You are looking at the documentation of a prior release. To read the documentation of the latest release, please
visit here.
New to KubeDB? Please start here.
Vertical Scale Kafka Combined Cluster
This guide will show you how to use KubeDB Ops-manager operator to update the resources of a Kafka combined cluster.
Before You Begin
At first, you need to have a Kubernetes cluster, and the
kubectlcommand-line tool must be configured to communicate with your cluster. If you do not already have a cluster, you can create one by using kind.Install
KubeDBProvisioner and Ops-manager operator in your cluster following the steps here.You should be familiar with the following
KubeDBconcepts:
To keep everything isolated, we are going to use a separate namespace called demo throughout this tutorial.
$ kubectl create ns demo
namespace/demo created
Note: YAML files used in this tutorial are stored in docs/examples/kafka directory of kubedb/docs repository.
Apply Vertical Scaling on Combined Cluster
Here, we are going to deploy a Kafka combined cluster using a supported version by KubeDB operator. Then we are going to apply vertical scaling on it.
Prepare Kafka Combined Cluster
Now, we are going to deploy a Kafka combined cluster database with version 3.6.1.
Deploy Kafka Combined Cluster
In this section, we are going to deploy a Kafka combined cluster. Then, in the next section we will update the resources of the database using KafkaOpsRequest CRD. Below is the YAML of the Kafka CR that we are going to create,
apiVersion: kubedb.com/v1
kind: Kafka
metadata:
name: kafka-dev
namespace: demo
spec:
replicas: 2
version: 3.6.1
storage:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 1Gi
storageClassName: standard
storageType: Durable
deletionPolicy: WipeOut
Let’s create the Kafka CR we have shown above,
$ kubectl create -f https://github.com/kubedb/docs/raw/v2024.11.8-rc.0/docs/examples/kafka/scaling/kafka-combined.yaml
kafka.kubedb.com/kafka-dev created
Now, wait until kafka-dev has status Ready. i.e,
$ kubectl get kf -n demo -w
NAME TYPE VERSION STATUS AGE
kafka-dev kubedb.com/v1 3.6.1 Provisioning 0s
kafka-dev kubedb.com/v1 3.6.1 Provisioning 24s
.
.
kafka-dev kubedb.com/v1 3.6.1 Ready 92s
Let’s check the Pod containers resources,
$ kubectl get pod -n demo kafka-dev-0 -o json | jq '.spec.containers[].resources'
{
"limits": {
"memory": "1Gi"
},
"requests": {
"cpu": "500m",
"memory": "1Gi"
}
}
This is the default resources of the Kafka combined cluster set by the KubeDB operator.
We are now ready to apply the KafkaOpsRequest CR to update the resources of this database.
Vertical Scaling
Here, we are going to update the resources of the combined cluster to meet the desired resources after scaling.
Create KafkaOpsRequest
In order to update the resources of the database, we have to create a KafkaOpsRequest CR with our desired resources. Below is the YAML of the KafkaOpsRequest CR that we are going to create,
apiVersion: ops.kubedb.com/v1alpha1
kind: KafkaOpsRequest
metadata:
name: kfops-vscale-combined
namespace: demo
spec:
type: VerticalScaling
databaseRef:
name: kafka-dev
verticalScaling:
node:
resources:
requests:
memory: "1.2Gi"
cpu: "0.6"
limits:
memory: "1.2Gi"
cpu: "0.6"
timeout: 5m
apply: IfReady
Here,
spec.databaseRef.namespecifies that we are performing vertical scaling operation onkafka-devcluster.spec.typespecifies that we are performingVerticalScalingon kafka.spec.VerticalScaling.nodespecifies the desired resources after scaling.
Let’s create the KafkaOpsRequest CR we have shown above,
$ kubectl apply -f https://github.com/kubedb/docs/raw/v2024.11.8-rc.0/docs/examples/kafka/scaling/vertical-scaling/kafka-vertical-scaling-combined.yaml
kafkaopsrequest.ops.kubedb.com/kfops-vscale-combined created
Verify Kafka Combined cluster resources updated successfully
If everything goes well, KubeDB Ops-manager operator will update the resources of Kafka object and related PetSets and Pods.
Let’s wait for KafkaOpsRequest to be Successful. Run the following command to watch KafkaOpsRequest CR,
$ kubectl get kafkaopsrequest -n demo
NAME TYPE STATUS AGE
kfops-vscale-combined VerticalScaling Successful 3m56s
We can see from the above output that the KafkaOpsRequest has succeeded. If we describe the KafkaOpsRequest we will get an overview of the steps that were followed to scale the cluster.
$ kubectl describe kafkaopsrequest -n demo kfops-vscale-combined
Name: kfops-vscale-combined
Namespace: demo
Labels: <none>
Annotations: <none>
API Version: ops.kubedb.com/v1alpha1
Kind: KafkaOpsRequest
Metadata:
Creation Timestamp: 2024-08-02T05:59:06Z
Generation: 1
Resource Version: 336197
UID: 5fd90feb-eed2-4130-8762-442f2f4d2698
Spec:
Apply: IfReady
Database Ref:
Name: kafka-dev
Timeout: 5m
Type: VerticalScaling
Vertical Scaling:
Node:
Resources:
Limits:
Cpu: 0.6
Memory: 1.2Gi
Requests:
Cpu: 0.6
Memory: 1.2Gi
Status:
Conditions:
Last Transition Time: 2024-08-02T05:59:06Z
Message: Kafka ops-request has started to vertically scaling the kafka nodes
Observed Generation: 1
Reason: VerticalScaling
Status: True
Type: VerticalScaling
Last Transition Time: 2024-08-02T05:59:09Z
Message: Successfully updated PetSets Resources
Observed Generation: 1
Reason: UpdatePetSets
Status: True
Type: UpdatePetSets
Last Transition Time: 2024-08-02T05:59:14Z
Message: get pod; ConditionStatus:True; PodName:kafka-dev-0
Observed Generation: 1
Status: True
Type: GetPod--kafka-dev-0
Last Transition Time: 2024-08-02T05:59:14Z
Message: evict pod; ConditionStatus:True; PodName:kafka-dev-0
Observed Generation: 1
Status: True
Type: EvictPod--kafka-dev-0
Last Transition Time: 2024-08-02T05:59:29Z
Message: check pod running; ConditionStatus:True; PodName:kafka-dev-0
Observed Generation: 1
Status: True
Type: CheckPodRunning--kafka-dev-0
Last Transition Time: 2024-08-02T05:59:34Z
Message: get pod; ConditionStatus:True; PodName:kafka-dev-1
Observed Generation: 1
Status: True
Type: GetPod--kafka-dev-1
Last Transition Time: 2024-08-02T05:59:34Z
Message: evict pod; ConditionStatus:True; PodName:kafka-dev-1
Observed Generation: 1
Status: True
Type: EvictPod--kafka-dev-1
Last Transition Time: 2024-08-02T06:00:59Z
Message: check pod running; ConditionStatus:True; PodName:kafka-dev-1
Observed Generation: 1
Status: True
Type: CheckPodRunning--kafka-dev-1
Last Transition Time: 2024-08-02T06:01:04Z
Message: Successfully Restarted Pods With Resources
Observed Generation: 1
Reason: RestartPods
Status: True
Type: RestartPods
Last Transition Time: 2024-08-02T06:01:04Z
Message: Successfully completed the vertical scaling for kafka
Observed Generation: 1
Reason: Successful
Status: True
Type: Successful
Observed Generation: 1
Phase: Successful
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Starting 2m38s KubeDB Ops-manager Operator Start processing for KafkaOpsRequest: demo/kfops-vscale-combined
Normal Starting 2m38s KubeDB Ops-manager Operator Pausing Kafka databse: demo/kafka-dev
Normal Successful 2m38s KubeDB Ops-manager Operator Successfully paused Kafka database: demo/kafka-dev for KafkaOpsRequest: kfops-vscale-combined
Normal UpdatePetSets 2m35s KubeDB Ops-manager Operator Successfully updated PetSets Resources
Warning get pod; ConditionStatus:True; PodName:kafka-dev-0 2m30s KubeDB Ops-manager Operator get pod; ConditionStatus:True; PodName:kafka-dev-0
Warning evict pod; ConditionStatus:True; PodName:kafka-dev-0 2m30s KubeDB Ops-manager Operator evict pod; ConditionStatus:True; PodName:kafka-dev-0
Warning check pod running; ConditionStatus:False; PodName:kafka-dev-0 2m25s KubeDB Ops-manager Operator check pod running; ConditionStatus:False; PodName:kafka-dev-0
Warning check pod running; ConditionStatus:True; PodName:kafka-dev-0 2m15s KubeDB Ops-manager Operator check pod running; ConditionStatus:True; PodName:kafka-dev-0
Warning get pod; ConditionStatus:True; PodName:kafka-dev-1 2m10s KubeDB Ops-manager Operator get pod; ConditionStatus:True; PodName:kafka-dev-1
Warning evict pod; ConditionStatus:True; PodName:kafka-dev-1 2m10s KubeDB Ops-manager Operator evict pod; ConditionStatus:True; PodName:kafka-dev-1
Warning check pod running; ConditionStatus:False; PodName:kafka-dev-1 2m5s KubeDB Ops-manager Operator check pod running; ConditionStatus:False; PodName:kafka-dev-1
Warning check pod running; ConditionStatus:True; PodName:kafka-dev-1 45s KubeDB Ops-manager Operator check pod running; ConditionStatus:True; PodName:kafka-dev-1
Normal RestartPods 40s KubeDB Ops-manager Operator Successfully Restarted Pods With Resources
Normal Starting 40s KubeDB Ops-manager Operator Resuming Kafka database: demo/kafka-dev
Normal Successful 40s KubeDB Ops-manager Operator Successfully resumed Kafka database: demo/kafka-dev for KafkaOpsRequest: kfops-vscale-combined
Now, we are going to verify from one of the Pod yaml whether the resources of the combined cluster has updated to meet up the desired state, Let’s check,
$ kubectl get pod -n demo kafka-dev-1 -o json | jq '.spec.containers[].resources'
{
"limits": {
"cpu": "600m",
"memory": "1288490188800m"
},
"requests": {
"cpu": "600m",
"memory": "1288490188800m"
}
}
The above output verifies that we have successfully scaled up the resources of the Kafka combined cluster.
Cleaning Up
To clean up the Kubernetes resources created by this tutorial, run:
kubectl delete mg -n demo kafka-dev
kubectl delete kafkaopsrequest -n demo kfops-vscale-combined
kubectl delete ns demo
Next Steps
- Detail concepts of Kafka object.
- Different Kafka topology clustering modes here.
- Monitor your Kafka database with KubeDB using out-of-the-box Prometheus operator.
- Want to hack on KubeDB? Check our contribution guidelines.