MCPcopy Index your code
hub / github.com/PaddlePaddle/FastDeploy / start

Method start

fastdeploy/engine/async_llm.py:106–122  ·  view source on GitHub ↗

Start engine service process

(self)

Source from the content-addressed store, hash-verified

104 llm_logger.info(f"EngineServiceClient initialized with engine_pid: {self.engine_pid}")
105
106 async def start(self):
107 """Start engine service process"""
108 try:
109 # Start independent engine process
110 self._start_engine_process()
111
112 # Wait for engine to be ready
113 if not self._wait_engine_ready():
114 raise EngineError("Engine failed to start within timeout", error_code=500)
115
116 self._running = True
117 llm_logger.info("EngineServiceClient started successfully")
118
119 except Exception as e:
120 llm_logger.error(f"Failed to start EngineServiceClient: {e}")
121 raise
122 return True
123
124 def _start_engine_process(self):
125 """Start engine process"""

Callers 2

_testMethod · 0.95
_start_engine_processMethod · 0.45

Calls 5

_start_engine_processMethod · 0.95
_wait_engine_readyMethod · 0.95
EngineErrorClass · 0.90
infoMethod · 0.45
errorMethod · 0.45

Tested by 1

_testMethod · 0.76