Rasheed Abdul-Aziz

Current state of the code: 1. There is code for success checks, and it copies forward the latest "success=true" stamped object's outputs to Runnable's Outputs. 1. Unfortunately, it also treats...

We can absolutely implement @jwntrs spec, I'll use this policy: * Output is from the most recent StampedObject where: * Success exists * Success: True * All output paths from...

Additional: * OutputPathNotSatisfiedCondition becomes a StampedObjectReady Condition * Yes we introduce StampedObjectReady * Yes that's a good top-level strategy. but implementing it in ConditionManager might change workload/deliverable behavior --- be...

Did #876 and #900 fix this? Check and resolve @squeedee

@cirocosta You want this: ```yaml resources: - name: source-provider selector: matchFields: - key: workload.spec.source.git.url operator: Exists templateRef: kind: ClusterSourceTemplate name: git-repository - name: image-builder templateRef: kind: ClusterImageTemplate name: kpack-image sources:...

Obviously the consistant test failures point to an issue with this change. We're not sure why it was introduced, but it feels valuable and would like to capture the reason...

# Pre-IPM discussion have we considered declarative "ignores": ``` ignore: - spec.syncInterval template: spec: syncInterval: 1m ``` we always apply it if not currently on the object ``` spec: syncInterval:...

This is not mergable due to realizers getting unified. Will re-implement

implemented here: https://github.com/vmware-tanzu/cartographer/pull/1040

Won't implement (at this time) Working with kpack is the common pain point, passing a secret to it's `services` key. These aren't really services, and can be implemented by using...