[system test] Update Azure KRaft pipelines to run KRaft

[see-quick]

Type of change

• Bugfix

Description

Currently, we do not run all Azure profiles as KRaft (as we should). This PR fixes such a problem.

Checklist

• Make sure all tests pass

[see-quick]

/azp list

[azure-pipelines]

CI/CD Pipelines for this repository:

• build
• system-tests
• upgrade
• helm-acceptance
• regression
• acceptance
• namespace-rbac-scope-acceptance
• namespace-rbac-scope-regression
• all-strimzi-system-tests
• feature-gates-regression
• operators-release
• operators-cve-rebuild
• kraft-regression
• bridge

[see-quick]

/azp run kraft-regression

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[scholzj]

Did I really set it only for one job and then randomly selected the same job to check it works? Sorry 🤦

[see-quick]

Did I really set it only for one job and then randomly selected the same job to check it works? Sorry 🤦

No problem, it was a bad review from my side also :/.

[see-quick]

There is a one-fail, which I fixed in the previous commit.

Conditions:
	Type: Warning
	Message: The .spec.zookeeper section in the Kafka custom resource is ignored in KRaft mode and should be removed from the custom resource.
	Type: Warning
	Message: The .spec.kafka.replicas property in the Kafka custom resource is ignored when node pools are used and should be removed from the custom resource.
	Type: Warning
	Message: The .spec.kafka.storage section in the Kafka custom resource is ignored when node pools are used and should be removed from the custom resource.
	Type: NotReady
	Message: The Kafka cluster my-cluster-74fcefe3 is invalid: [At least one KafkaNodePool with the controller role and at least one replica is required when KRaft mode is enabled]

I would not trigger the pipeline for this one change okay? :)

[im-konge]

@see-quick I think we should move the whole test to run in the KRaft mode, but for now, I guess that's good
Thanks

[see-quick]

Hmm it seems we have failed this test case but my change does not cause it

[ERROR] Errors: 
[ERROR] io.strimzi.operator.cluster.operator.assembly.KafkaAssemblyOperatorWithPoolsMockTest.testReconcileReplacesAllDeletedSecrets(VertxTestContext)
[ERROR]   Run 1: KafkaAssemblyOperatorWithPoolsMockTest.testReconcileReplacesAllDeletedSecrets(VertxTestContext) » FatalProblem Error while waiting for restarted pod my-cluster-pool-a-2 to become ready
[ERROR]   Run 2: KafkaAssemblyOperatorWithPoolsMockTest.testReconcileReplacesAllDeletedSecrets(VertxTestContext) » FatalProblem Error while waiting for restarted pod my-cluster-pool-b-3 to become ready
[ERROR]   Run 3: KafkaAssemblyOperatorWithPoolsMockTest.testReconcileReplacesAllDeletedSecrets(VertxTestContext) » FatalProblem Error while waiting for restarted pod my-cluster-pool-b-4 to become ready
[ERROR]   Run 4: KafkaAssemblyOperatorWithPoolsMockTest.testReconcileReplacesAllDeletedSecrets(VertxTestContext) » FatalProblem Error while waiting for restarted pod my-cluster-pool-b-4 to become ready
[ERROR]   Run 5: KafkaAssemblyOperatorWithPoolsMockTest.testReconcileReplacesAllDeletedSecrets(VertxTestContext) » FatalProblem Error while waiting for restarted pod my-cluster-pool-a-2 to become ready
[ERROR]   Run 6: KafkaAssemblyOperatorWithPoolsMockTest.testReconcileReplacesAllDeletedSecrets(VertxTestContext) » FatalProblem Error while waiting for restarted pod my-cluster-pool-a-2 to become ready

even 6 runs are not enough to get this test case passed. I would re-trigger the build, but we should fix such a test. :) I don't know if this is happening on other PRs but I would assume yes.