Closed mdbooth closed 2 years ago
/assign
I think it's related to BFV , we can replicate openshift to our code base
I think we don't have cinder interaction at all ...which means some e2e for cinder is helpful
I think we don't have cinder interaction at all ...which means some e2e for cinder is helpful
I'm working on it! I'm looking to create a 2 node devstack deployment so we can have 2 AZs with cinder in both.
ok, so I will hold this effort and glad to provide any help if I can :)
I'm working on test support in https://github.com/kubernetes-sigs/cluster-api-provider-openstack/pull/1026.
This is done by #1030
/kind feature
Describe the solution you'd like Complex OpenStack deployments will often have many availability zones for both compute and storage, and there is no requirement for there to be a 1-1 mapping between the two. Additionally most OpenStack deployments will have more than a single volume type, for example to differentiate storage with different performance. To take advantage of this, we need to allow the user to specify the availability zone and volume type of their root volumes, while still allowing Cinder to pick defaults if they are not given explicitly.
We currently have support for this in OpenShift implemented here: https://github.com/openshift/cluster-api-provider-openstack/blob/36ecd2d1cbd8dc997e88ebbf8ad6c86ff2fb1668/pkg/apis/openstackproviderconfig/v1alpha1/types.go#L245-L252