Use dns cluster info from lib common get function

[averdagu]

Openshift coreDNS creates the domain name using an string located in dnses.operator.openshift.io. This string can change in the future, calling lib-common/GetDNSClusterDomain the responsability of gathering this information correctly only falls under lib-common intead of all operators.

Resolves: OSPRH-3627
Depends-on: openstack-k8s-operators/lib-common#580

[softwarefactory-project-zuul]

Build failed (check pipeline). Post recheck (without leading slash)
to rerun all jobs. Make sure the failure cause has been resolved before
you rerun jobs.

https://softwarefactory-project.io/zuul/t/rdoproject.org/buildset/9a5458b6a1b54dd5a4786bf7dc0412cd

❌ openstack-k8s-operators-content-provider FAILURE in 6m 23s
⚠️ ovn-operator-tempest-multinode SKIPPED Skipped due to failed job openstack-k8s-operators-content-provider

[stuggi]

do we need this right now?

[averdagu]

Hmm.. I don't think we need it right now.
The idea behind watching this resource (On line 101) is to have the ovn-operator ready in case clusterDomain is parameterizable. If this resource is not watched, it will only be gathered when the ovn-operator is reconciled (for external reasons).

[stuggi]

but line 101 would not watch it. it just adds the permissions to be able to watch it. In my opinion we should not watch objects we don't use right now. we should add it later.

[averdagu]

Oh true, I forgot to add the watch part when I was copying the code. Instead of watching I will add a //TODO that if clusterLocal is parameterizable we need to watch the resource

[softwarefactory-project-zuul]

This change depends on a change that failed to merge.

Change openstack-k8s-operators/lib-common#580 is needed.

[booxter]

Code is fine but I think you need to pull the library in go.mod.

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: averdagu

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [averdagu]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[softwarefactory-project-zuul]

Build failed (check pipeline). Post recheck (without leading slash)
to rerun all jobs. Make sure the failure cause has been resolved before
you rerun jobs.

https://softwarefactory-project.io/zuul/t/rdoproject.org/buildset/9cca9f92a0de4f16a66c5d7baf3fe844

❌ openstack-k8s-operators-content-provider FAILURE in 6m 49s
⚠️ ovn-operator-tempest-multinode SKIPPED Skipped due to failed job openstack-k8s-operators-content-provider

[softwarefactory-project-zuul]

Build failed (check pipeline). Post recheck (without leading slash)
to rerun all jobs. Make sure the failure cause has been resolved before
you rerun jobs.

https://softwarefactory-project.io/zuul/t/rdoproject.org/buildset/d33f3a22d1a24a48ab232f7296bc21ed

❌ openstack-k8s-operators-content-provider FAILURE in 10m 44s
⚠️ ovn-operator-tempest-multinode SKIPPED Skipped due to failed job openstack-k8s-operators-content-provider

[karelyatin]

/retest

[karelyatin]

Also need to fix in openstack-operator to clear this atleast DNSSuffix usage

$ grep -r -e ClusterInternalDomain -e cluster.local -e DNSSuffix
grep: bin/k8s/1.29.3-linux-amd64/kube-apiserver: binary file matches
grep: bin/k8s/1.29.3-linux-amd64/kubectl: binary file matches
grep: .git/objects/pack/pack-c8e17e8731cb1dde52601f13e680585a00ec8588.pack: binary file matches
pkg/openstack/galera.go: fmt.Sprintf("%s.%s", hostname, ClusterInternalDomain),
pkg/openstack/galera.go: fmt.Sprintf("%s.%s", hostnameHeadless, ClusterInternalDomain),
pkg/openstack/galera.go: fmt.Sprintf(".%s.%s", hostnameHeadless, ClusterInternalDomain),
pkg/openstack/galera.go: Organizations: []string{fmt.Sprintf("%s.%s", instance.Namespace, ClusterInternalDomain)},
pkg/openstack/nova.go: fmt.Sprintf("%s.%s", hostname, ClusterInternalDomain),
pkg/openstack/nova.go: Organizations: []string{fmt.Sprintf("%s.%s", instance.Namespace, ClusterInternalDomain)},
pkg/openstack/neutron.go: fmt.Sprintf("%s.%s.svc.%s", serviceName, instance.Namespace, "cluster.local"),
pkg/openstack/memcached.go: fmt.Sprintf("%s.%s.svc.%s", name, instance.Namespace, ClusterInternalDomain),
pkg/openstack/memcached.go: fmt.Sprintf(".%s.%s.svc.%s", name, instance.Namespace, ClusterInternalDomain),
pkg/openstack/common.go: // ClusterInternalDomain - cluster internal dns domain
pkg/openstack/common.go: ClusterInternalDomain = "cluster.local"
pkg/openstack/common.go: fmt.Sprintf("%s.%s.svc.%s", ed.Name, instance.Namespace, ClusterInternalDomain),
pkg/openstack/common.go: fmt.Sprintf("%s.%s.svc.%s", ed.Name, instance.Namespace, ClusterInternalDomain),
pkg/openstack/rabbitmq.go: fmt.Sprintf("%s.%s", hostname, ClusterInternalDomain),
pkg/openstack/rabbitmq.go: fmt.Sprintf("%s.%s", hostnameHeadless, ClusterInternalDomain),
pkg/openstack/rabbitmq.go: Organizations: []string{fmt.Sprintf("%s.%s", rabbitmq.Namespace, ClusterInternalDomain)},
pkg/openstack/redis.go: fmt.Sprintf("redis-%s.%s.svc.%s", name, instance.Namespace, ClusterInternalDomain),
pkg/openstack/redis.go: fmt.Sprintf(".redis-%s.%s.svc.%s", name, instance.Namespace, ClusterInternalDomain),
pkg/openstack/redis.go: Organizations: []string{fmt.Sprintf("%s.%s", instance.Namespace, ClusterInternalDomain)},
pkg/openstack/ovn.go: fmt.Sprintf(".%s.svc.%s", instance.Namespace, ovnv1.DNSSuffix),
pkg/openstack/ovn.go: fmt.Sprintf("%s.%s.svc.%s", serviceName, instance.Namespace, ovnv1.DNSSuffix),
pkg/openstack/ovn.go: fmt.Sprintf("%s.%s.svc.%s", serviceName, instance.Namespace, ovnv1.DNSSuffix),
pkg/openstack/octavia.go: fmt.Sprintf("%s.%s.svc.%s", serviceName, instance.Namespace, ClusterInternalDomain),
config/certmanager/certificate.yaml: - $(SERVICE_NAME).$(SERVICE_NAMESPACE).svc.cluster.local

[averdagu]

Problem seen in this commit should be solved once this merges openstack-k8s-operators/openstack-operator#1215

[booxter]

/lgtm