servers to create a robust cluster system and avoid the single point of failure or split brain.
</para>
<para>
- <productname>PostgreSQL</productname> 15 is used in this configuration example.
+ <productname>PostgreSQL</productname> 16 is used in this configuration example.
All scripts have been tested with <productname>PostgreSQL</productname> 10 and later.
</para>
<sect2 id="example-cluster-requirement">
<sect2 id="example-cluster-structure">
<title>Cluster System Configuration</title>
<para>
- We use 3 servers with CentOS 7.9 installed. Let these servers be <literal>server1</literal>
- <literal>server2</literal>, <literal>server3</literal>.
+ We use three servers with Rocky Linux 8 installed and
+ the hostnames of the three servers are <literal>server1</literal>
+ <literal>server2</literal> and <literal>server3</literal> respectively.
We install <productname>PostgreSQL</productname> and <productname>Pgpool-II</productname> on each server.
</para>
<para>
</para>
<note>
<para>
- The roles of <literal>Active</literal>, <literal>Standby</literal>, <literal>Primary</literal>,
+ The roles of <literal>Leader</literal>, <literal>Standby</literal>, <literal>Primary</literal>,
<literal>Standby</literal> are not fixed and may be changed by further operations.
</para>
</note>
<tbody>
<row>
<entry>server1</entry>
- <entry>192.168.137.101</entry>
- <entry morerows="2">192.168.137.150</entry>
+ <entry>192.168.100.51</entry>
+ <entry morerows="2">192.168.100.50</entry>
</row>
<row>
<entry>server2</entry>
- <entry>192.168.137.102</entry>
+ <entry>192.168.100.52</entry>
</row>
<row>
<entry>server3</entry>
- <entry>192.168.137.103</entry>
+ <entry>192.168.100.53</entry>
</row>
</tbody>
</tgroup>
<tbody>
<row>
<entry>PostgreSQL Version</entry>
- <entry>15.0</entry>
+ <entry>16.0</entry>
<entry>-</entry>
</row>
<row>
</row>
<row>
<entry>$PGDATA</entry>
- <entry>/var/lib/pgsql/15/data</entry>
+ <entry>/var/lib/pgsql/16/data</entry>
<entry>-</entry>
</row>
<row>
<tbody>
<row>
<entry>Pgpool-II Version</entry>
- <entry>4.3.0</entry>
+ <entry>4.5.0</entry>
<entry>-</entry>
</row>
<row>
<entry>Pgpool-II config file</entry>
</row>
<row>
- <entry>Pgpool-II start user</entry>
+ <entry>User running Pgpool-II</entry>
<entry>postgres (Pgpool-II 4.1 or later)</entry>
- <entry>Pgpool-II 4.0 or before, the default startup user is root</entry>
+ <entry>Pgpool-II 4.0 or before, the default user running Pgpool-II is root</entry>
</row>
<row>
<entry>Running mode</entry>
<tbody>
<row>
<entry morerows='1'>Failover</entry>
- <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/failover.sh.sample;hb=refs/heads/V4_4_STABLE">/etc/pgpool-II/sample_scripts/failover.sh.sample</ulink></entry>
+ <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/failover.sh.sample;hb=refs/heads/V4_5_STABLE">/etc/pgpool-II/sample_scripts/failover.sh.sample</ulink></entry>
<entry>Run by <xref linkend="GUC-FAILOVER-COMMAND"> to perform failover</entry>
</row>
<row>
- <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/follow_primary.sh.sample;hb=refs/heads/V4_4_STABLE">/etc/pgpool-II/sample_scripts/follow_primary.sh.sample</ulink></entry>
+ <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/follow_primary.sh.sample;hb=refs/heads/V4_5_STABLE">/etc/pgpool-II/sample_scripts/follow_primary.sh.sample</ulink></entry>
<entry>Run by <xref linkend="GUC-FOLLOW-PRIMARY-COMMAND"> to synchronize the Standby with the new Primary after failover.</entry>
</row>
<row>
<entry morerows='1'>Online recovery</entry>
- <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/recovery_1st_stage.sample;hb=refs/heads/V4_4_STABLE">/etc/pgpool-II/sample_scripts/recovery_1st_stage.sample</ulink></entry>
+ <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/recovery_1st_stage.sample;hb=refs/heads/V4_5_STABLE">/etc/pgpool-II/sample_scripts/recovery_1st_stage.sample</ulink></entry>
<entry>Run by <xref linkend="GUC-RECOVERY-1ST-STAGE-COMMAND"> to recovery a Standby node</entry>
</row>
<row>
- <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/pgpool_remote_start.sample;hb=refs/heads/V4_4_STABLE">/etc/pgpool-II/sample_scripts/pgpool_remote_start.sample</ulink></entry>
+ <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/pgpool_remote_start.sample;hb=refs/heads/V4_5_STABLE">/etc/pgpool-II/sample_scripts/pgpool_remote_start.sample</ulink></entry>
<entry>Run after <xref linkend="GUC-RECOVERY-1ST-STAGE-COMMAND"> to start the Standby node</entry>
</row>
<row>
<entry morerows='1'>Watchdog</entry>
- <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/escalation.sh.sample;hb=refs/heads/V4_4_STABLE">/etc/pgpool-II/sample_scripts/escalation.sh.sample</ulink></entry>
- <entry>Run by <xref linkend="guc-wd-escalation-command"> to switch the Active/Standby Pgpool-II safely</entry>
+ <entry><ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/escalation.sh.sample;hb=refs/heads/V4_5_STABLE">/etc/pgpool-II/sample_scripts/escalation.sh.sample</ulink></entry>
+ <entry>Optional Configuration. Run by <xref linkend="guc-wd-escalation-command"> to switch the Leader/Standby Pgpool-II safely</entry>
</row>
</tbody>
</tgroup>
Install <productname>PostgreSQL</productname> from <productname>PostgreSQL</productname> YUM repository.
