed, so "ray.{name}" is unavailable. Please install Ray.'
+ )
|
PR_Github #20834 [ run ] completed with state |
|
/bot run |
|
PR_Github #20950 [ run ] triggered by Bot |
shuyixiong
force-pushed
the
user/shuyix/fix_ray_lora_test
branch
from
3ee0877 to
ccd59b1
Compare
|
/bot run |
|
PR_Github #21080 [ run ] triggered by Bot |
|
PR_Github #21080 [ run ] completed with state |
shuyixiong
force-pushed
the
user/shuyix/fix_ray_lora_test
branch
from
ccd59b1 to
397756c
Compare
|
/bot run |
|
PR_Github #21167 [ run ] triggered by Bot |
|
PR_Github #21167 [ run ] completed with state |
|
/bot run --reuse-test --disable-fail-fast |
|
PR_Github #21207 [ run ] triggered by Bot |
|
PR_Github #21207 [ run ] completed with state |
Signed-off-by: shuyix <[email protected]>
Signed-off-by: shuyix <[email protected]>
Signed-off-by: shuyix <[email protected]>
Signed-off-by: shuyix <[email protected]>
Signed-off-by: shuyix <[email protected]>
Signed-off-by: shuyix <[email protected]>
Signed-off-by: shuyix <[email protected]>
Signed-off-by: shuyix <[email protected]>
Signed-off-by: shuyix <[email protected]>
shuyixiong
force-pushed
the
user/shuyix/fix_ray_lora_test
branch
from
397756c to
80e6505
Compare
|
/bot run |
|
PR_Github #21278 [ run ] triggered by Bot |
shuyixiong
changed the title
|
PR_Github #21278 [ run ] completed with state |
|
/bot run --reuse-test --disable-fail-fast |
|
PR_Github #21338 [ run ] triggered by Bot |
|
PR_Github #21338 [ run ] completed with state |
5 participants
Summary by CodeRabbit
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.