笔者近来遇到一个任务：在多台Docker主机之间搭建PostgreSQL12的主从复制、读写分离高可用集群运行环境。确保即便在极端情况下(某台Docker主机宕机甚至磁盘损坏)依然能够读取PostgreSQL中的数据，而不至于导致数据丢失。
一、准备工作在两台linux主机(或虚拟机)上分别安装好Docker运行环境，建议Docker使用最新版。笔者使用了两个CentOS 8.3虚拟机。一个虚拟机(命名为dockerserver1)由Hyper-V提供，另一个(命名为虚拟机dockerserver2)由VMWare来提供。
在每个虚拟机里准备好PostgreSQL 12.3的镜像。
二、创建Master容器节点在dockerserver1上为Master和Slave节点创建数据文件夹master、slave和repl，并设置chown 999:999权限。这一步不是必须的，但是建议这样做。因为数据放在容器中在极端情况下很难被恢复。
[root@dockerserver1 postgredata]# mkdir master [root@dockerserver1 postgredata]# mkdir slave1 [root@dockerserver1 postgredata]# mkdir repl [root@dockerserver1 postgredata]# chown 999:999 master [root@dockerserver1 postgredata]# chown 999:999 slave1 [root@dockerserver1 postgredata]# chown 999:999 repl
通过PostgreSQL 12.3镜像创建master节点，使用上一步创建的master文件夹作为PostgreSQL的data文件夹：
[root@dockerserver1 postgredata]# docker run -d -p 5433:5432 -e POSTGRES_PASSWORD=postgres -v /DatabaseDatas/postgredata/master/:/var/lib/postgresql/data --name pg12master -h pg12master -e LANG="C.UTF-8" -e 'TZ=Asia/Shanghai' -m 256m --memory-swap -1 postgres:12.3 238b7234bd4b9662b90c695cbf0ba448a44276928b80deeed5d23382ef48e8c6 [root@dockerserver1 postgredata]# docker ps ConTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 238b7234bd4b postgres:12.3 "docker-entrypoint.s…" 27 hours ago Up 3 hours 0.0.0.0:5433->5432/tcp pg12master
说明：网上已有在Docker下创建PostgreSQL主从复制集群的教程多是基于网桥bridge来创建的，这里笔者不想这么复杂，所以省略了创建docker bridge这一步，直接基于默认的网桥172.17.0.1来创建容器。
三、在master节点中创建repuser的用户[root@dockerserver1 postgredata]# docker exec -it -u postgres pg12master /bin/bash postgres@pg12master:/$ psql psql (12.3 (Debian 12.3-1.pgdg100+1)) Type "help" for help. postgres=# CREATE ROLE repuser WITH LOGIN REPLICATION ConNECTION LIMIT 5 PASSWORD 'repuser'; CREATE ROLE四、创建slave1容器节点
在dockerserver1上创建slave1节点：
[root@dockerserver1 postgredata]# docker run -d -p 5434:5432 --name pg12slave1 -h pg12slave1 -e LANG="C.UTF-8" -e 'TZ=Asia/Shanghai' -e "POSTGRES_DB=postgres" -e "POSTGRES_USER=postgres" -e "POSTGRES_PASSWORD=postgres" -v /DatabaseDatas/postgredata/slave:/var/lib/postgresql/data -v /DatabaseDatas/postgredata/repl:/var/lib/postgresql/repl -m 256m --memory-swap -1 postgres:12.3 a8b1f661c9739e49b2432148e6c003c560d01bd1ae4019142681a877064c02d7
查看slave1容器的IP地址：
[root@dockerserver1 postgredata]# docker inspect --format '{{.NetworkSettings.IPAddress}}' pg12slave1
172.17.0.3
五、修改master的配置信息
1、将slave1的IP地址添加到master节点的信任域
打开master节点的pg_hba.conf，在最下面加入如下内容：
# TYPE DATAbase USER ADDRESS METHOD # "local" is for Unix domain socket connections only local all all trust # IPv4 local connections: host all all 127.0.0.1/32 trust # IPv6 local connections: host all all ::1/128 trust # Allow replication connections from localhost, by a user with the # replication privilege. local replication all trust host replication all 127.0.0.1/32 trust host replication all ::1/128 trust host replication all 0.0.0.0/0 md5 host replication repuser 172.17.0.0/24 md5
上述配置中的最后一行
host replication repuser 172.17.0.0/24 md5
