Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Flyte

hello! assuming that we're in a `@dynamic` workflow, is there a way to define dependencies between tasks without relying on their respective outputs? i.e, we would like to run `task2()` after `task1()` even though the latter does not output anything.
In a classic `@workflow` , we could rely on `create_node()` + `&gt;&gt;` but can't figure out the equivalent for `@dynamic`

It fails with `Cannot use explicit run to call Flyte entities workflows.test.task1`

This issue appears when we run a dynamic in local mode.

and the code is as simple as this

```@task
def task1():
    print("1")

@task
def task2():
    print("2")

@task
def task3():
    print("3") 

@dynamic
def dynamic_wf() -&gt; str:
    node_1 = create_node(
        task1
    )       
    node_2 = create_node(
        task2
    )
    node_3 = create_node(
        task3
    )

    node_3 &gt;&gt; node_2
    node_1 &gt;&gt; node_2

    return "Done"

@workflow
def wf2() -&gt; str:
    return dynamic_wf()    ```

after a further investigation, it looks like the execution mode is set to "TASK_EXECUTION" when we run `pyflyte run file.py wf2`

the execution mode is correctly set to "LOCAL_WORKFLOW_EXECUTION" when we target a workflow which doesn't return a dynamic.

Cc ohh and this runs in remote? The code looks right, will tal in a bit. Cc <@U0265RTUJ5B> if you are around

It looks like we a bug in `create_node` in `@dynamic` in local executions.
This works in a remote setting.

thanks Eduardo. Do you know if this bug has been introduced recently ? Is there a way to downgrade to an earlier version of Flyte, in order to make this work ?

I don't believe so, but give me a few minutes to confirm.

<@U03HT3VNN06> the fix should be very simple as local execution is actually very simple

thank you for pointing it out, do you mind filing it?

Yeah, this has been bug for a long time. I opened <https://github.com/flyteorg/flyte/issues/2692> to track it.

thanks a lot <@U0265RTUJ5B> and <@UNZB4NW3S>