k8s部署mysql主从

搭建一个主从复制（Master-Slave）的MySQL集群

从节点可以水平扩展
所有的写操作只能在MySQL主节点上执行
读操作可以在MySQL主从节点上执行
从节点能同步主节点的数据

78313fdc86be5439ca86daeeb3a49147.png

本地存储原理

这里使用了本地存储，也就是说，用户希望Kubernetes能够直接使用宿主机上的本地磁盘目录，而不依赖于远程存储服务，来提供持久化的容器Volume
1、我们把存储固定在一个节点上，但是Pod在调度的时候，是飘来飘去的，怎么能让Pod通过PVC也能固定在PVC上？
2、给这个Pod加上一个nodeAffinity行不行？

当然行，但是这变相破坏了开发人员对资源对象的定义规范了，开发人员应该不需要时刻考虑调度的细节。调度的改动应该交给运维就行。所以我们为了实现本地存储，我们采用了延迟绑定的方法。方法很简单，我们都知道StorageClass一般由运维人员设计，我们只需要在StorageClass指定no-provisioner。这是因为Local Persistent Volume目前尚不支持Dynamic Provisioning，所以它没办法在用户创建PVC的时候，就自动创建出对应的PV。与此同时，这个StorageClass还定义了一个volumeBindingMode=WaitForFirstConsumer的属性。它是Local Persistent Volume里一个非常重要的特性，即：延迟绑定。

创建PV

先在Node （实验用的Node节点IP是node1）节点上，预先分配几个PV（不建议在生产上这样操作），要保证在node1上有对应的目录

ssh root@node1
mkdir -pv /data/svr/projects/{mysql,mysql2,mysql3}

vim 01-persistentVolume-1.yaml

apiVersion: v1
kind: PersistentVolume
metadata:
  name: example-mysql-pv
spec:
  capacity:
    storage: 15Gi
  volumeMode: Filesystem
  accessModes:
  - ReadWriteOnce
  persistentVolumeReclaimPolicy: Delete
  storageClassName: local-storage
  local:
    path: /data/svr/projects/mysql
  nodeAffinity:
    required:
      nodeSelectorTerms:
      - matchExpressions:
        - key: kubernetes.io/hostname
          operator: In
          values:
          - node1

vim 01-persistentVolume-2.yaml

apiVersion: v1
kind: PersistentVolume
metadata:
  name: example-mysql-pv-2
spec:
  capacity:
    storage: 15Gi
  volumeMode: Filesystem
  accessModes:
  - ReadWriteOnce
  persistentVolumeReclaimPolicy: Delete
  storageClassName: local-storage
  local:
    path: /data/svr/projects/mysql2
  nodeAffinity:
    required:
      nodeSelectorTerms:
      - matchExpressions:
        - key: kubernetes.io/hostname
          operator: In
          values:
          - node1

vim 01-persistentVolume-3.yaml

apiVersion: v1
kind: PersistentVolume
metadata:
  name: example-mysql-pv-3
spec:
  capacity:
    storage: 15Gi
  volumeMode: Filesystem
  accessModes:
  - ReadWriteOnce
  persistentVolumeReclaimPolicy: Delete
  storageClassName: local-storage
  local:
    path: /data/svr/projects/mysql3
  nodeAffinity:
    required:
      nodeSelectorTerms:
      - matchExpressions:
        - key: kubernetes.io/hostname
          operator: In
          values:
          - node1

记住，这是在生产上不推荐的做法，我只是实验用途才这样手动预先创建，**正规的做法应该通过StorageClass采用Dynamic Provisioning， 而不是Static Provisioning机制生产PV

[root@master mysql]# kubectl apply -f 01-persistentVolume-1.yaml 
persistentvolume/example-mysql-pv created
[root@master mysql]# kubectl apply -f 01-persistentVolume-2.yaml 
persistentvolume/example-mysql-pv-2 created
[root@master mysql]# kubectl apply -f 01-persistentVolume-3.yaml 
persistentvolume/example-mysql-pv-3 created
[root@master mysql]# kubectl get pv
NAME                 CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS      CLAIM   STORAGECLASS    REASON   AGE
example-mysql-pv     15Gi       RWO            Delete           Available           local-storage            8s
example-mysql-pv-2   15Gi       RWO            Delete           Available           local-storage            6s
example-mysql-pv-3   15Gi       RWO            Delete           Available           local-storage            3s

