kubernetes中部署mysql集群并持久化存储

| 分类 linux  | 标签 docker  harbor  kubernetes  k8s  mysql集群 

简介：

参考：https://github.com/Yolean/kubernetes-mysql-cluster

环境

主节点：172.19.2.50
从节点：
172.19.2.51
172.19.2.140

部署完成后通过各节点的30336端口访问mysql

账号root，密码abcd1234
如：
mysql -h 172.19.2.50 -P 30336 -uroot -pabcd1234

部署完成后通过galera可以让集群3个节点间的数据一致

容器内访问mysql时，可以通过所有k8s节点的30336端口访问，也可以使用k8s服务中的内部入口访问，如mysql.mysql:3306，dns会自动解析mysql.mysql到对应的服务集群

一、在主节点创建目录

mkdir -pv /mysql_data/datadir-mariadb-0
mkdir -pv /mysql_data/datadir-mariadb-1
mkdir -pv /mysql_data/datadir-mariadb-2

二、修改部署文件

cd /opt/kubernetes-mysql-cluster

命名空间部署文件

vim 00namespace.yml
---
apiVersion: v1
kind: Namespace
metadata:
  name: mysql

pvc部署文件

vim 10pvc.yml
---
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: mysql-mariadb-0
  namespace: mysql
spec:
  accessModes:
	- ReadWriteOnce		#这里为pvc的访问模式
  resources:
	requests:
	  storage: 10Gi		#这里调整要挂载的pvc大小
  selector:
	matchLabels:		#这里要和pv的标签对应
	  app: mariadb
	  podindex: "0"
---
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: mysql-mariadb-1
  namespace: mysql
spec:
  accessModes:
	- ReadWriteOnce
  resources:
	requests:
	  storage: 10Gi
  selector:
	matchLabels:
	  app: mariadb
	  podindex: "1"
---
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: mysql-mariadb-2
  namespace: mysql
spec:
  accessModes:
	- ReadWriteOnce
  resources:
	requests:
	  storage: 10Gi
  selector:
	matchLabels:
	  app: mariadb
	  podindex: "2"

mariadb服务文件

vim 20mariadb-service.yml
# the "Headless Service, used to control the network domain"
---
apiVersion: v1
kind: Service
metadata:
  name: mariadb
  namespace: mysql
spec:
  clusterIP: None
  selector:
	app: mariadb
  ports:
	- port: 3306
	  name: mysql
	- port: 4444
	  name: sst
	- port: 4567
	  name: replication
	- protocol: UDP
	  port: 4567
	  name: replicationudp
	- port: 4568
	  name: ist

mysql服务文件

vim 30mysql-service.yml
---
apiVersion: v1
kind: Service
metadata:
  name: mysql
  namespace: mysql
spec:
  ports:
  - port: 3306
	name: mysql
	targetPort: 3306
	nodePort: 30336		#这里为porxy映射端口
  selector:
	app: mariadb
  type: NodePort

载入配置文件

vim 40configmap.sh
#!/bin/bash
DIR=`dirname "$BASH_SOURCE"`
kubectl create configmap "conf-d" --from-file="$DIR/conf-d/" --namespace=mysql

输入mysql初始密码文件

vim 41secret.sh
#!/bin/bash
echo -n Please enter mysql root password for upload to k8s secret:
read -s rootpw
echo
kubectl create secret generic mysql-secret --namespace=mysql --from-literal=rootpw=$rootpw

部署有状态集群文件

vim 50mariadb.yml
apiVersion: apps/v1beta1
kind: StatefulSet
metadata:
  name: mariadb
  namespace: mysql
spec:
  serviceName: "mariadb"
  replicas: 1		#这里是要启动的节点的数量
  template:
	metadata:
	  labels:
		app: mariadb
	spec:
	  terminationGracePeriodSeconds: 10
	  containers:
		- name: mariadb
		  image: mariadb:10.1.22	#这里修改使用的镜像文件
		  ports:
			- containerPort: 3306
			  name: mysql
			- containerPort: 4444
			  name: sst
			- containerPort: 4567
			  name: replication
			- containerPort: 4567
			  protocol: UDP
			  name: replicationudp
			- containerPort: 4568
			  name: ist
		  env:
			- name: MYSQL_ROOT_PASSWORD
			  valueFrom:
				secretKeyRef:
				  name: mysql-secret
				  key: rootpw
			- name: MYSQL_INITDB_SKIP_TZINFO
			  value: "yes"
		  args:
			- --character-set-server=utf8mb4
			- --collation-server=utf8mb4_unicode_ci
			# Remove after first replicas=1 create
			- --wsrep-new-cluster	#这里在执行的时候代表会创建新集群，新增节点的时候要注释掉
		  volumeMounts:
			- name: mysql
			  mountPath: /var/lib/mysql
			- name: conf
			  mountPath: /etc/mysql/conf.d
			- name: initdb
			  mountPath: /docker-entrypoint-initdb.d
	  volumes:
		- name: conf
		  configMap:
			name: conf-d
		- name: initdb
		  emptyDir: {}
  volumeClaimTemplates:
  - metadata:
	  name: mysql
	spec:
	  accessModes: [ "ReadWriteOnce" ]
	  resources:
		requests:
		  storage: 10Gi

