1

I'm trying to delete the argo app with argocd app delete Prometheus --cascade command.

It's prompting for confirmation after pressing y able to delete resources but some of the resources are not deleting.

Are you sure you want to delete 'prometheus' and all its resources? [y/n]

How can I confirm that some of the resources are not deleted ?

kubectl get all -n prometheus

NAME READY AGE statefulset.apps/alertmanager-prometheus-2-kube-promethe-alertmanager 1/1 39m

What I'm looking for ?

I need the solution for delete the all resources with argocd

user28434'mstep
  • 6,290
  • 2
  • 20
  • 35
Hack-Z
  • 47
  • 7

1 Answers1

1

Usually this is due to child resources created by an application that are pending deletion (probably due to a finalizer on that resource).
As a matter of fact, k get all wont show you everything at all; it wont show secrets, configmaps, serviceaccounts etc, and of course not any CR used by Prometheus. So to investigate further and see those resources that are still there, you will need to query specifically using k get <type>.
Or, use a hammer and query everything in that namespace

thikade
  • 370
  • 2
  • 4
  • It was conflicted because of the kube-promitheus stack, It's created the Long CRD Name that cause for breaking the application. I notice if something went wrong while deploying application via ArgoCD then Argo will fail to delete all the resources. I can't dig more now, shifted to another stack. Thanks @thikade – Hack-Z Jul 13 '22 at 16:08