[Core] "Reuse Ray Actors in DAGs" example does not seem to work when connecting to an existing ray instance #50328
Description
What happened + What you expected to happen
Running the example reuse-ray-actors-in-dags as shown on Ray Docs does not seem to work unless the code is run on a local Ray instance.
The error given is:
Traceback (most recent call last):
File "/home/******/ray/test.py", line 24, in <module>
dag = MultiOutputNode([worker.forward.bind(input_data)])
AttributeError: 'ClientRemoteMethod' object has no attribute 'bind'
Using bind() API on the Actor seems to work just fine even on an existing (non-local) Ray instance
Versions / Dependencies
Fresh conda environment running
Python version: 3.10.16
Ray version: 2.42.0
The cluster is running on microk8s, and installed via kuberay chart with image tag set to 2.42.0-py310
Reproduction script
# taken and adapted from Ray Docs @ https://docs.ray.io/en/releases-2.40.0/ray-core/ray-dag.html#reuse-ray-actors-in-dags
import ray
from ray.dag.input_node import InputNode
from ray.dag.output_node import MultiOutputNode
import ray
ray.init(address='ray://localhost:10001')
@ray.remote
class Worker:
def __init__(self):
self.forwarded = 0
def forward(self, input_data: int):
self.forwarded += 1
return input_data + 1
def num_forwarded(self):
return self.forwarded
# Create an actor via ``remote`` API not ``bind`` API to avoid
# killing actors when a DAG is finished.
worker = Worker.remote()
with InputNode() as input_data:
dag = MultiOutputNode([worker.forward.bind(input_data)])
# Actors are reused. The DAG definition doesn't include
# actor creation.
assert ray.get(dag.execute(1)) == [2]
assert ray.get(dag.execute(2)) == [3]
assert ray.get(dag.execute(3)) == [4]
Issue Severity
Low: It annoys or frustrates me.