02-storageclass.yaml

kind: StorageClass
apiVersion: storage.k8s.io/v1
metadata:
  name: local-storage
provisioner: kubernetes.io/no-provisioner
volumeBindingMode: WaitForFirstConsumer

[root@master mysql]# kubectl apply -f 02-storageclass.yaml 
storageclass.storage.k8s.io/local-storage created
[root@master mysql]# kubectl get sc
NAME            PROVISIONER                    RECLAIMPOLICY   VOLUMEBINDINGMODE      ALLOWVOLUMEEXPANSION   AGE
local-storage   kubernetes.io/no-provisioner   Delete          WaitForFirstConsumer   false                  5s

创建Namespace

03-mysql-namespace.yaml

apiVersion: v1
kind: Namespace
metadata:
  name: mysql
  labels:
    app: mysql

[root@master mysql]# kubectl apply -f 03-mysql-namespace.yaml 
namespace/mysql unchanged
[root@master mysql]# kubectl get ns -n mysql
NAME              STATUS   AGE
mysql             Active   3d

创建数据库的配置文件configmap

使用ConfigMap为Master/Slave节点分配不同的配置文件

vim 04-mysql-configmap.yaml

apiVersion: v1
kind: ConfigMap
metadata:
  name: mysql
  namespace: mysql
  labels:
    app: mysql
data:
  master.cnf: |
    # Master配置
    [mysqld]
    log-bin=mysqllog
    skip-name-resolve
  slave.cnf: |
    # Slave配置
    [mysqld]
    super-read-only
    skip-name-resolve
    log-bin=mysql-bin
    replicate-ignore-db=mysql

[root@master mysql]# kubectl apply -f 04-mysql-configmap.yaml 
configmap/mysql created
[root@master mysql]# kubectl get cm -n mysql
NAME               DATA   AGE
mysql              2      5s

创建MySQL密码Secret

05-mysql-secret.yaml

apiVersion: v1
kind: Secret
metadata:
  name: mysql-secret
  namespace: mysql
  labels:
    app: mysql
type: Opaque
data:
  password: MTIzNDU2 # echo -n "123456" | base64

[root@master mysql]# kubectl apply -f 05-mysql-secret.yaml 
secret/mysql-secret created
[root@master mysql]# kubectl get secret -n mysql
NAME                  TYPE                                  DATA   AGE
mysql-secret          Opaque                                1      18s

使用Service为MySQL提供读写分离

06-mysql-services.yaml

apiVersion: v1
kind: Service
metadata:
  name: mysql
  namespace: mysql
  labels:
    app: mysql
spec:
  ports:
  - name: mysql
    port: 3306
  clusterIP: None
  selector:
    app: mysql
---
apiVersion: v1
kind: Service
metadata:
  name: mysql-read
  namespace: mysql
  labels:
    app: mysql
spec:
  ports:
  - name: mysql
    port: 3306
  selector:
    app: mysql

用户所有写请求，必须以DNS记录的方式直接访问到Master节点，也就是mysql-0.mysql这条DNS记录。

用户所有读请求，必须访问自动分配的DNS记录可以被转发到任意一个Master或Slave节点上，也就是mysql-read这条DNS记录。

[root@master mysql]# kubectl apply -f 06-mysql-services.yaml 
service/mysql created
service/mysql-read created
[root@master mysql]# kubectl get svc -n mysql
NAME         TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)    AGE
mysql        ClusterIP   None                   3306/TCP   3s
mysql-read   ClusterIP   10.107.90.19           3306/TCP   3s

