Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[core][experimental] Propagate leaf node errors to users #47691

Open
kevin85421 opened this issue Sep 16, 2024 · 0 comments
Open

[core][experimental] Propagate leaf node errors to users #47691

kevin85421 opened this issue Sep 16, 2024 · 0 comments
Assignees
Labels
accelerated-dag enhancement Request for new feature and/or capability P0 Issues that should be fixed in short order

Comments

@kevin85421
Copy link
Member

kevin85421 commented Sep 16, 2024

Description

See this gist for more details. If a DAG node is not an ancestor of the MultiOutputNode, errors thrown by the DAG node will not be propagated to users.

In the following example, the assertion error of a.fail will not be propagated to users.

@ray.remote
class Actor:
  ...
  def fail(self, x):
    assert x != x 

def test_leaf_node_one_actor(self, ray_start_regular):
    a = Actor.remote(0)
    with InputNode() as i:
        input_data = a.read_input.bind(i)
        branch1 = a.fail.bind(input_data)  # branch1: leaf node
        branch2 = a.inc.bind(input_data)
        dag = MultiOutputNode([branch2])

    compiled_dag = dag.experimental_compile()

    ref = compiled_dag.execute(10)
    print(ray.get(ref))
    compiled_dag.teardown()

Possible solution is to add leaf node to MultiOutputNode implicitly.

Use case

No response

@kevin85421 kevin85421 added enhancement Request for new feature and/or capability triage Needs triage (eg: priority, bug/not-bug, and owning component) accelerated-dag P0 Issues that should be fixed in short order and removed triage Needs triage (eg: priority, bug/not-bug, and owning component) labels Sep 16, 2024
@kevin85421 kevin85421 self-assigned this Sep 16, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
accelerated-dag enhancement Request for new feature and/or capability P0 Issues that should be fixed in short order
Projects
None yet
Development

No branches or pull requests

1 participant