目录
8、改配置文件postgresql.conf（主库）... 3
11、配置PostgreSQL-MD5认证（主库）... 4
12、配置PostgreSQL-Configure TCP/IP（主库）... 5
20、从库上修改postgresql.conf（备库）... 7
21、从库增加recovery.conf配置文件（从库）... 7
1、环境规划
| ip |
os |
role |
| 10.45.11.177 |
CentOS7.6 |
master |
| 10.45.11.178 |
CentOS7.6 |
slave |
2、安装pg专用yum源（双节点）
Install the repository RPM:
yum install https://download.postgresql.org/pub/repos/yum/reporpms/EL-7-x86_64/pgdg-redhat-repo-latest.noarch.rpm
3、安装pg客户端（双节点）
Install the client packages:
yum install postgresql11
4、安装pgserver（双节点）
Optionally install the server packages:
yum install postgresql11-server
集群版额外安装（双节点）
yum -y install postgresql11-contrib
5、初始化数据库（双节点）
Optionally initialize the database and enable automatic start:
/usr/pgsql-11/bin/postgresql-11-setup initdb
/usr/pgsql-11/bin/initdb -D $PGDATA
postgres@ossec01[/postgres]$/usr/pgsql-11/bin/initdb -D $PGDATA
The files belonging to this database system will be owned by user "postgres".
This user must also own the server process.
The database cluster will be initialized with locale "en_US.UTF-8".
The default database encoding has accordingly been set to "UTF8".
The default text search configuration will be set to "english".
Data page checksums are disabled.
fixing permissions on existing directory /postgres/data ... ok
creating subdirectories ... ok
selecting default max_connections ... 100
selecting default shared_buffers ... 128MB
selecting default timezone ... Asia/Shanghai
selecting dynamic shared memory implementation ... posix
creating configuration files ... ok
running bootstrap script ... ok
performing post-bootstrap initialization ... ok
syncing data to disk ... ok
WARNING: enabling "trust" authentication for local connections
You can change this by editing pg_hba.conf or using the option -A, or
--auth-local and --auth-host, the next time you run initdb.
Success. You can now start the database server using:
/usr/pgsql-11/bin/pg_ctl -D /postgres/data -l logfile start
postgres@ossec01[/postgres]$
6、启动服务（双节点）
systemctl enable postgresql-11
systemctl start postgresql-11
7、配置postgres用户环境变量（双节点）
export PGHOME=/usr/pgsql-11
export PGDATA=/postgres/data
export PATH=$PATH:$HOME/bin:$PGHOME/bin
使用rpm包安装，会自动创建postgres用户。如果用源码包安装，需要自行创建postgres用户。
8、改配置文件postgresql.conf（主库）
配置文件就在data目录
postgres@ossec01[/postgres/data]$ls
base pg_dynshmem pg_logical pg_replslot pg_stat pg_tblspc pg_wal postgresql.conf
global pg_hba.conf pg_multixact pg_serial pg_stat_tmp pg_twophase pg_xact
pg_commit_ts pg_ident.conf pg_notify pg_snapshots pg_subtrans PG_VERSION postgresql.auto.conf
postgres@ossec01[/postgres/data]$pwd
/postgres/data
postgres@ossec01[/postgres/data]$vi postgresql.conf
data_directory = '/postgres/data'
unix_socket_directories = '/postgres/data, /tmp'
log_directory = '/postgres/logs'
9、pg_ctl启动pg服务（主库）
postgres@ossec01[/postgres/data]$pg_ctl start
waiting for server to start....2019-09-12 19:54:50.744 CST [8809] LOG: listening on IPv6 address "::1", port 5432
2019-09-12 19:54:50.744 CST [8809] LOG: listening on IPv4 address "127.0.0.1", port 5432
2019-09-12 19:54:50.770 CST [8809] LOG: listening on Unix socket "/postgres/data/.s.PGSQL.5432"
2019-09-12 19:54:50.772 CST [8809] LOG: listening on Unix socket "/tmp/.s.PGSQL.5432"
2019-09-12 19:54:50.782 CST [8809] LOG: redirecting log output to logging collector process
2019-09-12 19:54:50.782 CST [8809] HINT: Future log output will appear in directory "/postgres/logs".
done
server started
10、改用户密码（主库）
postgres@ossec01[/postgres/data]$psql
psql (11.5)
Type "help" for help.
