k8s HELM 安装Kafka Zookeeper集群

云计算 2023-07-27 张二河 手机阅读

很早之前文章介绍了通过Kafka 二进制安装集群，目前很多环境都是Kubernetes，追求快速部署、快速创建项目。 下面我们通过helm快速构建一套Kafka集群并配置持久化

关于k8s sc持久化和Kafka二进制安装，此处就不在介绍了，可以参考下面的文章

Kafka容器化会受底层物理机的配置影响，大并发常景还是要慎重考虑

Helm 安装

相关服务版本

  • Kubernetes 1.24.0
  • Containerd v1.6.4
  1. # 下载
  2. wget https://get.helm.sh/helm-v3.6.1-linux-amd64.tar.gz
  4. # 解压
  5. tar zxvf helm-v3.6.1-linux-amd64.tar.gz
  7. # 安装
  8. mv linux-amd64/helm /usr/local/bin/
  10. # 验证
  11. helm version

Helm 部署Zookeeper集群

  1. # 添加bitnami仓库
  2. helm repo add bitnami https://charts.bitnami.com/bitnami
  4. # 查询chart
  5. helm search repo bitnami
  7. # 拉取zookeeper
  8. helm pull bitnami/zookeeper
  10. # 解压
  11. tar zxvf zookeeper-11.4.2.tgz
  13. #进入Zookeeper
  14. cd zookeeper

接下来对Zookeeper进行时区、持久化存储、副本数等配置

  1. extraEnvVars: 
  2.   - name: TZ
  3.     value: "Asia/Shanghai"
  5. # 允许任意用户连接（默认开启）
  6. allowAnonymousLogin: true
  7. ---
  9. # 关闭认证（默认关闭）
  10. auth:
  11.   enable: false 
  12. ---
  14. # 修改副本数
  15. replicaCount: 3 
  16. ---
  18. # 4. 配置持久化，按需使用
  19. persistence:
  20.   enabled: true
  21.   storageClass: "rook-ceph-block"  # storageClass 如果有默认存储可以不写
  22.   accessModes:
  23.     - ReadWriteOnce

创建Kafka namespace

  1. [root@k8s-02 zookeeper]# kubectl create ns kafka

helm创建Zookeeper集群

  1. [root@k8s-02 zookeeper]# helm install zookeeper -n kafka .
  2. #此处下面环境版本信息
  3. NAME: zookeeper
  4. LAST DEPLOYED: Tue May 23 13:40:12 2023
  5. NAMESPACE: kafka
  6. STATUS: deployed
  7. REVISION: 1
  8. TEST SUITE: None
  9. NOTES:
  10. CHART NAME: zookeeper
  11. CHART VERSION: 11.4.2
  12. APP VERSION: 3.8.1
  14. #下面为相关注释，后面可以通过下面的命令查看Zookeeper集群状态
  15. ** Please be patient while the chart is being deployed **
  17. ZooKeeper can be accessed via port 2181 on the following DNS name from within your cluster:
  19.     zookeeper.kafka.svc.cluster.local
  21. To connect to your ZooKeeper server run the following commands:
  23.     export POD_NAME=$(kubectl get pods --namespace kafka -l "app.kubernetes.io/name=zookeeper,app.kubernetes.io/instance=zookeeper,app.kubernetes.io/component=zookeeper" -o jsonpath="{.items[0].metadata.name}")
  24.     kubectl exec -it $POD_NAME -- zkCli.sh
  26. To connect to your ZooKeeper server from outside the cluster execute the following commands:
  28.     kubectl port-forward --namespace kafka svc/zookeeper 2181:2181 &
  29.     zkCli.sh 127.0.0.1:2181
  30. [root@k8s-02 zookeeper]#

检查pod状态

  1. [root@k8s-02 zookeeper]# kubectl get all -n kafka
  2. NAME              READY   STATUS    RESTARTS   AGE
  3. pod/zookeeper-0   1/1     Running   0          52s
  4. pod/zookeeper-1   1/1     Running   0          51s
  5. pod/zookeeper-2   1/1     Running   0          49s
  7. NAME                         TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)                      AGE
  8. service/zookeeper            ClusterIP   10.110.142.203   <none>        2181/TCP,2888/TCP,3888/TCP   52s
  9. service/zookeeper-headless   ClusterIP   None             <none>        2181/TCP,2888/TCP,3888/TCP   52s
  11. NAME                         READY   AGE
  12. statefulset.apps/zookeeper   3/3     52s

检查pvc

  1. [root@k8s-02 zookeeper]# kubectl get pvc  |grep zook
  2. data-zookeeper-0   Bound    pvc-997a81c1-6986-4620-88f4-2270247354f5   8Gi        RWO            nfs-storage    7d23h
  3. data-zookeeper-1   Bound    pvc-a6012ebb-1f70-43d1-ac1b-8deaec660efe   8Gi        RWO            nfs-storage    7d23h
  4. data-zookeeper-2   Bound    pvc-f6300de4-8cd9-4807-a5fd-2655deb05139   8Gi        RWO            nfs-storage    7d23h
  5. [root@k8s-02 zookeeper]#

检查Zookeeper集群状态

  1. [root@k8s-02 zookeeper]# kubectl exec -it -n kafka zookeeper-0 -- bash
  2. I have no name!@zookeeper-0:/$
  3. I have no name!@zookeeper-0:/$ zkServer.sh status
  4. /opt/bitnami/java/bin/java
  5. ZooKeeper JMX enabled by default
  6. Using config: /opt/bitnami/zookeeper/bin/../conf/zoo.cfg
  7. Client port found: 2181. Client address: localhost. Client SSL: false.
  8. Mode: follower
  9. I have no name!@zookeeper-0:/$

