Description
I believe the issue has to do with the CRD scope change from being Namespaced to Cluster... but I could be totally wrong.
I updated from 0.0.9 using helm chart 0.2.1 to 0.0.10 using helm chart 0.4.1. I could not sync the CRD, since the scope field is immutable. So I deleted my clustersecret resources, and deleted the old CRD. Then I could successfully sync and everything appeared to be good.
I tried syncing one of my cluster-secret resources, and it would fail in Argo. I kept getting the error:
one or more objects failed to apply, reason: the server could not find the requested resource (post clustersecrets.clustersecret.io)
I could manually kubectl apply -f
the exact same yaml file and that would work, however syncing from argo seemed to indicate it could not validate the CRD.
I rolled back to 0.0.9 and rolled back the CRD and everything started working again as expected.