pg_standby备库搭建

1.主库

1.1主库参数文件修改

-- 该路径也需要在从库创建
mkdir -p /postgresql/archive
chown -R postgres.postgres /postgresql/archive

-- 主库配置归档
wal_level='replica'
archive_mode='on'
archive_command='cp %p /postgresql/archive/%f'
restore_command='cp /postgresql/archive/%f %p'

max_wal_senders=10
wal_sender_timeout=60s
wal_keep_size=256

--重启数据库
pg_ctl stop
pg_ctl start

1.2 查看pg进程信息

[root@test archive]# ps -ef|grep post
root      1485     1  0 08:57 ?        00:00:00 /usr/libexec/postfix/master -w
postfix   1502  1485  0 08:57 ?        00:00:00 qmgr -l -t unix -u
postfix  18200  1485  0 13:55 ?        00:00:00 pickup -l -t unix -u
pgsql    22160     1  0 14:12 ?        00:00:00 /postgresql/pg13/bin/postgres
pgsql    22162 22160  0 14:12 ?        00:00:00 postgres: logger
pgsql    22164 22160  0 14:12 ?        00:00:00 postgres: checkpointer
pgsql    22165 22160  0 14:12 ?        00:00:00 postgres: background writer
pgsql    22166 22160  0 14:12 ?        00:00:00 postgres: walwriter
pgsql    22167 22160  0 14:12 ?        00:00:00 postgres: autovacuum launcher
pgsql    22168 22160  0 14:12 ?        00:00:00 postgres: archiver
pgsql    22169 22160  0 14:12 ?        00:00:00 postgres: stats collector
pgsql    22170 22160  0 14:12 ?        00:00:00 postgres: logical replication launcher
pgsql    22172 22160  0 14:12 ?        00:00:00 postgres: walsender pgsql 192.168.56.125(40206) streaming 0/B0000D8
pgsql    22174  2756  0 14:12 pts/0    00:00:00 rlwrap psql -h 192.168.56.100 -p 5432 postgres pgsql
pgsql    22175 22174  0 14:12 pts/1    00:00:00 psql -h 192.168.56.100 -p 5432 postgres pgsql
pgsql    22180 22160  0 14:12 ?        00:00:00 postgres: pgsql postgres 192.168.56.100(64860) idle
root     22192 13050  0 14:12 pts/2    00:00:00 grep --color=auto post

1.3查看归档信息

select * from pg_settings where name in ('wal_level','archive_mode','archive_command');
--切换归档
select pg_switch_wal();
--查看归档状态
test=# select * from pg_stat_get_archiver();
-[ RECORD 1 ]------+-----------------------------------------
archived_count     | 8
last_archived_wal  | 000000010000000000000008.00000028.backup
last_archived_time | 2023-07-28 10:44:59.302381+08
failed_count       | 0
last_failed_wal    |
last_failed_time   |
stats_reset        | 2023-07-27 16:35:50.49841+08

--操作系统级别
[root@test archive]# pwd
/postgresql/archive
[root@test archive]#ll
-rw------- 1 pgsql pgsql 16777216 Jul 28 10:17 000000010000000000000001
-rw------- 1 pgsql pgsql 16777216 Jul 28 10:17 000000010000000000000002
-rw------- 1 pgsql pgsql 16777216 Jul 28 10:18 000000010000000000000003
-rw------- 1 pgsql pgsql 16777216 Jul 28 10:19 000000010000000000000004
-rw------- 1 pgsql pgsql 16777216 Jul 28 10:21 000000010000000000000005
-rw------- 1 pgsql pgsql 16777216 Jul 28 10:25 000000010000000000000006
-rw------- 1 pgsql pgsql 16777216 Jul 28 10:44 000000010000000000000007
-rw------- 1 pgsql pgsql 16777216 Jul 28 10:44 000000010000000000000008
-rw------- 1 pgsql pgsql      322 Jul 28 10:44 000000010000000000000008.00000028.backup

