Hey guys, having a problem using the

prefecthq/prefect:master

in flows - it has a feature I need [DFE execution on mapped tasks], but I can’t seem to figure out how to actually use master. I’ve updated both my Prefect Server and the image I base all our flows off [

FROM prefecthq/prefect:master

] but even with all that I still see the following when I build the flow storage:

Step 3/9 : RUN pip show prefect || pip install git+<https://github.com/PrefectHQ/prefect.git@0.11.5#egg=prefect[kubernetes]>

And in the UI it still reports 0.11.5 as Prefect Core Version, and the expected DFE mapping behaviour is not there, i.e. mapped tasks are executed one by one. .Any ideas on how to get around this?

🤔 If you provide

prefect_version="master"

to your Docker storage does it take?

Good spot, missed that one…

Also, are you on master locally? I can’t exactly recall if it’s the local version which determine’s your flow’s version or not

Not on master locally, but my assumption was that it’s the one built into docker storage that’s important?

Yes! It is supported through the GraphQL API and you can set it there if calling a mutation directly, when running in the UI there’s a box for it, or on the Client function

create_flow_run

it has

run_name: str

as a kwarg

Sorry for the delay @josh , completely lost track - I'm looking more for a naming schema I cna supply programmatically, as part of the flow?

Ah I actually don’t think that is supported. There is the flow’s

name

but I think I follow what you’re saying. A way for a name schema to be respected for auto generated flows?

Yeah pretty much