Lab 14 - Strimzi 0.18.0 / AMQ Streams 1.5.0

Install cluster operator

• Create namespace/project myproject and set it as default.
• On Openshift
  • oc new-project myproject
  • oc project myproject
• On Kubernetes
  • kubectl create ns myproject
  • kubectl config set-context --current --namespace=myproject

On OCP4 and OCP 3.11

• Install the operator:
  • kubectl apply -f amq-streams-1.5.0/

On Kubernetes

• Install the operator:
  • kubectl apply -f strimzi-0.18.0/

Improved TLS Configuration

• Check the configuration of Kafka in the kafka-tls-settings.yaml
  • Notice the settings in .spec.kafka.config which configures TLSv1.2 as the only allowed TLS/SSL protocol:

      ssl.protocol: TLSv1.2
      ssl.enabled.protocols: TLSv1.2

• Deploy the Kafka cluster

  • kubectl apply -f kafka-tls-settings.yaml
  • Wait for it to be ready

• Check the configuration of Kafka Connect in the kafka-connect-tls-settings.yaml

  • Notice the settings in .spec.config which configures TLSv1.1 as the only allowed TLS/SSL protocol:

      ssl.protocol: TLSv1.1
      ssl.enabled.protocols: TLSv1.1

• Deploy the Kafka Connect cluster
  • kubectl apply -f kafka-connect-tls-settings.yaml
• Because we configured Kafka brokers to use TLSv1.2 and Connect to use TLSv1.1, they should never connect and the Connect pod should be looping
• Edit the Connect CR (kubectl edit kc my-connect-cluster) and change the options to TLSv1.2
  • Watch as the operator applies the changes and gets the Connect deployment running

Operator Metrics

• On OpenShift, install the Prometheus operators from Operator Hub and deploy Grafana and Prometheus
  • kubectl apply -f metrics/prometheus-server.yaml
  • kubectl apply -f metrics/grafana-operator.yaml
  • kubectl apply -f metrics/grafana-server.yaml
• On Kubernetes, install the Grafana and Prometheus operators from YAMLs and deploy Grafana and Prometheus
  • kubectl apply -f metrics/prometheus-operator.yaml
  • kubectl apply -f metrics/prometheus-server.yaml
  • kubectl apply -f metrics/grafana-operator.yaml
  • kubectl apply -f metrics/grafana-server.yaml
• Open the Grafana using the OpenShift Route or using port-forward
  • Check the dashboard with operator metrics
  • Look at the different metrics
• Deploy the example clients with topics and users
  • kubectl apply -f client-applications.yaml
• Check how the new users and topics show up in the dashboard

Cluster Balancing / Cruise Control

• Scale the Kafka broker to 3 nodes
  • Edit the Kafka CR with kubectl edit kafka my-cluster and set .spec.kafka.replicas to 3
  • Wait until the new nodes are started and ready
• Check the existing topics and notice how they are all on node 0, since they were created when we had only 1 broker
  • kubectl exec -ti my-cluster-kafka-0 -c kafka -- bin/kafka-topics.sh --bootstrap-server localhost:9092 --describe
• Deploy Cruise Control
  • Edit the Kafka CR with kubectl edit kafka my-cluster and add .spec.cruiseControl section set to {}. E.g.

apiVersion: kafka.strimzi.io/v1beta1
kind: Kafka
metadata:
  name: my-cluster
spec:
  kafka:
    # ...
  zookeeper:
    # ...
  entityOperator:
    # ...
  cruiseControl: {}

• Let's get a rebalance proposal from Cruise Control. Check the KafkaRebalance resource in rebalance.yaml before applying it and look at the goals.
  • kubectl apply -f rebalance.yaml
• Wait until the proposal status is ProposalReady
• Check the status of the KafkaProposal resource
  • Notice the information about the replicas to be moved etc.
• Approve the rebalance kubectl annotate kafkarebalance my-rebalance strimzi.io/rebalance=approve
• Wait until the proposal status is Ready which means that the rebalancing is finished
• Check the new topic distribution after the rebalance
  • kubectl exec -ti my-cluster-kafka-0 -c kafka -- bin/kafka-topics.sh --bootstrap-server localhost:9092 --describe