Can't delete rabbitmq-cluster

[marczahn]

I am not sure whether this is a bug or I am doing sth. wrong. The strange thing is it worked before:

I set up a rabbitmq cluster

apiVersion: rabbitmq.com/v1beta1
kind: RabbitmqCluster
metadata:
  name: new-rabbitmq
spec:
  replicas: 1
  rabbitmq:
    additionalPlugins:
      - rabbitmq_federation
      - rabbitmq_federation_management
      - rabbitmq_management
  service:
    type: ClusterIP

Then, for some test purposes, I added another cluster:

apiVersion: rabbitmq.com/v1beta1
kind: RabbitmqCluster
metadata:
  name: rabbitmq
spec:
  replicas: 5
  rabbitmq:
    additionalPlugins:
      - rabbitmq_federation
      - rabbitmq_federation_management
      - rabbitmq_management
  service:
    type: ClusterIP

Now, I want to delete new-rabbitmq. I tried out some things earlier and back then I could start deleting the services, then the pods and so on. But now, it doesn't work. The services come back again and again.

[lukebakken]

Just FYI, if a github repository has both Issues and Discussions available, always use Discussions unless you are certain you have found a bug, and provide information to reproduce it.

How do you expect us to continue given the information you have provided?

[marczahn]

unless you are certain you have found a bug

Sorry for that.

How do you expect us to continue given the information you have provided?

What further information do you need? I'd love to provide you with what you need. Or you tell me what I am doing wrong. Or what I missed in the documentation.

[lukebakken]

A good starting point would be:

• A transcript of all the commands you ran (since we have no idea what you have done), starting with the creation of the first cluster, and their full output.
• Logs while running those commands.

I'm not a k8s expert, but I would guess that the k8s logs show why it thinks it should re-create the new-rabbitmq cluster even though you are trying to delete it.

The strange thing is it worked before

"it worked" - what worked? Creating the first single-node environment, then the five-nodes, then deleting the five-node cluster?

[lukebakken]

Here is what I tried:

https://github.com/lukebakken/cluster-operator-1410#reproduction-attempt

[marczahn]

"it worked" - what worked? Creating the first single-node environment, then the five-nodes, then deleting the five-node cluster?
Means, I deleted the services and then all belonging resources. At this time, the resources were not recreated. Now they do.

Here is what I tried:

https://github.com/lukebakken/cluster-operator-1410#reproduction-attempt

That worked! I am using Kustomize. So I created a template-yaml and ran that delete command. Thanks for that! I think that's a way cleaner way than what I did.