Has anyone done

FlyteDirectory

inputs to a

ContainerTask

? It seems like the download step doesn't correctly handle directories? with

inputs=dict(path=FlyteDirectory)

we're getting

Failed to download from ref [<gs://cradle-bio-pipelines/i9/f7aac4450113a425b905-n0-0/path>]

where that path does exist.

Ohh this is a bug, we are tracking to fix this month

but the input is downloaded by the init container right? the sidecar is only for uploading outputs...?

@Eli Bixby and @Felix Ruess can you provide examples of the ContainerTask definition where these issues are occurring?

@Dan Rammer (hamersaw) I know

I have this on my TODO list for today, just easier to get test cases rather than write my own 😛

added a simple container tasks without outputs to the issue

yeah, i agree. i'll take try to get them both wrapped up in the fix.

1.24 or to be specific v1.24.6+k3s

Same

Hey @Eli Bixby, wondering if you could share your code for how you got this working with a FlyteDirectory input? I've tried it every way I can think of and nothing shows up in the container itself...

bt = ContainerTask(
    name="basic-test",
    input_data_dir="/var/inputs",
    output_data_dir="/var/outputs",
    inputs=kwtypes(indir=FlyteDirectory),
    outputs=kwtypes(),
    image="<http://ghcr.io/flyteorg/rawcontainers-shell:v2|ghcr.io/flyteorg/rawcontainers-shell:v2>",
    command=[
        "ls",
        "-la",
        "/var/inputs",
    ],
)

@task
def get_dir(dirpath: str) -> FlyteDirectory:
    fd = FlyteDirectory(path=dirpath)
    return fd

@workflow
def wf():
    fd = get_dir(dirpath='<s3://my-s3-bucket/cv-in>')
    bt(indir=fd)

Hey @Pryce, I think we decided to go a different way with the workflow we needed this for. I'll double check with my team, but I'm fairly sure we never got this working.

Ah shame - oh well, thanks for following up!