postgres=# alter user postgres with encrypted password 'postgres';
ALTER ROLE
postgres=#
11、配置PostgreSQL-MD5认证（主库）
vi /postgres/data/pg_hba.conf
最后加入如下配置：
host all all 0.0.0.0/0 md5
12、配置PostgreSQL-Configure TCP/IP（主库）
默认情况下，TCP/IP连接是不可行的，所以其他计算机用户不能连接到postgresql。
修改文件：vi /postgres/data/postgresql.conf 增加如下配置：
vi /postgres/data/postgresql.conf
listen_addresses = '*'
port = 5432
max_connections = 500
13、重启pg服务（主库）
postgres@ossec02[/postgres]$pg_ctl restart
waiting for server to shut down.... done
server stopped
waiting for server to start....2019-09-16 19:06:24.896 CST [7931] LOG: listening on IPv4 address "0.0.0.0", port 5432
2019-09-16 19:06:24.896 CST [7931] LOG: listening on IPv6 address "::", port 5432
2019-09-16 19:06:24.898 CST [7931] LOG: listening on Unix socket "/postgres/data/.s.PGSQL.5432"
2019-09-16 19:06:24.901 CST [7931] LOG: listening on Unix socket "/tmp/.s.PGSQL.5432"
2019-09-16 19:06:25.008 CST [7931] LOG: redirecting log output to logging collector process
2019-09-16 19:06:25.008 CST [7931] HINT: Future log output will appear in directory "/postgres/logs".
done
server started
14、配置流复制（主库）
至此前期准备工作已经完成。
下面就是本文的重点了，主备流复制的配置。
vi /postgres/data/postgresql.conf
在最后添加如下属性：
wal_level = hot_standby
#checkpoint_segments = 16 #该配置项在11.5版本中不存在，否则报错
checkpoint_timeout = 5min
archive_mode = on
archive_command ='cp -i %p /postgres/data/arch/%f </dev/null'
max_wal_senders = 3
wal_keep_segments = 16
15、创建流复制账号（主库）
创建具有replication权限的用户
postgres@ossec02[/postgres/data]$psql
psql (11.5)
Type "help" for help.
postgres=# create user rep replication login encrypted password 'postgres';
CREATE ROLE
postgres=#
16、修改配置文件pg_hba.conf
增加复制属性
vi /postgres/data/pg_hba.conf
host replication rep 0.0.0.0/0 md5
17、停止备库的PostgreSQL服务（备库）
postgres@ossec02[/postgres/data]$pg_ctl stop
waiting for server to shut down..... done
server stopped
18、清空备库之前的数据文件（备库）
postgres@ossec02[/postgres/data]$rm -rf /postgres/data/*
19、恢复备库（备库）
从主库上恢复数据，输入下面命令，密码是之前配置的rep复制用户的密码
在从库上执行的，-h后面跟的ip是主库的ip
pg_basebackup -D /postgres/data/ -F p -X stream -v -P -h 10.45.11.177 -U rep
postgres@ossec02[/postgres/data]$pg_basebackup -D /postgres/data/ -F p -X stream -v -P -h 10.45.11.177 -U rep
Password:
pg_basebackup: initiating base backup, waiting for checkpoint to complete
WARNING: skipping special file "./.s.PGSQL.5432"
pg_basebackup: checkpoint completed
pg_basebackup: write-ahead log start point: 0/2000028 on timeline 1
pg_basebackup: starting background WAL receiver
pg_basebackup: created temporary replication slot "pg_basebackup_8551"
WARNING: skipping special file "./.s.PGSQL.5432"000000010000000000000001)
40934/40934 kB (100%), 1/1 tablespace
pg_basebackup: write-ahead log end point: 0/2000130
pg_basebackup: waiting for background process to finish streaming ...
pg_basebackup: base backup completed
20、从库上修改postgresql.conf（备库）
vi /postgres/data/postgresql.conf
增加如下配置项：
hot_standby = on
21、从库增加recovery.conf配置文件（从库）
从库增加recovery.conf配置文件：
vi /postgres/data/recovery.conf
standby_mode = on
primary_conninfo = 'host=10.45.11.177 port=5432 user=rep password=postgres'
trigger_file = '/postgres/data/postgresql.trigger.5432'
chmod 777 /postgres/data/recovery.conf
22、启动PostgreSQL服务
postgres@ossec02[/postgres/data]$pg_ctl start
waiting for server to start....2019-09-16 23:44:18.715 CST [8746] LOG: listening on IPv4 address "0.0.0.0", port 5432
2019-09-16 23:44:18.715 CST [8746] LOG: listening on IPv6 address "::", port 5432
2019-09-16 23:44:18.717 CST [8746] LOG: listening on Unix socket "/postgres/data/.s.PGSQL.5432"
2019-09-16 23:44:18.719 CST [8746] LOG: listening on Unix socket "/tmp/.s.PGSQL.5432"
2019-09-16 23:44:19.034 CST [8746] LOG: redirecting log output to logging collector process
2019-09-16 23:44:19.034 CST [8746] HINT: Future log output will appear in directory "/postgres/logs".
