clusters apply fails if context name has underscore

[carlossg]

GKE clusters typically have underscore in them (ie. gke_ACCOUNT_ZONE_CLUSTERNAME), and that causes the apply to fail

Workaround is to rename contexts - context - name in ~/.kube/config

shipperctl admin clusters apply -f
...
Joining management cluster gke_account_us-central1-a_somename to application cluster gke_account_us-central1-a_somename:
Creating or updating the cluster object for cluster gke_account_us-central1-a_somename on the management cluster... Error! Cluster.shipper.booking.com "gke_account_us-central1-a_somename" is invalid: metadata.name: Invalid value: "gke_account_us-central1-a_somename": a DNS-1123 subdomain must consist of lower case alphanumeric characters, '-' or '.', and must start and end with an alphanumeric character (e.g. 'example.com', regex used for validation is '[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*')

[kanatohodets]

Hey! Thanks!

This made me recall that when I was setting up that GKE demo, I did indeed edit my ~/.kube/config context names.

Another way to solve this is to do something like this in your clusters.yaml:

managementClusters:
- name: my-management-cluster
  context: gke_ACCOUNT_ZONE_CLUSTERNAME_FOR_MANAGEMENT
applicationClusters:
- name: my-application-cluster
  context: gke_ACCOUNT_ZONE_CLUSTERNAME_FOR_WORKLOAD

This way Shipper will use name to create the Cluster object, based on the context named in the context key.

In terms of making this obvious going forward, I think the easiest thing would be for us to put that regex in shipperctl, and it if fails to match, give a hint to add a distinct context: gke_ACCOUNT_ZONE_CLUSTERNAME from name: my-cluster in the clusters.yaml.

Does that sound sensible to you?

[carlossg]

yep that works