We have been using ArgoCD in a proof of concept EKS Cluster (running 1.18) and yesterday we tried to update it from ArgoCD 1.7.10 to 1.8.7. Our first attempt was just by doing:
kubectl apply -n argocd -f https://raw.githubusercontent.com/argoproj/argo-cd/v1.8.7/manifests/install.yaml
That just hung, without giving any output at all. Since then we've established that the first thing that manifest tries to install is a couple of CRDs and any attempt to install any CRDs has been very unreliable since. Even trying to install the basic CRD from https://www.eksworkshop.com/intermediate/270_custom_resource_definition/creating_crd/ will timeout, but only sometimes.
We deleted the entire argocd namespace and were unable to reinstall ArgoCD using:
kubectl apply -n argocd -f https://raw.githubusercontent.com/argoproj/argo-cd/v1.7.10/manifests/install.yaml
it would just hang in the same way.
I extracted the CRD from the full manifest and tried applying that on its own - that does at least give a timeout error from kubectl. And I've tried stripping down the CRD (to a minimal schema and removing optional components) which makes no difference.
I can't find any reason for the error and can't find anything in the logs on AWS. So I'm a bit stumped as to where to look for a reason for the problem.
Does anyone have any suggestion as to what the problem could be or where I can find more clues?
Thanks
