Configure a StreamingReplication cluster in PostgreSQL

Configure a StreamingReplication cluster in PostgreSQL

Configure a Streaming Replication cluster in PostgreSQL

2. Install PostgreSQL

1) install PostgreSQL software on the Primary and Standy nodes, installation path is/opt/pgsql-9.1.2

2) Set postgres user environment variables

PGHOME =/opt/pgsql-9.1.2

PGDATA =/storage0/database/postgres/main

PATH = $ PG_HOME/bin: $ PATH: $ HOME/bin

3. Primary Node

1) switch to ipvs user

$ Su-postgres

2) initialize the database

$ Initdb

3) Configure pg_cmd.conf

Add a line under # IPv4 local connections to set PostgreSQL access and permissions

Host all 192.168.111.1/24 trust

Add a row under # replication privilege. And set the replication user and permissions.

Host replication ipvs 192.168.111.1/24 trust

4) Configure postgresql. conf

Configure the listener and modify listen_addresses = 'localhost'

Listen_addresses = '*' # what IP address (es) to listen on;

Configure Primary Replication Parameters

Wal_level = hot_standby

Max_wal_senders = 5

Wal_keep_segments = 32

Archive_mode = on

Archive_command = 'cp % p/storage0/database/S/archive/% f </dev/null'

"/Storage0/database/S/archive" is the storage path of the Replication archive. PostgreSQL stores the Replication WAL in the "/storage0/database/S/archive" path.

5) Start the PostgreSQL database on Primary

$ Pg_ctl start

6) run the following command on primary:

$ Psql-c "SELECT pg_start_backup ('label', true )"

Save the files in the PGDATA directory of Primary, except postmaster. copy the pid to the/storage0/database/postgres/main directory of the Standby node, which is the PGDATA directory of the PostgreSQL database on the Standby node.

$ Rsync-a $ {PGDATA}/postgres@192.168.111.151:/storage0/database/S/main -- exclude postmaster. pid

$ Psql-c "SELECT pg_stop_backup ()"

The content in the/storage0/database/postgres/main directory of 192.168.111.151 is

For more details, please continue to read the highlights on the next page: