Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Prefect Community

Have a few questions on flow boilerplate and k8's templates

Some constraints:
1. We can’t pull code from Github during a flowRun
2. We don’t want to store flowRun results in a place like S3 since we plan on using prefect cloud
What storage type and flow declaration should we use if we assume our flow-run docker image has all the source code in it needed to run? And is that doable with a k8's spec that references that image (full image of resolved dependencies and flow-run tasks, modules etc)

After discussing with our SRE we basically can’t go the approach we were going to of pulling the task code from GH when the flow kicks off

Local storage using `stored_as_script=True` and `path=flow_path_in_container.py` and then it will look for the Flow relative to the image

<https://github.com/anna-geller/packaging-prefect-flows/blob/master/flows/local_script_kubernetes_run_custom_ecr_image.py|this> is a good example