Problem with setting up redpanda operator on a airgapped environment

[alidhl]

In our airgapped kubernetes environment, we're deploying the Redpanda custom resource managed by the Redpanda Operator. However, the CR triggers the creation of a HelmRepository resource that references a hardcoded URL:

https://charts.redpanda.com/index.yaml

Because the cluster is airgapped and has no internet access, the operator fails to fetch the Helm chart, resulting in the Redpanda CR remaining in a non-ready state.

This is the HelmRepository manifest that I'm referencing

apiVersion: source.toolkit.fluxcd.io/v1beta2
kind: HelmRepository
metadata:
  finalizers:
  - finalizers.fluxcd.io
  generation: 1
  name: redpanda-repository
  namespace: redpanda
  ownerReferences:
  - apiVersion: cluster.redpanda.com/v1alpha2
    controller: true
    kind: Redpanda
    name: redpanda
  resourceVersion: "69055321"
spec:
  interval: 30s
  provider: generic
  timeout: 60s
  url: https://charts.redpanda.com/

I have a internal helm repository but it isn't clear to us how to overwrite the hard coded url in the HelmRepository CR.

We are really interested on using redpanda operator and wanted to know if there is any suggestions on how to tackle this problem.

[david-yu]

Hi @alidhl we should probably have these documented clearly on our docs but you can see the Release note changes here that should help you configure the Operator for internal Helm and docker image repositories: https://github.com/redpanda-data/redpanda-operator/blob/main/.changes/operator/v2.3.6-24.3.3.md#changed

[chrisseto]

You could also consider using v2.4.1 which won't delegate to helm by default, so there will be no need to worry about extraneous requests 🎉

[alidhl]

Got it to work, appreciate the help thanks!