65 lines
1.8 KiB
Python
65 lines
1.8 KiB
Python
# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
|
|
import asyncio
|
|
import signal
|
|
|
|
import uvloop
|
|
|
|
from dynamo.runtime import DistributedRuntime, dynamo_worker
|
|
|
|
|
|
class RequestHandler:
|
|
"""
|
|
Request handler for the generate endpoint
|
|
"""
|
|
|
|
async def generate(self, request):
|
|
print(f"Received request: {request}")
|
|
for char in request:
|
|
await asyncio.sleep(1)
|
|
yield char
|
|
|
|
|
|
@dynamo_worker()
|
|
async def worker(runtime: DistributedRuntime):
|
|
# Set up signal handler for graceful shutdown
|
|
loop = asyncio.get_running_loop()
|
|
|
|
def signal_handler():
|
|
# Schedule the shutdown coroutine instead of calling it directly
|
|
asyncio.create_task(graceful_shutdown(runtime))
|
|
|
|
for sig in (signal.SIGTERM, signal.SIGINT):
|
|
loop.add_signal_handler(sig, signal_handler)
|
|
|
|
print("Signal handlers registered for graceful shutdown")
|
|
await init(runtime, "dynamo")
|
|
|
|
|
|
async def graceful_shutdown(runtime: DistributedRuntime):
|
|
print("Received shutdown signal, shutting down DistributedRuntime")
|
|
runtime.shutdown()
|
|
print("DistributedRuntime shutdown complete")
|
|
|
|
|
|
async def init(runtime: DistributedRuntime, ns: str):
|
|
"""
|
|
Instantiate a `backend` component and serve the `generate` endpoint
|
|
A `Component` can serve multiple endpoints
|
|
"""
|
|
component = runtime.namespace(ns).component("backend")
|
|
await component.create_service()
|
|
|
|
endpoint = component.endpoint("generate")
|
|
print("Started server instance")
|
|
|
|
# the server will gracefully shutdown (i.e., keep opened TCP streams finishes)
|
|
# after the lease is revoked
|
|
await endpoint.serve_endpoint(RequestHandler().generate)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
uvloop.install()
|
|
asyncio.run(worker())
|