[Bugfix] Fix ray instance detect issue (#9439)
This commit is contained in:
parent
feb92fbe4a
commit
2adb4409e0
@ -232,6 +232,13 @@ def initialize_ray_cluster(
|
|||||||
|
|
||||||
# Connect to a ray cluster.
|
# Connect to a ray cluster.
|
||||||
if current_platform.is_rocm() or current_platform.is_xpu():
|
if current_platform.is_rocm() or current_platform.is_xpu():
|
||||||
|
# Try to connect existing ray instance and create a new one if not found
|
||||||
|
try:
|
||||||
|
ray.init("auto")
|
||||||
|
except ConnectionError:
|
||||||
|
logger.warning(
|
||||||
|
"No existing RAY instance detected. "
|
||||||
|
"A new instance will be launched with current node resources.")
|
||||||
ray.init(address=ray_address,
|
ray.init(address=ray_address,
|
||||||
ignore_reinit_error=True,
|
ignore_reinit_error=True,
|
||||||
num_gpus=parallel_config.world_size)
|
num_gpus=parallel_config.world_size)
|
||||||
|
|||||||
Loading…
Reference in New Issue
Block a user