调整集群节点到3个的文件

vim 70unbootstrap.sh
#!/bin/bash
DIR=`dirname "$BASH_SOURCE"`
set -e
set -x
cp "$DIR/50mariadb.yml" "$DIR/50mariadb.yml.unbootstrap.yml"
sed -i 's/replicas: 1/replicas: 3/' "$DIR/50mariadb.yml.unbootstrap.yml"
sed -i 's/- --wsrep-new-cluster/#- --wsrep-new-cluster/' "$DIR/50mariadb.yml.unbootstrap.yml"
kubectl apply -f "$DIR/50mariadb.yml.unbootstrap.yml"
rm "$DIR/50mariadb.yml.unbootstrap.yml"

创建目录

mkdir -pv bootstrap conf-d

创建pv脚本

vim bootstrap/pv.sh
#!/bin/bash
echo "Note that in for example GKE a PetSet will have PersistentVolume(s) and PersistentVolumeClaim(s) created for it automatically"
dir="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && cd .. && pwd )"
path="$dir/data"
echo "Please enter a path where to store data during local testing: ($path)"
read newpath
[ -n "$newpath" ] && path=$newpath
cat bootstrap/pv-template.yml | sed "s|/tmp/k8s-data|$path|" | kubectl create -f -

创建挂载pv文件

vim bootstrap/pv-template.yml
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: datadir-mariadb-0
  labels:				#这里的标签要和pvc的matchLabels对应
	app: mariadb
	podindex: "0"
spec:
  accessModes:
  - ReadWriteOnce		#这里是pv的访问模式，必须要与pvc相同
  capacity:
	storage: 10Gi		#这里是要创建的pv的大小
  hostPath:
	path: /mysql_data/datadir-mariadb-0		#这里为挂载到本地的路径
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: datadir-mariadb-1
  labels:
	app: mariadb
	podindex: "1"
spec:
  accessModes:
  - ReadWriteOnce
  capacity:
	storage: 10Gi
  hostPath:
	path: /mysql_data/datadir-mariadb-1
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: datadir-mariadb-2
  labels:
	app: mariadb
	podindex: "2"
spec:
  accessModes:
  - ReadWriteOnce
  capacity:
	storage: 10Gi
  hostPath:
	path: /mysql_data/datadir-mariadb-2

删除pv脚本

vim bootstrap/rm.sh
#!/bin/bash
echo "Note that in for example GKE a PetSet will have PersistentVolume(s) and PersistentVolumeClaim(s) created for it automatically"
dir="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && cd .. && pwd )"
path="$dir/data"
echo "Please enter a path where to store data during local testing: ($path)"
read newpath
[ -n "$newpath" ] && path=$newpath
cat bootstrap/pv-template.yml | sed "s|/tmp/k8s-data|$path|" | kubectl delete -f -

集群同步galera的配置文件

vim conf-d/galera.cnf
[server]
[mysqld]
[galera]
wsrep_on=ON
wsrep_provider="/usr/lib/galera/libgalera_smm.so"
wsrep_cluster_address="gcomm://mariadb-0.mariadb,mariadb-1.mariadb,mariadb-2.mariadb"
binlog_format=ROW
default_storage_engine=InnoDB
innodb_autoinc_lock_mode=2
wsrep-sst-method=rsync
bind-address=0.0.0.0
[embedded]
[mariadb]
[mariadb-10.1]

三、在kubernetes上部署mariadb集群

kubernetes所有节点执行

docker pull mariadb:10.1.22

主节点执行

cd /opt/kubernetes-mysql-cluster
sh bootstrap/pv.sh
kubectl create -f 00namespace.yml
kubectl create -f 10pvc.yml
./40configmap.sh
./41secret.sh
设置数据库root密码为abcd1234
kubectl create -f 20mariadb-service.yml
kubectl create -f 30mysql-service.yml
kubectl create -f 50mariadb.yml

四、增加集群节点至3个

./70unbootstrap.sh

五、清理kubernetes上的mysql集群

主节点上执行

cd /opt/kubernetes-mysql-cluster
kubectl delete -f ./
sh bootstrap/rm.sh

所有节点执行

rm -rf /mysql_data/datadir-mariadb-0/*
rm -rf /mysql_data/datadir-mariadb-1/*
rm -rf /mysql_data/datadir-mariadb-2/*

上一篇     下一篇