Docker部署Always on集群

SQL Server在2016年开始支持Linux。随着2017和2019版本的发布,它开始支持Linux和容器平台上的HA/DR、Kubernetes和大数据集群解决方案。

在本文中,我们将在3个节点的Docker容器上安装SQL Server 2019,并创建AlwaysOn可用性组。

我们的目标是使用单个配置文件快速准备好环境。因此,开发人员或测试团队可以快速执行诸如兼容性、连通性、代码功能等测试。

在本节中,我们将首先准备一个基于Ubuntu的映像,以便能够在容器上安装可用性组。然后我们将执行必要的安装。

重要提示:不建议在生产环境中执行操作。安装是在Ubuntu 18.04上执行的。

安装Docker

安装Docker就不介绍了,自行安装即可.

架构

主机名IP端口角色
sqlNode1宿主机IP1501:1433
sqlNode2宿主机IP1502:1433副本
sqlNode3宿主机IP1503:1433副本

端口表示:外网端口:内网端口

准备相关容器镜像

拉取操作系统和数据库的Docker镜像,如下

操作系统

docker pull ubuntu:18.04

SQL Server 2019

docker pull mcr.microsoft.com/mssql/server:2019-latest

可通过docker images来查看已下载的镜像信息。

镜像地址:https://hub.docker.com/_/microsoft-mssql-server

开始配置-容器

环境准备完毕后,开始正式的配置安装。

步骤1:创建Dockerfile

创建目录用于存放dockerfile、docker-compose.yml等文件。

mkdir /sql2019ha
cd /sql2019ha
touch dockerfile
vi dockerfile
  • dockerfile内容如下

    FROM ubuntu:18.04
     
    RUN apt-get update
     
    RUN apt install sudo wget curl gnupg gnupg1 gnupg2 -y
    RUN apt install software-properties-common systemd vim -y
    RUN wget -qO- https://packages.microsoft.com/keys/microsoft.asc | sudo apt-key add -
     
    RUN add-apt-repository "$(wget -qO- https://packages.microsoft.com/config/ubuntu/18.04/mssql-server-2019.list)"
    RUN apt-get update
    RUN apt-get install -y mssql-server
     
    RUN /opt/mssql/bin/mssql-conf set hadr.hadrenabled  1
    RUN /opt/mssql/bin/mssql-conf set sqlagent.enabled true
     
    EXPOSE 1433
     
    ENTRYPOINT /opt/mssql/bin/sqlservr
    

    说明

    FROM:表示基于什么镜像进行安装的

    RUN:在镜像中进行的操作

    EXPOSE:指定服务端口

    ENTRYPOINT:运行命令

步骤2:编译镜像

通过dockerfile来编译镜像,用于后面的安装,命令:docker build -t sqlag2019:ha .