创建MySQL集群实例

使用StatefulSet搭建MySQL主从集群

07-mysql-statefulset.yaml

apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: mysql
  namespace: mysql
  labels:
    app: mysql
spec:
  selector:
    matchLabels:
      app: mysql
  serviceName: mysql
  replicas: 2
  template:
    metadata:
      labels:
        app: mysql
    spec:
      initContainers:
      - name: init-mysql
        image: mysql:5.7
        env:
        - name: MYSQL_ROOT_PASSWORD
          valueFrom:
            secretKeyRef:
              name: mysql-secret
              key: password
        command:
        - bash
        - "-c"
        - |
          set -ex
          # 从Pod的序号，生成server-id
          [[ $(hostname) =~ -([0-9]+)$ ]] || exit 1
          ordinal=${BASH_REMATCH[1]}
          echo [mysqld] > /mnt/conf.d/server-id.cnf
          # 由于server-id不能为0，因此给ID加100来避开它
          echo server-id=$((100 + $ordinal)) >> /mnt/conf.d/server-id.cnf
          # 如果Pod的序号为0，说明它是Master节点，从ConfigMap里把Master的配置文件拷贝到/mnt/conf.d目录下
          # 否则，拷贝ConfigMap里的Slave的配置文件
          if [[ ${ordinal} -eq 0 ]]; then
            cp /mnt/config-map/master.cnf /mnt/conf.d
          else
            cp /mnt/config-map/slave.cnf /mnt/conf.d
          fi
        volumeMounts:
        - name: conf
          mountPath: /mnt/conf.d
        - name: config-map
          mountPath: /mnt/config-map
      - name: clone-mysql
        image: gcr.io/google-samples/xtrabackup:1.0
        env:
        - name: MYSQL_ROOT_PASSWORD
          valueFrom:
            secretKeyRef:
              name: mysql-secret
              key: password
        command:
        - bash
        - "-c"
        - |
          set -ex
          # 拷贝操作只需要在第一次启动时进行，所以数据已经存在则跳过
          [[ -d /var/lib/mysql/mysql ]] && exit 0
          # Master 节点（序号为 0）不需要这个操作
          [[ $(hostname) =~ -([0-9]+)$ ]] || exit 1
          ordinal=${BASH_REMATCH[1]}
          [[ $ordinal == 0 ]] && exit 0
          # 使用ncat指令，远程地从前一个节点拷贝数据到本地
          ncat --recv-only mysql-$(($ordinal-1)).mysql 3307 | xbstream -x -C /var/lib/mysql
          # 执行 --prepare，这样拷贝来的数据就可以用作恢复了
          xtrabackup --prepare --target-dir=/var/lib/mysql
        volumeMounts:
        - name: data
          mountPath: /var/lib/mysql
          subPath: mysql
        - name: conf
          mountPath: /etc/mysql/conf.d
      containers:
      - name: mysql
        image: mysql:5.7
        env:
 #        - name: MYSQL_ALLOW_EMPTY_PASSWORD
 #          value: "1"
        - name: MYSQL_ROOT_PASSWORD
          valueFrom:
            secretKeyRef:
              name: mysql-secret
              key: password
        ports:
        - name: mysql
          containerPort: 3306
        volumeMounts:
        - name: data
          mountPath: /var/lib/mysql
          subPath: mysql
        - name: conf
          mountPath: /etc/mysql/conf.d
        resources:
          requests:
            cpu: 500m
            memory: 1Gi
        livenessProbe:
          exec:
            command: ["mysqladmin", "ping", "-uroot", "-p${MYSQL_ROOT_PASSWORD}"]
          initialDelaySeconds: 30
          periodSeconds: 10
          timeoutSeconds: 5
        readinessProbe:
          exec:
            command: ["mysqladmin", "ping", "-uroot", "-p${MYSQL_ROOT_PASSWORD}"]
