本文演示如何在K8s集群下部署Kafka集群,并且搭建后除了可以K8s内部访问Kafka服务,也支持从K8s集群外部访问Kafka服务。服务的集群部署通常有两种方式:一种是 StatefulSet,另一种是 Service&Deployment。本次我们使用 StatefulSet 方式搭建 ZooKeeper 集群,使用 Service&Deployment 搭建 Kafka 集群。
一、创建 NFS 存储
NFS 存储主要是为了给 Kafka、ZooKeeper 提供稳定的后端存储,当 Kafka、ZooKeeper 的 Pod 发生故障重启或迁移后,依然能获得原先的数据。
1,安装 NFS
这里我选择在 master 节点创建 NFS 存储,首先执行如下命令安装 NFS:
yum -y install nfs-utils
yum -y install rpcbind
2,创建共享文件夹
(1)执行如下命令创建 6 个文件夹:
mkdir -p /usr/local/k8s/zookeeper/pv{1..3}
mkdir -p /usr/local/k8s/kafka/pv{1..3}
(2)编辑 /etc/exports 文件:
vim /etc/exports
(3)在里面添加如下内容:
/usr/local/k8s/kafka/pv1 *(rw,sync,no_root_squash)
/usr/local/k8s/kafka/pv2 *(rw,sync,no_root_squash)
/usr/local/k8s/kafka/pv3 *(rw,sync,no_root_squash)
/usr/local/k8s/zookeeper/pv1 *(rw,sync,no_root_squash)
/usr/local/k8s/zookeeper/pv2 *(rw,sync,no_root_squash)
/usr/local/k8s/zookeeper/pv3 *(rw,sync,no_root_squash)
(4)保存退出后执行如下命令重启服务:
如果执行 systemctl restart nfs 报"Failed to restart nfs.service: Unit nfs.service not found."错误,可以尝试改用如下命令:
- sudo service nfs-server start
systemctl restart rpcbind
systemctl restart nfs
systemctl enable nfs
(5)执行 exportfs -v 命令可以显示出所有的共享目录:
(6)而其他的 Node 节点上需要执行如下命令安装 nfs-utils 客户端:
yum -y install nfs-util
(7)然后其他的 Node 节点上可执行如下命令(ip 为 Master 节点 IP)查看 Master 节点上共享的文件夹:
showmount -e 107.106.37.33(nfs服务端的IP)
二、创建 ZooKeeper 集群
1,创建 ZooKeeper PV
(1)首先创建一个 zookeeper-pv.yaml 文件,内容如下:
注意:170.106.37.33 需要改成实际 NFS 服务器地址:
apiVersion: v1
kind: PersistentVolume
metadata:
name: k8s-pv-zk01
labels:
app: zk
annotations:
volume.beta.kubernetes.io/storage-class: "anything"
spec:
capacity:
storage: 1Gi
accessModes:
- ReadWriteOnce
nfs:
server: 170.106.37.33
path: "/usr/local/k8s/zookeeper/pv1"
persistentVolumeReclaimPolicy: Recycle
---
apiVersion: v1
kind: PersistentVolume
metadata:
name: k8s-pv-zk02
labels:
app: zk
annotations:
volume.beta.kubernetes.io/storage-class: "anything"
spec:
capacity:
storage: 1Gi
accessModes:
- ReadWriteOnce
nfs:
server: 170.106.37.33
path: "/usr/local/k8s/zookeeper/pv2"
persistentVolumeReclaimPolicy: Recycle
---
apiVersion: v1
kind: PersistentVolume
metadata:
name: k8s-pv-zk03
labels:
app: zk
annotations:
volume.beta.kubernetes.io/storage-class: "anything"
spec:
capacity:
storage: 1Gi
accessModes:
- ReadWriteOnce
nfs:
server: 170.106.37.33
path: "/usr/local/k8s/zookeeper/pv3"
persistentVolumeReclaimPolicy: Recycle
(2)然后执行如下命令创建 PV:
kubectl apply -f zookeeper-pv.yaml
(3)执行如下命令可以查看是否创建成功:
kubectl get pv
2,创建 ZooKeeper 集群
(1)我们这里要搭建一个包含 3 个节点的 ZooKeeper 集群。首先创建一个 zookeeper.yaml 文件,内容如下:
apiVersion: v1
kind: Service
metadata:
name: zk-hs
labels:
app: zk
spec:
selector:
app: zk
clusterIP: None
ports:
- name: server
port: 2888
- name: leader-election
port: 3888
---
apiVersion: v1
kind: Service
metadata:
name: zk-cs
labels:
app: zk
spec:
selector:
app: zk
type: NodePort
ports:
- name: client
port: 2181
nodePort: 31811
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: zk
spec:
serviceName: "zk-hs"
replicas: 3 # by default is 1
selector:
matchLabels:
app: zk # has to match .spec.template.metadata.labels
updateStrategy:
type: RollingUpdate
podManagementPolicy: Parallel
template:
metadata:
labels:
app: zk # has to match .spec.selector.matchLabels
spec:
containers:
- name: zk
