K8SPS-212: Add validation rules

[inelpandzic]

CHANGE DESCRIPTION

Problem: Some CR fields and their values are related in the sense when one field value is specified, some other should be a certain value. For example, if we set clusterType=group-replication then haproxy.enabled should be false. Now you can set it to enabled, but HAProxy would not be started at all since it does not work with GR cluster type.

Cause: Lack of validation when applying the CR.

Solution: Utilise CRD Validation Rules to validate CR upon creation and reject the create request if certain rules are not met.

Now when the user wants to create a CR with GR cluster type and he enables HAProxy, he will get a response like this.

➜  k apply -f deploy/cr.yaml
The PerconaServerMySQL "cluster1" is invalid: spec: Invalid value: "object": HAProxy can not be enabled with 'group-replication' clusterType

CHECKLIST

Jira

• [x] Is the Jira ticket created and referenced properly?
• [x] Does the Jira ticket have the proper statuses for documentation (Needs Doc) and QA (Needs QA)?
• [x] Does the Jira ticket link to the proper milestone (Fix Version field)?

Tests

• [ ] Is an E2E test/test case added for the new feature/change?
• [ ] Are unit tests added where appropriate?

Config/Logging/Testability

• [x] Are all needed new/changed options added to default YAML files?
• [x] Are the manifests (crd/bundle) regenerated if needed?
• [x] Did we add proper logging messages for operator actions?
• [x] Did we ensure compatibility with the previous version or cluster upgrade process?
• [x] Does the change support oldest and newest supported PS version?
• [x] Does the change support oldest and newest supported Kubernetes version?

[JNKPercona]

Test name	Status
version-service	failure
async-ignore-annotations	failure
auto-config	failure
config	failure
config-router	failure
demand-backup	failure
gr-demand-backup	failure
gr-demand-backup-haproxy	failure
gr-finalizer	failure
gr-haproxy	failure
gr-ignore-annotations	failure
gr-init-deploy	failure
gr-one-pod	failure
gr-recreate	failure
gr-scaling	failure
gr-scheduled-backup	failure
gr-security-context	failure
gr-self-healing	failure
gr-tls-cert-manager	failure
gr-users	failure
haproxy	failure
init-deploy	failure
limits	failure
monitoring	failure
one-pod	failure
operator-self-healing	failure
recreate	failure
scaling	failure
scheduled-backup	failure
service-per-pod	failure
sidecars	failure
smart-update	failure
tls-cert-manager	failure
users	failure
We run 34 out of 34

commit: https://github.com/percona/percona-server-mysql-operator/pull/308/commits/a4df79110848973bd012ab8e61c837b9fed7af7d image: perconalab/percona-server-mysql-operator:PR-308-a4df7911