1.4切换日之后查看pg进程信息

[root@test archive]# ps -ef|grep post
root      1485     1  0 08:57 ?        00:00:00 /usr/libexec/postfix/master -w
postfix   1502  1485  0 08:57 ?        00:00:00 qmgr -l -t unix -u
postfix  18200  1485  0 13:55 ?        00:00:00 pickup -l -t unix -u
pgsql    22160     1  0 14:12 ?        00:00:00 /postgresql/pg13/bin/postgres
pgsql    22162 22160  0 14:12 ?        00:00:00 postgres: logger
pgsql    22164 22160  0 14:12 ?        00:00:00 postgres: checkpointer
pgsql    22165 22160  0 14:12 ?        00:00:00 postgres: background writer
pgsql    22166 22160  0 14:12 ?        00:00:00 postgres: walwriter
pgsql    22167 22160  0 14:12 ?        00:00:00 postgres: autovacuum launcher
pgsql    22168 22160  0 14:12 ?        00:00:00 postgres: archiver last was 00000001000000000000000B
pgsql    22169 22160  0 14:12 ?        00:00:00 postgres: stats collector
pgsql    22170 22160  0 14:12 ?        00:00:00 postgres: logical replication launcher
pgsql    22172 22160  0 14:12 ?        00:00:00 postgres: walsender pgsql 192.168.56.125(40206) streaming 0/C000060
pgsql    22174  2756  0 14:12 pts/0    00:00:00 rlwrap psql -h 192.168.56.100 -p 5432 postgres pgsql
pgsql    22175 22174  0 14:12 pts/1    00:00:00 psql -h 192.168.56.100 -p 5432 postgres pgsql
pgsql    22180 22160  0 14:12 ?        00:00:00 postgres: pgsql postgres 192.168.56.100(64860) idle
root     22324 13050  0 14:15 pts/2    00:00:00 grep --color=auto post

1.5 备份

--主库创建复制用户
create role repuser login encrypted password 'reppasswd' replication;

2.standby库

--创建备份路径
mkdir -p /bk
chown pgsql.pgsql /bk

-- 远程备份
su - postgres
pg_basebackup -h 192.168.56.100 -p 5432 -U pgsql -l bk20230728 -F p -P -R -D /bk

[root@pg ~]# ll /bk/
-rw-------. 1 pgsql pgsql      0 Jul 27 22:44 standby.signal

-- 关闭从库，删除从库的数据文件，并且将备份文件覆盖从库的数据文件
pg_ctl stop
mv /postgresql/pgdata   /postgresql/20230728pgdata
mkdir /postgresql/pgdata
chmod 700 /postgresql/pgdata
cp -r /bk/* /postgresql/pgdata

--修改配置文件
cat >> /postgresql/pgdata/postgresql.conf <<"EOF"
wal_level='replica'
archive_mode='on'
archive_command='cp %p /postgresql/archive/%f'
restore_command='cp /postgresql/archive/%f %p'

max_wal_senders=10
wal_sender_timeout=60s
wal_keep_size=256
primary_conninfo = 'host=192.168.56.100 port=5432 user=repuser password=reppasswd'

EOF

--重启
pg_ctl stop
pg_ctl start

3.主备状态查看

3.1日志状态查看

--主库查看wal日志发送状态
postgres=# select * from pg_stat_replication;
-[ RECORD 1 ]----+------------------------------
pid              | 22172
usesysid         | 16384
usename          | pgsql
application_name | walreceiver
client_addr      | 192.168.56.125
client_hostname  |
client_port      | 40206
backend_start    | 2023-07-28 14:12:26.157227+08
backend_xmin     |
state            | streaming
sent_lsn         | 0/C000148
write_lsn        | 0/C000148
flush_lsn        | 0/C000148
replay_lsn       | 0/C000148
write_lag        |
flush_lag        |
replay_lag       |
sync_priority    | 0
sync_state       | async
reply_time       | 2023-07-28 14:23:09.779059+08

--从库查看wal日志接收状态
test=# select * from pg_stat_wal_receiver;
-[ RECORD 1 ]---------+------------------------------------------------------------------------------------------------------------
pid                   | 2106
status                | streaming
receive_start_lsn     | 0/B000000
receive_start_tli     | 1
written_lsn           | 0/C000148
flushed_lsn           | 0/C000148
received_tli          | 1
last_msg_send_time    | 2023-07-28 14:25:29.789323+08
last_msg_receipt_time | 2023-07-28 14:25:29.7901+08
latest_end_lsn        | 0/C000148
latest_end_time       | 2023-07-28 14:17:29.470132+08
slot_name             |
sender_host           | 192.168.56.100
sender_port           | 5432
conninfo              | user=pgsql password=******** channel_binding=disable dbname=replication host=192.168.56.100 port=5432 fallback_application_name=walreceiver sslmode=disable sslcompression=0 ssl_min_protocol_version=TLSv1.2 gssencmode=disable krbsrvname=postgres target_session_attrs=any

3.2 通过控制文件查看主备状态

--主
[pgsql@test ~]$ /postgresql/pg13/bin/pg_controldata | grep state
Database cluster state:               in production

--备
[pgsql@pg pgdata]$  /postgresql/pg13/bin/pg_controldata | grep state
Database cluster state:               in archive recovery

3.3通过恢复状态确认主备

--主
postgres=# select pg_is_in_recovery();
-[ RECORD 1 ]-----+--
pg_is_in_recovery | f


--备
test=# select pg_is_in_recovery();
-[ RECORD 1 ]-----+--
pg_is_in_recovery | t

至此，standby部署完成