xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-07 08:59:09 +08:00
Code Issues Packages Projects Releases Wiki Activity

[BugFix] Avoid error traceback in logs when V1 LLM terminates (#13565)

Browse Source 
Signed-off-by: Nick Hill <nhill@redhat.com>
This commit is contained in:
Nick Hill 2025-02-19 16:49:01 -08:00 committed by GitHub
parent 550d97eb58
commit a4c402a756
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 12 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
vllm/v1/engine/core_client.py
View File

@ -252,14 +252,18 @@ class SyncMPClient(MPClient):
outputs_queue = self.outputs_queue outputs_queue = self.outputs_queue
def process_outputs_socket(): def process_outputs_socket():
while True: try:
(frame, ) = output_socket.recv_multipart(copy=False) while True:
outputs = decoder.decode(frame.buffer) (frame, ) = output_socket.recv_multipart(copy=False)
if outputs.utility_output: outputs = decoder.decode(frame.buffer)
_process_utility_output(outputs.utility_output, if outputs.utility_output:
utility_results) _process_utility_output(outputs.utility_output,
else: utility_results)
outputs_queue.put_nowait(outputs) else:
outputs_queue.put_nowait(outputs)
except zmq.error.ContextTerminated:
# Expected when the class is GC'd / during process termination.
pass
# Process outputs from engine in separate thread. # Process outputs from engine in separate thread.
Thread(target=process_outputs_socket, daemon=True).start() Thread(target=process_outputs_socket, daemon=True).start()