done
server started
23、验证配置
23.1、主库检查复制进程
在主库上检查wal进程,会看到多出一个wal sender process的进程
postgres@ossec01[/postgres/data]$ps -aux | grep postgres
root 7749 0.0 0.0 191772 2348 pts/0 S 18:41 0:00 su - postgres
postgres 7750 0.0 0.0 115436 2032 pts/0 S 18:41 0:00 -bash
postgres 8107 0.0 1.0 415068 29448 pts/0 S 19:31 0:00 /usr/pgsql-11/bin/postgres
postgres 8108 0.0 0.0 249356 1924 ? Ss 19:31 0:00 postgres: logger
postgres 8110 0.0 0.1 415316 3904 ? Ss 19:31 0:00 postgres: checkpointer
postgres 8111 0.0 0.1 415200 3604 ? Ss 19:31 0:00 postgres: background writer
postgres 8112 0.0 0.2 415068 6240 ? Ss 19:31 0:05 postgres: walwriter
postgres 8113 0.0 0.1 415836 3204 ? Ss 19:31 0:00 postgres: autovacuum launcher
postgres 8114 0.0 0.0 251476 2024 ? Ss 19:31 0:00 postgres: archiver last was 000000010000000000000002.00000028.backup
postgres 8115 0.0 0.0 251608 2192 ? Ss 19:31 0:00 postgres: stats collector
postgres 8116 0.0 0.0 415680 2808 ? Ss 19:31 0:00 postgres: logical replication launcher
postgres 8582 0.0 0.1 415844 3656 ? Ss 23:44 0:00 postgres: walsender rep 10.45.11.178(26317) streaming 0/3000140
postgres 8586 0.0 0.0 155360 1880 pts/0 R+ 23:45 0:00 ps -aux
postgres 8587 0.0 0.0 112708 972 pts/0 S+ 23:45 0:00 grep --color=auto postgres
23.2、备库检查复制进程
在从库上有个walreceiver streaming的进程
postgres@ossec02[/postgres/data]$ps -aux | grep postgres
root 7824 0.0 0.0 191772 2348 pts/0 S 18:41 0:00 su - postgres
postgres 7825 0.0 0.0 115436 2016 pts/0 S 18:41 0:00 -bash
postgres 8746 0.0 1.0 415068 29336 pts/0 S 23:44 0:00 /usr/pgsql-11/bin/postgres
postgres 8747 0.0 0.0 249356 1920 ? Ss 23:44 0:00 postgres: logger
postgres 8748 0.0 0.0 415232 2788 ? Ss 23:44 0:00 postgres: startup recovering 000000010000000000000003
postgres 8749 0.0 0.1 415068 3360 ? Ss 23:44 0:00 postgres: checkpointer
postgres 8750 0.0 0.1 415068 3112 ? Ss 23:44 0:00 postgres: background writer
postgres 8751 0.0 0.0 251476 1788 ? Ss 23:44 0:00 postgres: stats collector
postgres 8752 0.1 0.1 421908 3668 ? Ss 23:44 0:01 postgres: walreceiver streaming 0/3000140
postgres 8757 0.0 0.0 155360 1880 pts/0 R+ 23:54 0:00 ps -aux
postgres 8758 0.0 0.0 112708 972 pts/0 S+ 23:54 0:00 grep --color=auto postgres
23.3、主库上建表和插入数据
postgres@ossec01[/postgres/data]$psql
psql (11.5)
Type "help" for help.
postgres=# \c
You are now connected to database "postgres" as user "postgres".
postgres=# \d
Did not find any relations. ^
postgres=# create table tmp_list_001(id int ,name varchar(10));
CREATE TABLE
postgres=# insert into tmp_list_001 values(1,'haha');
INSERT 0 1
postgres=#
23.4、从库上验证是否有对应表和数据
postgres@ossec02[/postgres/logs]$psql
psql (11.5)
Type "help" for help.
postgres=# \d
List of relations
Schema | Name | Type | Owner
--------+--------------+-------+----------
public | tmp_list_001 | table | postgres
(1 row)
postgres=# select * from tmp_list_001 ;
id | name
----+------
1 | haha
(1 row)
postgres=#
测试完成。
