A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 23:45 UTC
[Bugfix] Fix example disagg_example_p2p_nccl_xpyd.sh zombie process (#21437)
Signed-off-by: David Chen <530634352@qq.com>
WeiQing Chen committed
Jul 24, 2025 at 11:42 UTC
f8c15c4efb90d3c6aa879e6fa0f5abad9f80b9aa
1 file changed
+1
examples/online_serving/disaggregated_serving_p2p_nccl_xpyd/disagg_example_p2p_nccl_xpyd.sh
+1
@@ -93,6 +93,7 @@ ensure_python_library_installed() {
93
cleanup() {
94
echo "Stopping everything…"
95
trap - INT TERM # prevent re-entrancy
96
+ pkill -9 -f "disagg_proxy_p2p_nccl_xpyd.py"
97
kill -- -$$ # negative PID == "this whole process-group"
98
wait # reap children so we don't leave zombies
99
exit 0