          initialDelaySeconds: 5
          periodSeconds: 2
          timeoutSeconds: 1
      - name: xtrabackup
        image: gcr.io/google-samples/xtrabackup:1.0
        ports:
        - name: xtrabackup
          containerPort: 3307
        env:
        - name: MYSQL_ROOT_PASSWORD
          valueFrom:
            secretKeyRef:
              name: mysql-secret
              key: password
        command:
        - bash
        - "-c"
        - |
          set -ex
          cd /var/lib/mysql
          # 从备份信息文件里读取MASTER_LOG_FILE和MASTER_LOG_POS这2个字段的值，用来拼装集群初始化SQL
          if [[ -f xtrabackup_slave_info ]]; then
            # 如果xtrabackup_slave_info文件存在，说明这个备份数据来自于另一个Slave节点
            # 这种情况下，XtraBackup工具在备份的时候，就已经在这个文件里自动生成了“CHANGE MASTER TO”SQL语句
            # 所以，只需要把这个文件重命名为change_master_to.sql.in，后面直接使用即可
            mv xtrabackup_slave_info change_master_to.sql.in
            # 所以，也就用不着xtrabackup_binlog_info了
            rm -f xtrabackup_binlog_info
          elif [[ -f xtrabackup_binlog_info ]]; then
            # 如果只是存在xtrabackup_binlog_info文件，说明备份来自于Master节点，就需要解析这个备份信息文件，读取所需的两个字段的值
            [[ $(cat xtrabackup_binlog_info) =~ ^(.*?)[[:space:]]+(.*?)$ ]] || exit 1
            rm xtrabackup_binlog_info
            # 把两个字段的值拼装成SQL，写入change_master_to.sql.in文件
            echo "CHANGE MASTER TO MASTER_LOG_FILE='${BASH_REMATCH[1]}',\
                  MASTER_LOG_POS=${BASH_REMATCH[2]}" > change_master_to.sql.in
          fi
          # 如果存在change_master_to.sql.in，就意味着需要做集群初始化工作
          if [[ -f change_master_to.sql.in ]]; then
            # 但一定要先等MySQL容器启动之后才能进行下一步连接MySQL的操作
            echo "Waiting for mysqld to be ready（accepting connections）"
            until mysql -h 127.0.0.1 -uroot -p${MYSQL_ROOT_PASSWORD} -e "SELECT 1"; do sleep 1; done
            echo "Initializing replication from clone position"
            # 将文件change_master_to.sql.in改个名字
            # 防止这个Container重启的时候，因为又找到了change_master_to.sql.in，从而重复执行一遍初始化流程
            mv change_master_to.sql.in change_master_to.sql.orig
            # 使用change_master_to.sql.orig的内容，也就是前面拼装的SQL，组成一个完整的初始化和启动Slave的SQL语句
            mysql -h 127.0.0.1 -uroot -p${MYSQL_ROOT_PASSWORD} << EOF
          $(< change_master_to.sql.orig),
            MASTER_HOST='mysql-0.mysql.mysql',
            MASTER_USER='root',
            MASTER_PASSWORD='${MYSQL_ROOT_PASSWORD}',
            MASTER_CONNECT_RETRY=10;
          START SLAVE;
          EOF
          fi
          # 使用ncat监听3307端口。
          # 它的作用是，在收到传输请求的时候，直接执行xtrabackup --backup命令，备份MySQL的数据并发送给请求者
          exec ncat --listen --keep-open --send-only --max-conns=1 3307 -c \
            "xtrabackup --backup --slave-info --stream=xbstream --host=127.0.0.1 --user=root --password=${MYSQL_ROOT_PASSWORD}"
        volumeMounts:
        - name: data
          mountPath: /var/lib/mysql
          subPath: mysql
        - name: conf
          mountPath: /etc/mysql/conf.d
      volumes:
      - name: conf
        emptyDir: {}
      - name: config-map
        configMap:
          name: mysql
  volumeClaimTemplates:
  - metadata:
      name: data
    spec:
      accessModes:
      - "ReadWriteOnce"
      storageClassName: local-storage
      resources:
        requests:
          storage: 3Gi