imagePullPolicy: Always
image: leolee32/kubernetes-library:kubernetes-zookeeper1.0-3.4.10
ports:
- containerPort: 2181
name: client
- containerPort: 2888
name: server
- containerPort: 3888
name: leader-election
command:
- sh
- -c
- "start-zookeeper \
--servers=3 \
--data_dir=/var/lib/zookeeper/data \
--data_log_dir=/var/lib/zookeeper/data/log \
--conf_dir=/opt/zookeeper/conf \
--client_port=2181 \
--election_port=3888 \
--server_port=2888 \
--tick_time=2000 \
--init_limit=10 \
--sync_limit=5 \
--heap=4G \
--max_client_cnxns=60 \
--snap_retain_count=3 \
--purge_interval=12 \
--max_session_timeout=40000 \
--min_session_timeout=4000 \
--log_level=INFO"
readinessProbe:
exec:
command:
- sh
- -c
- "zookeeper-ready 2181"
initialDelaySeconds: 10
timeoutSeconds: 5
livenessProbe:
exec:
command:
- sh
- -c
- "zookeeper-ready 2181"
initialDelaySeconds: 10
timeoutSeconds: 5
volumeMounts:
- name: datadir
mountPath: /var/lib/zookeeper
volumeClaimTemplates:
- metadata:
name: datadir
annotations:
volume.beta.kubernetes.io/storage-class: "anything"
spec:
accessModes: [ "ReadWriteOnce" ]
resources:
requests:
storage: 1Gi
(2)然后执行如下命令开始创建:
kubectl apply -f zookeeper.yaml
(3)执行如下命令可以查看是否创建成功:
kubectl get pods
kubectl get service
三、创建 Kafka 集群
(1)我们这里要搭建一个包含 3 个节点的 Kafka 集群。首先创建一个 kafka.yaml 文件,内容如下:
注意:
- nfs 地址需要改成实际 NFS 服务器地址。
- status.hostIP 表示宿主机的 IP,即 Pod 实际最终部署的 Node 节点 IP(本文我是直接部署到 Master 节点上),将 KAFKA_ADVERTISED_HOST_NAME 设置为宿主机 IP 可以确保 K8s 集群外部也可以访问 Kafka
apiVersion: v1
kind: Service
metadata:
name: kafka-service-1
labels:
app: kafka-service-1
spec:
type: NodePort
ports:
- port: 9092
name: kafka-service-1
targetPort: 9092
nodePort: 30901
protocol: TCP
selector:
app: kafka-1
apiVersion: v1
kind: Service
metadata:
name: kafka-service-2
labels:
app: kafka-service-2
spec:
type: NodePort
ports:
- port: 9092
name: kafka-service-2
targetPort: 9092
nodePort: 30902
protocol: TCP
selector:
app: kafka-2
apiVersion: v1
kind: Service
metadata:
name: kafka-service-3
labels:
app: kafka-service-3
spec:
type: NodePort
ports:
- port: 9092
name: kafka-service-3
targetPort: 9092
nodePort: 30903
protocol: TCP
selector:
app: kafka-3
apiVersion: apps/v1
kind: Deployment
metadata:
name: kafka-deployment-1
spec:
replicas: 1
selector:
matchLabels:
app: kafka-1
template:
metadata:
labels:
app: kafka-1
spec:
containers:
- name: kafka-1
image: wurstmeister/kafka
imagePullPolicy: IfNotPresent
ports:
- containerPort: 9092
env:
- name: KAFKA_ZOOKEEPER_CONNECT
value: zk-0.zk-hs.default.svc.cluster.local:2181,zk-1.zk-hs.default.svc.cluster.local:2181,zk-2.zk-hs.default.svc.cluster.local:2181
- name: KAFKA_BROKER_ID
value: "1"
- name: KAFKA_CREATE_TOPICS
value: mytopic:2:1
- name: KAFKA_LISTENERS
value: PLAINTEXT://0.0.0.0:9092
- name: KAFKA_ADVERTISED_PORT
value: "30901"
- name: KAFKA_ADVERTISED_HOST_NAME
valueFrom:
fieldRef:
fieldPath: status.hostIP
volumeMounts:
- name: datadir
mountPath: /var/lib/kafka
volumes:
- name: datadir
nfs:
server: 170.106.37.33
path: "/usr/local/k8s/kafka/pv1"
apiVersion: apps/v1
kind: Deployment
metadata:
name: kafka-deployment-2
spec:
replicas: 1
selector:
matchLabels:
app: kafka-2
template:
metadata:
labels:
app: kafka-2
spec:
containers:
- name: kafka-2
image: wurstmeister/kafka