即为新添加的配置。意为只要是同一个网段下的容器都可以访问。网上的教程基本都是配置的固定IP(即pg12slave1节点的IP地址172.17.0.3)。这里为了能够在同一个服务器的Docker环境下创建多个slave容器，所以直接将容器所在的网段172.17.0.*添加到了master的信任域里。
2、修改postgresql.conf，开启master的主从复制功能查找文件中以下几个参数，并调整如下：
archive_mode = on archive_command = '/bin/date' max_wal_senders = 10 wal_keep_segments = 16 synchronous_standby_names = '*'
将上述两个文件保存后，需要重启master节点：
[root@dockerserver1 postgredata]# docker restart pg12master pg12master六、将master的数据同步到slave1中
进入pg12slave1容器内部，执行如下命令：
[root@dockerserver1 postgredata]# docker exec -it -u postgres pg12slave1 /bin/bash postgres@pg12slave1:/$ pg_basebackup -R -D /var/lib/postgresql/repl -Fp -Xs -v -P -h 172.17.0.1 -p 5433 -U repuser Password: pg_basebackup: initiating base backup, waiting for checkpoint to complete pg_basebackup: checkpoint completed pg_basebackup: write-ahead log start point: 0/90000D8 on timeline 1 pg_basebackup: starting background WAL receiver pg_basebackup: created temporary replication slot "pg_basebackup_86" 78736/78736 kB (100%), 1/1 tablespace pg_basebackup: write-ahead log end point: 0/90001E8 pg_basebackup: waiting for background process to finish streaming ... pg_basebackup: syncing data to disk ... pg_basebackup: base backup completed
说明：经笔者反复测试，上述pg_basebackup命令中的-h和-p必须使用pg12master容器所在主机的IP地址，以及映射到主机的端口号，否则会提示连接不到服务器。
七、使用同步到pg12slave1节点的数据重建pg12slave1容器[root@dockerserver1 postgredata]# docker stop pg12slave1 && docker rm pg12slave1 pg12slave1 [root@dockerserver1 postgredata]# rm -rf slave1 [root@dockerserver1 postgredata]# mv repl slave1 [root@dockerserver1 postgredata]# docker run -d --ip 172.17.0.5 -p 5434:5432 --name pg12slave -h pg12slave -e LANG="C.UTF-8" -e 'TZ=Asia/Shanghai' -e "POSTGRES_DB=postgres" -e "POSTGRES_USER=postgres" -e "POSTGRES_PASSWORD=postgres" -v /DatabaseDatas/postgredata/slave1:/var/lib/postgresql/data -m 256m --memory-swap -1 postgres:12.3 a8b1f661c9739e49b2432148e6c003c560d01bd1ae4019142681a877064c02d7
说明：注意此时pg12slave1容器已经不需要再挂载repl目录了。
八、验证pg12master-pg12slave1主从复制集群效果用postgres用户登入pg12master容器内部，执行如下SQL语句，查看关联的从节点信息：
[root@dockerserver1 postgredata]# docker exec -it -u postgres pg12master /bin/bash postgres@pg12master:/$ psql psql (12.3 (Debian 12.3-1.pgdg100+1)) Type "help" for help. postgres=# select * from pg_stat_replication; pid | usesysid | usename | application_name | client_addr | client_hostname | client_port | backend_start | backend_xmin | state | sent_lsn | write_lsn | flush_lsn | replay_lsn | write_l ag | flush_lag | replay_lag | sync_priority | sync_state | reply_time -----+----------+---------+------------------+--------------+-----------------+-------------+-------------------------------+--------------+-----------+-----------+-----------+-----------+------------+-------- ---+-----------+------------+---------------+------------+------------------------------- 149 | 16384 | repuser | walreceiver | 172.17.0.1 | | 51752 | 2021-11-25 15:21:28.80868+08 | | streaming | 0/A004518 | 0/A004518 | 0/A004518 | 0/A004518 | | | | 1 | sync | 2021-11-25 18:42:51.27092+08 (2 rows) postgres=#
能够看到上述记录时说明pg12master-pg12slave1主从复制集群环境搭建成功了。
可以按照上述的操作再在同一个服务器(虚拟机)dockerserver1上再创建更多的slave节点。