其中sqlag2019为镜像名称,ha是镜像标签,.表示在当前目录下编译,因为dockerfile就在当前目录下。

以下输出是精简的,实际上输出非常多…也需要一定时间(安装一些包、数据库等),由网速决定

$ docker build -t sqlag2019:ha .
Sending build context to Docker daemon   2.56kB
Step 1/12 : FROM ubuntu:18.04
 ---> c3c304cb4f22
Step 2/12 : RUN apt-get update
 ---> Running in 950e50f80f00
Get:1 http://archive.ubuntu.com/ubuntu bionic InRelease [242 kB]
Get:2 http://security.ubuntu.com/ubuntu bionic-security InRelease [88.7 kB]
Get:3 http://security.ubuntu.com/ubuntu bionic-security/main amd64 Packages [932 kB]
Get:4 http://archive.ubuntu.com/ubuntu bionic-updates InRelease [88.7 kB]...
Step 3/12 : RUN apt install sudo wget curl gnupg gnupg1 gnupg2 -y
 ---> Running in edc9d15b2383
..
..
Step 8/12 : RUN sudo apt-get install -y mssql-server
 ---> Running in 43d82a503f8a
Reading package lists...
Building dependency tree...
Reading state information...
The following additional packages will be installed:
Step 9/12 : RUN sudo /opt/mssql/bin/mssql-conf set hadr.hadrenabled  1
 ---> Running in 166c6596d2dd
SQL Server needs to be restarted in order to apply this setting. Please run
'systemctl restart mssql-server.service'.
Removing intermediate container 166c6596d2dd
 ---> bcdb057fed43
Step 10/12 : RUN sudo /opt/mssql/bin/mssql-conf set sqlagent.enabled true
 ---> Running in 22dd6a93d1ef
SQL Server needs to be restarted in order to apply this setting. Please run
'systemctl restart mssql-server.service'.
Removing intermediate container 22dd6a93d1ef
 ---> 6b90afbaf94e
Step 11/12 : EXPOSE 1433
 ---> Running in bcc14f3b0bad
Removing intermediate container bcc14f3b0bad
 ---> 4aae1563aa74
Step 12/12 : ENTRYPOINT /opt/mssql/bin/sqlservr
 ---> Running in 68b6ed45ff6a
Removing intermediate container 68b6ed45ff6a
 ---> b7467618c371
Successfully built b7467618c371
Successfully tagged sqlag2019:ha

最后出现Successfully表示编译成功,否则根据错误信息进行解决。

步骤3:创建容器

现在镜像编译好了,下面再通过docker-compose文件来创建、配置3个容器,具体内容如下:

$ touch docker-compose.yml
$ vi docker-compose.yml
version: '3'
 
services:
  db1:
    container_name: sqlNode1
    image: sqlag2019:ha
    hostname: sqlNode1
    domainname: lab.local
    environment:
      SA_PASSWORD: "MyPassWord123"
      ACCEPT_EULA: "Y"
    ports:
    - "1501:1433"
    extra_hosts:
      sqlNode2.labl.local: "172.16.238.22"
      sqlNode3.labl.local: "172.16.238.23"
    networks:
        internal:
                ipv4_address: 172.16.238.21
 
  db2:
    container_name: sqlNode2
    image: sqlag2019:ha
    hostname: sqlNode2
    domainname: lab.local
    environment:
      SA_PASSWORD: "MyPassWord123"
      ACCEPT_EULA: "Y"
    ports:
    - "1502:1433"
    extra_hosts:
      sqlNode1.lab.local: "172.16.238.21"
      sqlNode3.lab.local: "172.16.238.23"
    networks:
        internal:
                ipv4_address: 172.16.238.22
 
  db3:
    container_name: sqlNode3
    image: sqlag2019:ha
    hostname: sqlNode3
    domainname: lab.local
    environment:
      SA_PASSWORD: "MyPassWord123"
      ACCEPT_EULA: "Y"
    ports:
    - "1503:1433"
    extra_hosts:
      sqlNode1.lab.local: "172.16.238.21"
      sqlNode2.lab.local: "172.16.238.22"
    networks:
        internal:
                ipv4_address: 172.16.238.23
 
networks:
    internal:
      ipam:
            driver: default
            config:
                - subnet: 172.16.238.0/24

步骤4:启动容器

然后通过docker-compose up -d命令启动三个容器,其中-d表示在后台运行。

$ docker-compose up -d
Creating network "sql2019hademo_internal" with the default driver
Creating sqlNode2 ...
Creating sqlNode1 ...
Creating sqlNode2
Creating sqlNode3 ...
Creating sqlNode1
Creating sqlNode2 ... done

注意:docker-compose是需要单独安装的,就是一个可执行文件。可通过apt、yum来安装。

查看容器状态

$ docker-compose ps
Name                Command               State           Ports
--------------------------------------------------------------------------
sqlNode1   /bin/sh -c /opt/mssql/bin/ ...   Up      0.0.0.0:1501->1433/tcp
sqlNode2   /bin/sh -c /opt/mssql/bin/ ...   Up      0.0.0.0:1502->1433/tcp
sqlNode3   /bin/sh -c /opt/mssql/bin/ ...   Up      0.0.0.0:1503->1433/tcp