imagePullPolicy: IfNotPresent
ports:
- containerPort: 9092
env:
- name: KAFKA_ZOOKEEPER_CONNECT
value: zk-0.zk-hs.default.svc.cluster.local:2181,zk-1.zk-hs.default.svc.cluster.local:2181,zk-2.zk-hs.default.svc.cluster.local:2181
- name: KAFKA_BROKER_ID
value: "2"
- name: KAFKA_LISTENERS
value: PLAINTEXT://0.0.0.0:9092
- name: KAFKA_ADVERTISED_PORT
*value: "30902"
- name: KAFKA_ADVERTISED_HOST_NAME
valueFrom:
fieldRef:
fieldPath: status.hostIP
volumeMounts: - name: datadir
mountPath: /var/lib/kafka
volumes: - name: datadir
nfs:
server: 170.106.37.33
path: "/usr/local/k8s/kafka/pv2"
apiVersion: apps/v1
kind: Deployment
metadata:
name: kafka-deployment-3
spec:
replicas: 1
selector:
matchLabels:
app: kafka-3
template:
metadata:
labels:
app: kafka-3
spec:
containers:
- name: kafka-3
image: wurstmeister/kafka
imagePullPolicy: IfNotPresent
ports: - containerPort: 9092
env: - name: KAFKA_ZOOKEEPER_CONNECT
value: zk-0.zk-hs.default.svc.cluster.local:2181,zk-1.zk-hs.default.svc.cluster.local:2181,zk-2.zk-hs.default.svc.cluster.local:2181 - name: KAFKA_BROKER_ID
value: "3" - name: KAFKA_LISTENERS
value: PLAINTEXT://0.0.0.0:9092 - name: KAFKA_ADVERTISED_PORT
value: "30903" - name: KAFKA_ADVERTISED_HOST_NAME
valueFrom:
fieldRef:
fieldPath: status.hostIP
volumeMounts: - name: datadir
mountPath: /var/lib/kafka
volumes: - name: datadir
nfs:
server: 170.106.37.33
path: "/usr/local/k8s/kafka/pv3"*
(2)然后执行如下命令开始创建:
kubectl apply -f kafka.yaml
(3)执行如下命令可以查看是否创建成功:
kubectl get pods
kubectl get service
四、开始测试
1,K8s 集群内部测试
(1)首先执行如下命令进入一个容器:
kubectl exec -it kafka-deployment-1-59f87c7cbb-99k46 /bin/bash
(2)接着执行如下命令创建一个名为 test_topic 的 topic:
kafka-topics.sh --create --topic test_topic --zookeeper zk-0.zk-hs.default.svc.cluster.local:2181,zk-1.zk-hs.default.svc.cluster.local:2181,zk-2.zk-hs.default.svc.cluster.local:2181 --partitions 1 --replication-factor 1
(3)创建后执行如下命令开启一个生产者,启动后可以直接在控制台中输入消息来发送,控制台中的每一行数据都会被视为一条消息来发送。
kafka-console-producer.sh --broker-list kafka-service-1:9092,kafka-service-2:9092,kafka-service-3:9092 --topic test_topic
(4)重新再打开一个终端连接服务器,然后进入容器后执行如下命令开启一个消费者:
kafka-console-consumer.sh --bootstrap-server kafka-service-1:9092,kafka-service-2:9092,kafka-service-3:9092 --topic test_topic
(5)再次打开之前的消息生产客户端来发送消息,并观察消费者这边对消息的输出来体验 Kafka 对消息的基础处理。
2,集群外出测试
使用 Kafka 客户端工具(Offset Explorer)连接 Kafka 集群(可以通过 zookeeper 地址连接,也可以通过 kafka 地址连接),可以连接成功并能查看到数据。
更多的测试命令参考:
二、Kafka生产者消费者实例(基于命令行)
1.创建一个itcasttopic的主题
代码如下(示例):
kafka-topics.sh --create --topic itcasttopic --partitions 3 --replication-factor 2 -zookeeper 10.0.0.27:2181,10.0.0.103:2181,10.0.0.37:2181
2.hadoop01当生产者
代码如下(示例):
kafka-console-producer.sh --broker-list kafka-service-1:9092,kafka-service-2:9092,kafka-service-3:9092 --topic itcasttopic
3.hadoop02当消费者
代码如下(示例):
kafka-console-consumer.sh --from-beginning --topic itcasttopic --bootstrap-server kafka-service-1:9092,kafka-service-2:9092,kafka-service-3:9092
3.--list查看所有主题
代码如下(示例):
kafka-topics.sh --list --zookeeper 10.0.0.27:2181,10.0.0.103:2181,10.0.0.37:2181
4.删除主题
代码如下(示例):
kafka-topics.sh --delete --zookeeper 10.0.0.27:2181,10.0.0.103:2181,10.0.0.37:2181 --topic itcasttopic
5.关闭kafka
代码如下(示例):
bin/kafka-server-stop.sh config/server.properties