九、在dockerserver2上创建pg12slave2容器在dockerserver2服务器上创建pg12slave2的步骤与在dockerserver1上创建pg12slave1的步骤完全相同。只是在pg12slave2内执行pg_basebackup命令之前需要将dockerserver2的IP地址添加到pg12master的信任域中，即在pg_hba.conf中添加如下内容并重启pg12master容器：
# TYPE DATAbase USER ADDRESS METHOD # "local" is for Unix domain socket connections only local all all trust # IPv4 local connections: host all all 127.0.0.1/32 trust # IPv6 local connections: host all all ::1/128 trust # Allow replication connections from localhost, by a user with the # replication privilege. local replication all trust host replication all 127.0.0.1/32 trust host replication all ::1/128 trust host replication all 0.0.0.0/0 md5 host replication repuser 172.17.0.0/24 md5 host replication repuser 192.169.33.0/24 md5
说明：上述内容中最下面一行
host replication repuser 192.169.33.0/24 md5
其中的192.169.33.0即为dockerserver2的主机IP地址。
然后按照同样的步骤执行pg_basebackup命令：
postgres@pg12slave2:/$ pg_basebackup -R -D /var/lib/postgresql/repl -Fp -Xs -v -P -h 192.169.33.2 -p 5433 -U repuser Password: pg_basebackup: initiating base backup, waiting for checkpoint to complete pg_basebackup: checkpoint completed pg_basebackup: write-ahead log start point: 0/90000D8 on timeline 1 pg_basebackup: starting background WAL receiver pg_basebackup: created temporary replication slot "pg_basebackup_86" 78736/78736 kB (100%), 1/1 tablespace pg_basebackup: write-ahead log end point: 0/90001E8 pg_basebackup: waiting for background process to finish streaming ... pg_basebackup: syncing data to disk ... pg_basebackup: base backup completed
然后重新创建pg12slave2容器。成功后在pg12master节点内执行pg_stat_replication SQL查询，就可以看到如下信息：
postgres=# select * from pg_stat_replication; pid | usesysid | usename | application_name | client_addr | client_hostname | client_port | backend_start | backend_xmin | state | sent_lsn | write_lsn | flush_lsn | replay_lsn | write_l ag | flush_lag | replay_lag | sync_priority | sync_state | reply_time -----+----------+---------+------------------+--------------+-----------------+-------------+-------------------------------+--------------+-----------+-----------+-----------+-----------+------------+-------- ---+-----------+------------+---------------+------------+------------------------------- 99 | 16384 | repuser | walreceiver | 192.169.33.3 | | 47044 | 2021-11-25 14:59:24.511927+08 | 1099 | streaming | 0/A004518 | 0/A004518 | 0/A004518 | 0/A004518 | | | | 1 | potential | 2021-11-26 02:42:57.577931+08 149 | 16384 | repuser | walreceiver | 172.17.0.1 | | 51752 | 2021-11-25 15:21:28.80868+08 | | streaming | 0/A004518 | 0/A004518 | 0/A004518 | 0/A004518 | | | | 1 | sync | 2021-11-25 18:42:51.27092+08 (2 rows) postgres=#
OK，大功告成！
十、验证主从同步功能在dockerserver1和dockerserver2内将对应的端口从防火墙中开放出来。然后使用数据库可视化工具去连接后，在主库pg12master中进行数据库创建，即可看到从库pg12slave1和pg12slave2均可实时同步变化后的信息：
以上就是在多个Docker主机之间创建一主多从PostgreSQL集群的搭建步骤。其中，只有master节点是可写的，其他的slave节点都是只读的。从而搭建了读写分离主从同步集群。
大家可以按照这样的步骤扩展其他的slave节点。
参考过的网址：
Docker创建postgesql主从复制保姆级示例_无问西东j的博客-CSDN博客
(57条消息) 使用 Docker 搭建 PostgreSQL 12 主从环境_寒冰屋的专栏-CSDN博客_docker postgres