整体的StatefulSet有两个Replicas，一个Master，一个Slave，然后使用init-mysql这个initContainers进行配置文件的初始化。接着使用clone-mysql这个initContainers进行数据的传输；同时使用xtrabackup这个sidecar容器进行SQL初始化和数据传输功能。
创建 StatefulSet

kubectl apply -f 07-mysql-statefulset.yaml

可以看到，StatefulSet启动成功后，会有两个Pod运行。接下来，我们可以尝试向这个MySQL集群发起请求，执行一些SQL操作来验证它是否正常。整个过程因为拉取mysql和一个gcr.io/google-samples/xtrabackup:1.0国外的镜像会很慢,但是在创建mysql-0拉取一次之后，后续创建m>ysql-1就相对很快了。
最后，容器检查pod的运行状态

[root@master mysql]# kubectl get po -n mysql
NAME      READY   STATUS    RESTARTS   AGE
mysql-0   2/2     Running   0          2m7s
mysql-1   2/2     Running   0          116s

服务验证

验证主从状态

[root@master mysql]# kubectl -n mysql exec mysql-1 -c mysql -- bash -c "mysql -uroot -p123456 -e 'show slave status \G'"
*************************** 1. row ***************************
               Slave_IO_State: Waiting for master to send event
                  Master_Host: mysql-0.mysql.mysql
                  Master_User: root
                  Master_Port: 3306
                Connect_Retry: 10
              Master_Log_File: mysqllog.000003
          Read_Master_Log_Pos: 154
               Relay_Log_File: mysql-1-relay-bin.000002
                Relay_Log_Pos: 319
        Relay_Master_Log_File: mysqllog.000003
             Slave_IO_Running: Yes
            Slave_SQL_Running: Yes
              Replicate_Do_DB: 
          Replicate_Ignore_DB: mysql0
           Replicate_Do_Table: 
       Replicate_Ignore_Table: 
      Replicate_Wild_Do_Table: 
  Replicate_Wild_Ignore_Table: 
                   Last_Errno: 0
                   Last_Error: 
                 Skip_Counter: 0
          Exec_Master_Log_Pos: 154
              Relay_Log_Space: 528
              Until_Condition: None
               Until_Log_File: 
                Until_Log_Pos: 0
           Master_SSL_Allowed: No
           Master_SSL_CA_File: 
           Master_SSL_CA_Path: 
              Master_SSL_Cert: 
            Master_SSL_Cipher: 
               Master_SSL_Key: 
        Seconds_Behind_Master: 0
Master_SSL_Verify_Server_Cert: No
                Last_IO_Errno: 0
                Last_IO_Error: 
               Last_SQL_Errno: 0
               Last_SQL_Error: 
  Replicate_Ignore_Server_Ids: 
             Master_Server_Id: 100
                  Master_UUID: f8c7768a-6702-11ec-a648-c204e42fb62e
             Master_Info_File: /var/lib/mysql/master.info
                    SQL_Delay: 0
          SQL_Remaining_Delay: NULL
      Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates
           Master_Retry_Count: 86400
                  Master_Bind: 
      Last_IO_Error_Timestamp: 
     Last_SQL_Error_Timestamp: 
               Master_SSL_Crl: 
           Master_SSL_Crlpath: 
           Retrieved_Gtid_Set: 
            Executed_Gtid_Set: 
                Auto_Position: 0
         Replicate_Rewrite_DB: 
                 Channel_Name: 
           Master_TLS_Version: 
mysql: [Warning] Using a password on the command line interface can be insecure.

接下来，我们通过Master容器创建数据库和表、插入数据库

[root@master mysql]# kubectl -n mysql exec mysql-0 -c mysql -- bash -c "mysql -uroot -p123456 -e 'create database test'"

[root@master mysql]# kubectl -n mysql exec mysql-0 -c mysql -- bash -c "mysql -uroot -p123456 -e 'use test;create table counter(c int);'"

[root@master mysql]# kubectl -n mysql exec mysql-0 -c mysql -- bash -c "mysql -uroot -p123456 -e 'use test;insert into counter values(123)'"

[root@master mysql]# kubectl -n mysql exec mysql-1 -c mysql -- bash -c "mysql -uroot -p123456 -e 'use test;select * from counter'"  
c
123

当看到输出结果，主从同步正常了

扩展节点

[root@master mysql]# kubectl -n mysql scale statefulset mysql -—replicas=3
[root@master mysql]# kubectl get po -n mysql
NAME      READY   STATUS    RESTARTS   AGE
mysql-0   2/2     Running   0          11m
mysql-1   2/2     Running   0          11m
mysql-2   2/2     Running   0          20s

查看扩容节点数据是否已同步

kubectl -n mysql exec mysql-2 -c mysql -- bash -c "mysql -uroot -p123456 -e 'use test;select * from counter’"  
c
123