Helm 部署Kafka集群

拉取kafka

  1. helm pull bitnami/kafka

解压kafka

  1. [root@k8s-02 ~]# tar xf kafka-22.1.3.tgz

进入Kafka目录

  1. [root@k8s-02 ~]# cd kafka
  2. [root@k8s-02 kafka]# ls
  3. Chart.lock  charts  Chart.yaml  README.md  templates  values.yaml

修改values.yaml

  1. extraEnvVars: 
  2.   - name: TZ
  3.     value: "Asia/Shanghai"
  4. ---
  6. # 副本数
  7. replicaCount: 3                    # 副本数
  8. ---
  10. # 持久化存储
  11. persistence:
  12.   enabled: true
  13.   storageClass: "rook-ceph-block"  # sc 有默认sc可以不写
  14.   accessModes:
  15.     - ReadWriteOnce
  16.   size: 8Gi
  17. ---
  18. kraft:
  19.   ## @param kraft.enabled Switch to enable or disable the Kraft mode for Kafka
  20.   ##
  21.   enabled: false   #设置为false
  22. ---
  23. # 配置zookeeper外部连接
  24. zookeeper:
  25.   enabled: false                   # 不使用内部zookeeper，默认是false
  27. externalZookeeper:                 # 外部zookeeper
  28.   servers: zookeeper            #Zookeeper svc名称

可选配置

  1. ## 允许删除topic（按需开启）
  2. deleteTopicEnable: true
  4. ## 日志保留时间（默认一周）
  5. logRetentionHours: 168
  7. ## 自动创建topic时的默认副本数
  8. defaultReplicationFactor: 2
  10. ## 用于配置offset记录的topic的partition的副本个数
  11. offsetsTopicReplicationFactor: 2
  13. ## 事务主题的复制因子
  14. transactionStateLogReplicationFactor: 2
  16. ## min.insync.replicas
  17. transactionStateLogMinIsr: 2
  19. ## 新建Topic时默认的分区数
  20. numPartitions: 3

创建Kafka集群

  1. [root@k8s-02 kafka]# helm install kafka -n kafka .
  3. #输出结果如下
  4. [root@k8s-02 kafka]# helm install kafka -n kafka .
  5. W0523 13:52:58.673090   28827 warnings.go:70] spec.template.spec.containers[0].env[39].name: duplicate name "KAFKA_ENABLE_KRAFT"
  6. NAME: kafka
  7. LAST DEPLOYED: Tue May 23 13:52:58 2023
  8. NAMESPACE: kafka
  9. STATUS: deployed
  10. REVISION: 1
  11. TEST SUITE: None
  12. NOTES:
  13. CHART NAME: kafka
  14. CHART VERSION: 22.1.3
  15. APP VERSION: 3.4.0
  17. ** Please be patient while the chart is being deployed **
  19. Kafka can be accessed by consumers via port 9092 on the following DNS name from within your cluster:
  21.     kafka.kafka.svc.cluster.local
  23. Each Kafka broker can be accessed by producers via port 9092 on the following DNS name(s) from within your cluster:
  25.     kafka-0.kafka-headless.kafka.svc.cluster.local:9092
  26.     kafka-1.kafka-headless.kafka.svc.cluster.local:9092
  27.     kafka-2.kafka-headless.kafka.svc.cluster.local:9092
  29. To create a pod that you can use as a Kafka client run the following commands:
  31.     kubectl run kafka-client --restart='Never' --image docker.io/bitnami/kafka:3.4.0-debian-11-r33 --namespace kafka --command -- sleep infinity
  32.     kubectl exec --tty -i kafka-client --namespace kafka -- bash
  34.     PRODUCER:
  35.         kafka-console-producer.sh \
  36.             --broker-list kafka-0.kafka-headless.kafka.svc.cluster.local:9092,kafka-1.kafka-headless.kafka.svc.cluster.local:9092,kafka-2.kafka-headless.kafka.svc.cluster.local:9092 \
  37.             --topic test
  39.     CONSUMER:
  40.         kafka-console-consumer.sh \
  41.             --bootstrap-server kafka.kafka.svc.cluster.local:9092 \
  42.             --topic test \
  43.             --from-beginning

进入Kafka集群，创建topic查看

  1. ##进入Kafka集群
  2. kubectl exec -it -n kafka kafka-0 -- bash
  4. #创建topic
  5. kafka-topics.sh --create --bootstrap-server kafka:9092  --topic abcdocker
  7. #查看topic列表
  8. kafka-topics.sh --list --bootstrap-server kafka:9092 
  10. #查看topic详细信息
  11. kafka-topics.sh --bootstrap-server kafka:9092  --describe --topic abcdocker
  13. #配置文件配置已经生效，默认分区为3，副本为3，过期时间为168小时
  15. I have no name!@kafka-0:/$ kafka-topics.sh --bootstrap-server kafka:9092  --describe --topic abcdocker
  16. Topic: abcdocker    TopicId: jcJtxY1NSr-nSloax8oPnA PartitionCount: 3   ReplicationFactor: 3    Configs: flush.ms=1000,segment.bytes=1073741824,flush.messages=10000,max.message.bytes=1000012,retention.bytes=1073741824
  17.     Topic: abcdocker    Partition: 0    Leader: 1   Replicas: 1,2,0 Isr: 1,2,0
  18.     Topic: abcdocker    Partition: 1    Leader: 0   Replicas: 0,1,2 Isr: 0,1,2
  19.     Topic: abcdocker    Partition: 2    Leader: 2   Replicas: 2,0,1 Isr: 2,0,1