Are there any details anyone needs me to fill in? Is this the right place to be posting this? Don’t know how we like to “bump” messages here…

hey @Mark Li ! When you deploy the Orion helm chart, are you using a

values.yaml

that sets a specific prefect base image? my first guess is that you're right, the issue is when

prefect orion start

is run, and the environment might have the wrong version of prefect

Hi Nate! Yes, I’m using a very lightly modified version of the values.yaml file here: https://github.com/PrefectHQ/prefect-helm/blob/main/charts/prefect-orion/values.yaml Only made two modifications - Adding the “primary” section for postgres to correctly identify persistence or no persistence and adding in resource limits. So looking at the values.yaml file here, the image version would be latest-python3.9 - seems funky when looking at it. What’s the current latest version? I couldn’t find it when searching.

I usually go to prefecthq/prefect on dockerhub and check out the tags to figure out what's available, for example

2.3-python3.9

is a very recent image that could be one that works for you Also just a note on helm, you only need to include the portion of the values.yaml that differs from the default chart, all the yaml you don't provide will default to the values you linked above

I’ll specifically reference that version and see if that resolves my issue.

I'm not certain on the tagging name convention, but I'd guess that the one you used might have been the latest python 3.9 image for prefect 1, causing the

no command “orion”

error

That was my initial thoughts. Wondering if there’s a way to specify latest the prefect 2 image?

you can find the latest version for each python version here on our dockerhub