容器运行polardb-pg配置主从

# 容器运行polardb-pg配置主从

环境：

操作系统：ubuntu:22.04
docker: 28.4.0
polardb镜像: polardb/polardb_pg_local_instance:11
网段：192.168.5.0/24，主：192.168.5.118

# 主服务器配置

# docker-compose.yaml

cat >docker-compose.yaml<<'EOF'
version: '3'
services:
  polardb-11:
    restart: always
    container_name: polardb-11
    #image: polardb/polardb_pg_local_instance:11
    image: docker.1ms.run/polardb/polardb_pg_local_instance:11
    volumes:
      - ./polardb:/var/polardb
      - /etc/localtime:/etc/localtime
    ports:
      - "5432:5432"
    environment:
      - TZ=Asiz/Shanghai
    privileged: true
    cap_add:
      - SYS_PTRACE
    network_mode: bridge
    entrypoint: ["./docker-entrypoint.sh","postgres"]
EOF

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

启动

docker-compose -f docker-compose.yaml up -d

主库创建复制槽

poladb-11]# docker exec -it polardb-11 bash

postgres@2636328ce7b3:~$ psql -U postgres


postgres=# SELECT pg_create_physical_replication_slot('replica1_slot', true);
postgres=# SELECT pid,usename,application_name,state,sync_state FROM pg_stat_replication;

1
2
3
4
5
6
7
8

查看复制流

SELECT pid,usename,application_name,state,sync_state FROM pg_stat_replication;

修改配置

类似和postgresql,配置主从同步

[root@location-197 poladb-11]# tail -2 ./polardb/primary_datadir/pg_hba.conf
host    replication     all             192.168.5.0/24                 trust

1
2

重启服务

docker restart poladb-11

# 从服务器配置

mkdir -p /data/polardb-11
cd /data/polardb-11

1
2

# 基于现有的docker镜像，构建一个初始环境的镜像

cat >Dockerfile<<'EOF'
FROM docker.1ms.run/polardb/polardb_pg_local_instance:11

ENTRYPOINT ["/bin/bash"]
EOF

#修改默认启动方式
docker build . -t docker.1ms.run/polardb/polardb_pg_local_instance:11-2

1
2
3
4
5
6
7
8

# 启动一个初始的

cd /data/polardb-11

mkdir -p `pwd`/polardb/replica
chmod 750 `pwd`/polardb/replica
chown -R 1000:999 `pwd`/polardb

#启动一个初始的
docker run -d --name polar11_replica1   -v `pwd`/polardb:/var/polardb  -p 5433:5432   -e POSTGRES_PASSWORD=Yfklife#csdm8888 polardb/polardb_pg_local_instance:11-2  -c "tail -F test"

1
2
3
4
5
6
7
8

# 开始复制

docker exec -it  polar11_replica1  bash

export POSTGRES_PASSWORD=Dimine#csdm8888
pg_basebackup -h 192.168.5.118 -U postgres -Fp -Xs -P -R -D /var/polardb/replica -v


#退出初始容器,在宿主机执行
exit

1
2
3
4
5
6
7
8

# 创建软链接，启动需要

docker exec -it polar11_replica1 ln -s /var/polardb/replica/polar_shared_data /var/polardb/shared_datadir

# 进入容器，启动-检查服务是否正常

docker exec -it polar11_replica1  bash
pg_ctl -D /var/polardb/replica start

#等待大概一两分钟，登录，测试在主库创建库，是否同步
psql -U postgres

1
2
3
4
5

# 改成docker-compose启动

cat >docker-compose.yaml<<'EOF'
version: '3'
services:
  polardb-11:
    restart: always
    container_name: polardb-11
    image: docker.1ms.run/polardb/polardb_pg_local_instance:11-2
    volumes:
      - ./polardb:/var/polardb
      - /etc/localtime:/etc/localtime
    ports:
      - "5432:5432"
    environment:
      - TZ=Asiz/Shanghai
    privileged: true
    cap_add:
      - SYS_PTRACE
    entrypoint: ["postgres","-D","/var/polardb/replica"]
    network_mode: bridge
EOF

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

启动数据库

docker rm -f polar11_replica1
docker-compose -f docker-compose.yaml up -d

1
2

再次检查服务是否正常
连上主库查询流

SELECT pid,usename,application_name,state,sync_state FROM pg_stat_replication;

# 从库切换为主库

docker exec -it poladb-11 pg_ctl -D /var/polardb/replica promote

# 遇到的问题

环境：

centos7.9 
docker version: 19.03

启动报错信息：怀疑是docker版本原因，或者centos系统，换回ubuntu22.04 正常启动

postgres@f6a56b85d5b0:/var/polardb/replica$ pg_ctl  -D /var/polardb/replica/ start
popen failure: Cannot allocate memory
The program "postgres" is needed by pg_ctl but was not found in the
same directory as "/home/postgres/tmp_basedir_polardb_pg_1100_bld/bin/pg_ctl".
Check your installation.


postgres@f6a56b85d5b0:/var/polardb/replica$ ldd  /home/postgres/tmp_basedir_polardb_pg_1100_bld/bin/postgres
ldd: error: you do not have read permission for `/home/postgres/tmp_basedir_polardb_pg_1100_bld/bin/postgres'

1
2
3
4
5
6
7
8
9

如果此文章对您有帮助，点击 -->> 请博主喝咖啡

← 容器运行polardb-pg