updated

[robertgshaw2-redhat]

SUMMARY:

• handle exception in multiproc worker busy loop (currently, we hang if this happens)

This is what the stack trace looks like after this PR:

INFO:     Started server process [4028321]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://0.0.0.0:8001 (Press CTRL+C to quit)
INFO:     127.0.0.1:58610 - "GET /v1/models HTTP/1.1" 200 OK
INFO 01-03 16:21:03 logger.py:37] Received request cmpl-ac2ac919986b405fb3528f1a0e8e1613-0: prompt: 'Hello my name is', params: SamplingParams(n=1, presence_penalty=0.0, frequency_penalty=0.0, repetition_penalty=1.0, temperature=1.0, top_p=1.0, top_k=-1, min_p=0.0, seed=None, stop=[], stop_token_ids=[], bad_words=[], include_stop_str_in_output=False, ignore_eos=False, max_tokens=100, min_tokens=0, logprobs=None, prompt_logprobs=None, skip_special_tokens=True, spaces_between_special_tokens=True, truncate_prompt_tokens=None, guided_decoding=None), prompt_token_ids: [128000, 9906, 856, 836, 374], lora_request: None, prompt_adapter_request: None.
INFO:     127.0.0.1:58610 - "POST /v1/completions HTTP/1.1" 200 OK
INFO 01-03 16:21:04 async_llm.py:191] Added request cmpl-ac2ac919986b405fb3528f1a0e8e1613-0.
ERROR 01-03 16:21:05 core.py:200] EngineCore hit an exception: Traceback (most recent call last):
ERROR 01-03 16:21:05 core.py:200]   File "/home/rshaw/vllm/vllm/v1/engine/core.py", line 193, in run_engine_core
ERROR 01-03 16:21:05 core.py:200]     engine_core.run_busy_loop()
ERROR 01-03 16:21:05 core.py:200]   File "/home/rshaw/vllm/vllm/v1/engine/core.py", line 231, in run_busy_loop
ERROR 01-03 16:21:05 core.py:200]     outputs = self.step()
ERROR 01-03 16:21:05 core.py:200]               ^^^^^^^^^^^
ERROR 01-03 16:21:05 core.py:200]   File "/home/rshaw/vllm/vllm/v1/engine/core.py", line 124, in step
ERROR 01-03 16:21:05 core.py:200]     output = self.model_executor.execute_model(scheduler_output)
ERROR 01-03 16:21:05 core.py:200]              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 01-03 16:21:05 core.py:200]   File "/home/rshaw/vllm/vllm/v1/executor/multiproc_executor.py", line 163, in execute_model
ERROR 01-03 16:21:05 core.py:200]     model_output = self.collective_rpc("execute_model",
ERROR 01-03 16:21:05 core.py:200]                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 01-03 16:21:05 core.py:200]   File "/home/rshaw/vllm/vllm/v1/executor/multiproc_executor.py", line 157, in collective_rpc
ERROR 01-03 16:21:05 core.py:200]     raise e
ERROR 01-03 16:21:05 core.py:200]   File "/home/rshaw/vllm/vllm/v1/executor/multiproc_executor.py", line 146, in collective_rpc
ERROR 01-03 16:21:05 core.py:200]     raise result
ERROR 01-03 16:21:05 core.py:200] ValueError: SIMULATE CUDA EXCEPTION
ERROR 01-03 16:21:05 core.py:200] 
CRITICAL 01-03 16:21:05 async_llm.py:53] AsyncLLM got SIGQUIT from worker processes, shutting down. See stack trace above for root cause issue.
Killed

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

• Add ready label to the PR
• Enable auto-merge.

🚀

[robertgshaw2-redhat]

NOTE: removed this because this creates a circular reference that can prevent us from being gced (finalizer shutdown function cannot be a bound method of self + EngineCore calls already calls executor.shutdown() at its exit.