至此容器已经启动完成,下面通过SSMS连接数据库进行相关检查和配置ALWAYSON。

步骤5:SSMS连接MSSQL

通过宿主机的外网IP+端口连接相应的数据库,如下:

在这里插入图片描述

注意:IP和端口之间是逗号

在这里插入图片描述
可以看到数据库的图标也是Linux的图标。

配置-数据库

这部分就是在数据库中进行相关配置,如:创建KEY加密文件,管理用户、可用组等。

步骤1:连接主库-sqlNode1

主库也就是节点1,端口是1501,连接方法如上图。

我们将证书和私钥提取到/tmp/dbm_certificate.cer/tmp/dbm_certificate.pvk文件中。

我们将这些文件复制到其他节点,并根据以下文件创建主密钥和证书:执行以下脚本

USE master
GO
 
CREATE LOGIN dbm_login WITH PASSWORD = 'MyStr0ngPa$w0rd';
CREATE USER dbm_user FOR LOGIN dbm_login;
GO
 
CREATE MASTER KEY ENCRYPTION BY PASSWORD = 'MyStr0ngPa$w0rd';
go
CREATE CERTIFICATE dbm_certificate WITH SUBJECT = 'dbm';
BACKUP CERTIFICATE dbm_certificate
TO FILE = '/tmp/dbm_certificate.cer'
WITH PRIVATE KEY (
        FILE = '/tmp/dbm_certificate.pvk',
        ENCRYPTION BY PASSWORD = 'MyStr0ngPa$w0rd'
    );
GO

将文件拷贝到其他两个节点:

$ docker cp sqlNode1:/tmp/dbm_certificate.cer .
$ docker cp sqlNode1:/tmp/dbm_certificate.pvk .
$ docker cp dbm_certificate.cer sqlNode2:/tmp/
$ docker cp dbm_certificate.pvk sqlNode2:/tmp/
$ docker cp dbm_certificate.cer sqlNode3:/tmp/
$ docker cp dbm_certificate.pvk sqlNode3:/tmp/

步骤2:连接从库-sqlNode2和sqlNode3

两个从库的端口分别是:1502和1503.然后重复主库执行的操作,如下:

CREATE LOGIN dbm_login WITH PASSWORD = 'MyStr0ngPa$w0rd';
CREATE USER dbm_user FOR LOGIN dbm_login;
GO
 
CREATE MASTER KEY ENCRYPTION BY PASSWORD = 'MyStr0ngPa$w0rd';
GO
CREATE CERTIFICATE dbm_certificate   
    AUTHORIZATION dbm_user
    FROM FILE = '/tmp/dbm_certificate.cer'
    WITH PRIVATE KEY (
    FILE = '/tmp/dbm_certificate.pvk',
    DECRYPTION BY PASSWORD = 'MyStr0ngPa$w0rd'
);
GO

步骤3:所有节点

在所有节点上执行以下命令

CREATE ENDPOINT [Hadr_endpoint]
    AS TCP (LISTENER_IP = (0.0.0.0), LISTENER_PORT = 5022)
    FOR DATA_MIRRORING (
        ROLE = ALL,
        AUTHENTICATION = CERTIFICATE dbm_certificate,
        ENCRYPTION = REQUIRED ALGORITHM AES
        );
ALTER ENDPOINT [Hadr_endpoint] STATE = STARTED;
GRANT CONNECT ON ENDPOINT::[Hadr_endpoint] TO [dbm_login];

启用开机自启动ALWAYON,在所有节点执行以下命令

ALTER EVENT SESSION  AlwaysOn_health ON SERVER WITH (STARTUP_STATE=ON);
GO

步骤4:创建高可用组

可以用SSMS工具和T-SQL两种方式,下面以T-SQL为例:

运行以下脚本在主节点中创建一个可用性组。 请注意,选择CLUSTER_TYPE = NONE选项是因为它是在没有诸如Pacemaker或Windows Server故障转移群集之类的群集管理平台的情况下安装的。

如果要在Linux上安装AlwaysOn AG,则应为Pacemaker选择CLUSTER_TYPE = EXTERNAL:

CREATE AVAILABILITY GROUP [AG1]
        WITH (CLUSTER_TYPE = NONE)
        FOR REPLICA ON
        N'sqlNode1'
            WITH (
            ENDPOINT_URL = N'tcp://sqlNode1:5022',
            AVAILABILITY_MODE = ASYNCHRONOUS_COMMIT,
                SEEDING_MODE = AUTOMATIC,
                FAILOVER_MODE = MANUAL,
            SECONDARY_ROLE (ALLOW_CONNECTIONS = ALL)
                ),
        N'sqlNode2'
            WITH (
            ENDPOINT_URL = N'tcp://sqlNode2:5022',
            AVAILABILITY_MODE = ASYNCHRONOUS_COMMIT,
                SEEDING_MODE = AUTOMATIC,
                FAILOVER_MODE = MANUAL,
            SECONDARY_ROLE (ALLOW_CONNECTIONS = ALL)
                ),
        N'sqlNode3'
            WITH (
            ENDPOINT_URL = N'tcp://sqlNode3:5022',
            AVAILABILITY_MODE = ASYNCHRONOUS_COMMIT,
                SEEDING_MODE = AUTOMATIC,
                FAILOVER_MODE = MANUAL,
            SECONDARY_ROLE (ALLOW_CONNECTIONS = ALL)
                );
GO

在从库中执行以下命令,将从库加入到AG组中

ALTER AVAILABILITY GROUP [ag1] JOIN WITH (CLUSTER_TYPE = NONE);
ALTER AVAILABILITY GROUP [ag1] GRANT CREATE ANY DATABASE;
GO

至此在Docker容器中安装SQL Server Alwayson集群已经完成了!

注意:当指定CLUSTER_TYPE = NONE创建可用组时,在执行故障转移时需执行以下命令

ALTER AVAILABILITY GROUP [ag1] FORCE_FAILOVER_ALLOW_DATA_LOSS

测试

在主库上创建一个数据库,并加入到可用组AG中。

CREATE DATABASE agtestdb;
GO
ALTER DATABASE agtestdb SET RECOVERY FULL;
GO
BACKUP DATABASE agtestdb TO DISK = '/var/opt/mssql/data/agtestdb.bak';
GO
ALTER AVAILABILITY GROUP [ag1] ADD DATABASE [agtestdb];
GO

通过SSMS查看同步状态是否正常.

参考连接

  1. https://docs.microsoft.com/en-us/sql/linux/quickstart-install-connect-docker?view=sql-server-ver15
  2. https://docs.microsoft.com/en-us/sql/linux/quickstart-install-connect-ubuntu?view=sql-server-ver15
  3. https://docs.microsoft.com/en-us/sql/linux/sql-server-linux-create-availability-group?view=sql-server-ver15
  4. https://docs.microsoft.com/en-us/sql/linux/sql-server-linux-configure-mssql-conf?view=sql-server-ver15
  5. https://docs.microsoft.com/en-us/sql/linux/sql-server-linux-configure-environment-variables?view=sql-server-ver15
  6. https://docs.microsoft.com/en-us/sql/linux/sql-server-linux-availability-group-cluster-ubuntu?view=sql-server-linux-ver15
  7. https://docs.docker.com/engine/install/ubuntu/
  8. https://docs.docker.com/compose/
GitHub 加速计划 / li / linux-dash
6
1
下载
A beautiful web dashboard for Linux
最近提交(Master分支:3 个月前 )
186a802e added ecosystem file for PM2 4 年前
5def40a3 Add host customization support for the NodeJS version 4 年前
Logo

旨在为数千万中国开发者提供一个无缝且高效的云端环境,以支持学习、使用和贡献开源项目。

更多推荐