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 17, 08:00 UTC
fix(ray_utils): ignore re-init error (#465)
Massimiliano Pronesti committed
Jul 20, 2023 at 02:01 UTC
16c3e295a877580867c8b90ace86acf3d1841455
1 file changed
+1
-1
vllm/engine/ray_utils.py
+1
-1
@@ -44,7 +44,7 @@ def initialize_cluster(
44
"Ray is not installed. Please install Ray to use distributed "
45
"serving.")
46
# Connect to a ray cluster.
47
- ray.init(address=ray_address)
47
+ ray.init(address=ray_address, ignore_reinit_error=True)
48
49
if not parallel_config.worker_use_ray:
50
# Initialize cluster locally.