@michael-gvdw did you run into a specific problem, or is this a general question?

If you have a self-signed object store already set up, please test it to see if anything fails so we can document it.

Just follow the normal docs for connecting an object store and tell us which parts don't work (if any).

I managed to get it working (kinda). Since it was a ssl error I created a configmap with the .crt file. After I edited the Deployment of "ml-pipeline" in which i mounted the configmap as a volume. This way the ssl crt was available to the pod and it was able to connect successfully to the external object store.

I have two follow up questions:

1. Which resource would I need to alter in the deploykf repo in order to keep using the sync_script? (now i first deploy and edit the deployed resource)

2. For some reason unkown to me the pipeline is registerd as failed in the UI but the kubernetes pod has status completed and also the s3 bucket is populated with data of the pipeline, is there a known issue about this behaviour?