CentOS7下PostgreSQL主从流复制部署

一、准备环境

操作系统：CentOS Linux release 7.4
数据库版本：PostgreSQL 9.6.12
主库:192.168.189.152
从库:192.168.189.153

二、主机数据库部署

以下操作步骤请在两个节点进行，主库需要安装数据库软件以及初始化数据库，从库仅需要安装数据库软件即可无需初始化。
1、添加RPM源

[root@localhost opt]# yum install https://download.postgresql.org/pub/repos/yum/9.6/redhat/rhel-7-x86_64/pgdg-centos96-9.6-3.noarch.rpm

2、使用yum直接安装postgresql

[root@localhost opt]# yum -y install postgresql96-server postgresql96-contrib postgresql96-devel

3、创建postgresql文件目录以及用户组

创建数据文件目录
[root@localhost opt]# mkdir -p /data/pgdata/

创建pg用户组
[root@localhost opt]# groupadd pg

[root@localhost opt]# useradd -g pg pg

对数据文件目录授予权限
[root@localhost opt]# chown -R pg:pg /data/
[root@localhost opt]# chown -R pg:pg /var/run/postgresql

4、初始化数据库
注意：此步骤仅主库需要初始化，从库不需要

[pg@localhost ~]$ /usr/pgsql-9.6/bin/initdb -D /data/pgdata/

初始化过程：

The files belonging to this database system will be owned by user "pg".
This user must also own the server process.

The database cluster will be initialized with locale "en_US.UTF-8".
The default database encoding has accordingly been set to "UTF8".
The default text search configuration will be set to "english".

Data page checksums are disabled.

fixing permissions on existing directory /data/pgdata ... ok
creating subdirectories ... ok
selecting default max_connections ... 100
selecting default shared_buffers ... 128MB
selecting dynamic shared memory implementation ... posix
creating configuration files ... ok
running bootstrap script ... ok
performing post-bootstrap initialization ... ok
syncing data to disk ... ok

WARNING: enabling "trust" authentication for local connections
You can change this by editing pg_hba.conf or using the option -A, or
--auth-local and --auth-host, the next time you run initdb.

Success. You can now start the database server using:

初始化完毕之后启动数据库

-->/usr/pgsql-9.6/bin/pg_ctl -D /data/pgdata/ -l logfile start

 附停止命令：[postgres@localhost pgsql]$ pg_ctl stop -D /mydata/pgdata/备注：     

5、添加环境变量

[pg@localhost ~]$ vi ~/.bash_profile    
PGDATA=/data/pgdata
PGHOST=127.0.0.1
PGDATABASE=postgres
PGUSER=pg
PGPORT=5432
PATH=/usr/pgsql-9.6/bin:$PATH
export PATH
export PGDATA PGHOST PGDATABASE PGUSER PGPORT
export TMOUT=1000
export LD_LIBRARY_PATH=/usr/pgsql-9.6/lib

步骤6、7仅需在主库操作

6、修改监听配置

修改如下监听参数，是监听允许任何地址使用

[pg@localhost ~]$ vi /data/pgdata/postgresql.conf
listen_addresses = '*'

7、添加允许访问数据库的网段

在127.0.0.1下一段添加网段
 [pg@localhost ~]$ vi /data/pgdata/pg_hba.conf 
 # IPv4 local connections:
host    all             all             127.0.0.1/32        trust
添加此行—->host    all             all             0.0.0.0/0            trust

三、主备数据库配置

3.1 主库配置

(1)创建复制用户，专门进行主从同步使用

[root@localhost ~]# su - pg
[pg@localhost ~]$ psql
postgres=# create user rpl superuser password '11111';

(2)主库上配置从库允许的网段
添加从库网段

[pg@localhost ~]$  vi /data/pgdata/pg_hba.conf
host    replication     rpl            192.168.0.0/0         md5

(3)修改主库参数文件

[pg@localhost ~]$ vi /data/pgdata/postgresql.conf
wal_level = hot_standby   -->启用流复制
max_wal_senders=2	  -->流复制允许连接进程
wal_keep_segments =64-->xlog目录中最多容纳多少个wal日志文件，超过了则删掉最初的几个。
max_connections = 1000 -->主库最大连接数

(4)重启主库服务

[pg@localhost ~]$  pg_ctl  -D /data/pgdata/ -l logfile restart

3.2 从库配置

(1)基础备份
在从库无需拷贝主库的数据文件到本地，使用pg_basebackup可将主库的数据文件通过网络复制到本地、包括目录结构。

备注：如果主库存在自定义表空间，此操作会失败

[pg@localhost ~]$ pg_basebackup -h 192.168.212.152 -p 5432 -U rpl -F p -x -P -R -D /data/pgdata
Password: 
46694/46694 kB (100%), 1/1 tablespace
传输完毕

(2) 从库配置文件配置

[pg@localhost ~]$ vi /data/pgdata/postgresql.conf
注释这三个参数wal_level，max_wal_senders,wal_keep_segments

修改
hot_standby = on ———>从库上执行只读操作
max_standby_streaming_delay = 30s -->从库接收主库日志最大超时时间
wal_receiver_status_interval = 10s -->从库向主库报告的最大时间间隔
hot_standby_feedback = off -->从库与主库发生冲突的反馈信息
max_connections = 1500 -->数据库最大连接数

3)创建恢复文件recovery.conf
recovery.conf 用于主库，从库切换时的参数配置。如果从库没有recovery.conf文件可拷贝模板

[pg@localhost pgdata]$ cp /usr/pgsql-9.6/share/recovery.conf.sample /data/pgdata/recovery.conf

[pg@localhost pgdata]$ vi /data/pgdata/recovery.conf
调整参数：
recovery_target_timeline = 'latest'  moren
standby_mode = on
primary_conninfo = 'host=192.168.212.152 port=5432 user=rpl password=11111'

(4)启动从库服务

[pg@localhost pgdata]$ pg_ctl -D /data/pgdata/ -l logfile start

四、主从状态检查

查看主库sender进程信息
[root@localhost ~]# ps -ef|grep "wal sender process"
pg        7031  7015  0 17:04 ?        00:00:00 postgres: wal sender process rpl 192.168.212.153(49562) streaming 0/50001E8

查看从库receiver进程信息
[root@localhost ~]# ps -ef|grep "wal receiver"
pg        2360  2354  0 17:04 ?        00:00:00 postgres: wal receiver process   streaming 0/50002C8


psql查询主从同步状态
postgres=# select usename,application_name,client_addr,state,sync_state from  pg_stat_replication; 
 usename | application_name |   client_addr   |   state   | sync_state 
---------+------------------+-----------------+-----------+------------
 rpl    | walreceiver      | 192.168.212.153 | streaming | async-->异步
(1 row)

主库集群状态
[pg@localhost ~]$ pg_controldata /data/pgdata/|grep "Database cluster state"
Database cluster state:               in production

从库集群状态
[pg@localhost ~]$ pg_controldata /data/pgdata/|grep "Database cluster state"
Database cluster state:               in archive recovery

主从数据同步验证
主库建表、从库进行查询测试

主库：
pg=# insert into test values(1);

pg=# select * from test;        
 id 
----
  1
(1 row)

从库：
pg=# insert into test values(2);
ERROR:  cannot execute INSERT in a read-only transaction
pg=# select * from test;        
 id 
----
  1
(1 row)
从库只有只读功能，无法做写入操作