From a5a306ed50cb026665c2aa52a1f40574ed26ba8a Mon Sep 17 00:00:00 2001 From: "rshaw@neuralmagic.com" Date: Fri, 3 Jan 2025 19:31:12 +0000 Subject: [PATCH] fixed! --- vllm/v1/engine/core_client.py | 10 +++++----- vllm/v1/worker/gpu_worker.py | 2 +- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/vllm/v1/engine/core_client.py b/vllm/v1/engine/core_client.py index 22ed69e8eb2c2..6a40c961fc1d7 100644 --- a/vllm/v1/engine/core_client.py +++ b/vllm/v1/engine/core_client.py @@ -10,8 +10,8 @@ from vllm.config import VllmConfig from vllm.logger import init_logger -from vllm.utils import (get_open_zmq_ipc_path, make_zmq_socket, - kill_process_tree) +from vllm.utils import (get_open_zmq_ipc_path, kill_process_tree, + make_zmq_socket) from vllm.v1.engine import (EngineCoreOutput, EngineCoreOutputs, EngineCoreProfile, EngineCoreRequest, EngineCoreRequestType, EngineCoreRequestUnion) @@ -145,10 +145,10 @@ def __init__( # handle at the API server level so we can return a better # error code to the clients calling VLLM. def sigusr1_handler(signum, frame): - logger.fatal( - "Got fatal signal from worker processes, shutting " - "down. See stack trace above for root cause issue.") + logger.fatal("Got fatal signal from worker processes, shutting " + "down. See stack trace above for root cause issue.") kill_process_tree(os.getpid()) + signal.signal(signal.SIGUSR1, sigusr1_handler) # Serialization setup. diff --git a/vllm/v1/worker/gpu_worker.py b/vllm/v1/worker/gpu_worker.py index fcef37371a6b9..af438f7d5820c 100644 --- a/vllm/v1/worker/gpu_worker.py +++ b/vllm/v1/worker/gpu_worker.py @@ -34,7 +34,7 @@ def __init__( rank: int, distributed_init_method: str, ): - + # TODO: use WorkerBase.__init__(self, vllm_config=vllm_config) self.vllm_config = vllm_config self.model_config = vllm_config.model_config