</para>
<programlisting>
-[all servers]# yum install -y https://download.postgresql.org/pub/repos/yum/reporpms/EL-7-x86_64/pgdg-redhat-repo-latest.noarch.rpm
-[all servers]# yum install -y postgresql15-server
+[all servers]# dnf install -y https://download.postgresql.org/pub/repos/yum/reporpms/EL-8-x86_64/pgdg-redhat-repo-latest.noarch.rpm
+[all servers]# dnf -qy module disable postgresql
+[all servers]# dnf install -y postgresql16-server
</programlisting>
<para>
Since <productname>Pgpool-II</productname> related packages are also included in <productname>PostgreSQL</productname> YUM repository,
...
exclude=pgpool*
+[pgdg16]
+...
+exclude=pgpool*
+
[pgdg15]
...
exclude=pgpool*
[pgdg11]
...
-exclude=pgpool*
-
-[pgdg10]
-...
exclude=pgpool*
</programlisting>
Install <productname>Pgpool-II</productname> from Pgpool-II YUM repository.
</para>
<programlisting>
-[all servers]# yum install -y https://www.pgpool.net/yum/rpms/4.4/redhat/rhel-7-x86_64/pgpool-II-release-4.4-1.noarch.rpm
-[all servers]# yum install -y pgpool-II-pg15-*
+[all servers]# dnf install -y https://www.pgpool.net/yum/rpms/4.5/redhat/rhel-8-x86_64/pgpool-II-release-4.5-1.noarch.rpm
+[all servers]# dnf install -y pgpool-II-pg16-*
</programlisting>
</sect2>
Before you start the configuration process, please check the following prerequisites.
</para>
- <itemizedlist>
- <listitem>
-
Set up <productname>PostgreSQL</productname> streaming replication on the primary server.
- In this example, we use WAL archiving.
- </para>
- First, we create the directory <filename>/var/lib/pgsql/archivedir</filename> to store
-
<acronym>WAL</acronym> segments on all servers. In this example, only Primary node archives
-
<acronym>WAL</acronym> locally.
- </para>
+ <sect3 id="example-cluster-before-starting-primary">
+ <title>Setting up streaming replication on Primary</title>
+ <para>
+ Set up <productname>PostgreSQL</productname> streaming replication on the primary server.
+ In this example, we use WAL archiving.
+ </para>
+ <para>
+ First, we create the directory <filename>/var/lib/pgsql/archivedir</filename> to store
+ <acronym>WAL</acronym> segments on all servers. In this example, only Primary node archives
+ <acronym>WAL</acronym> locally.
+ </para>
+ <programlisting>
[all servers]# su - postgres
[all servers]$ mkdir /var/lib/pgsql/archivedir
- </programlisting>
+ </programlisting>
-
Initialize <productname>PostgreSQL</productname> on the primary server.
- </para>
+ <para>
+ Initialize <productname>PostgreSQL</productname> on the primary server.
+ </para>
+ <programlisting>
[server1]# su - postgres
-[server1]$ /usr/pgsql-15/bin/initdb -D $PGDATA
+[server1]$ /usr/pgsql-16/bin/initdb -D $PGDATA
</programlisting>
<para>
wal_log_hints = on
</programlisting>
<para>
- We use the online recovery functionality of <productname>Pgpool-II</productname> to setup standby server after the primary server is started.
+ Start PostgreSQL primary server on <literal>server1</literal>.
</para>
- </listitem>
+ <programlisting>
+[server1]# su - postgres
+[server1]$ /usr/pgsql-16/bin/pg_ctl start -D $PGDATA
+ </programlisting>
+ </sect3>
- <listitem>
- <para>
- Because of the security reasons, we create a user <literal>repl</literal> solely used
- for replication purpose, and a user <literal>pgpool</literal> for streaming
- replication delay check and health check of <productname>Pgpool-II</productname>.
+ <sect3 id="example-cluster-before-starting-standby">
+ <title>Setting up streaming replication on Standby</title>
+ <para>
+ There are multiple methods to setup a standby server, such as:
+ <itemizedlist>
+ <listitem>
+ <para>
+ use pg_basebackup to backup the data directory of the primary from the standby.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ use <productname>Pgpool-II</productname>'s online recovery feature
+ (<xref linkend="runtime-online-recovery">) to automatically
+ setup a standby server.
+ </para>
+ </listitem>
+ </itemizedlist>
+ </para>
+ <para>
+ In this example, we use <productname>Pgpool-II</productname>'s
+ online recovery to setup the standby server
+ in section <xref linkend="example-cluster-verify-standby">
+ after the configuration of <productname>Pgpool-II</productname>
+ is completed.
+ </para>
+ </sect3>
+
+ <sect3 id="example-cluster-before-starting-users">
+ <title>Setting up PostgreSQL users</title>
+ <para>
+ Because of the security reasons, we create a dedicated user named <literal>repl</literal>
+ for replication and a dedicated user named <literal>pgpool</literal> for streaming
+ replication delay check and health check of <productname>Pgpool-II</productname>.
+ Because online recovery feature requires superuser privilege,
+ we use <literal>postgres</literal> user here.
</para>
<table id="example-cluster-user">
<tgroup cols="3">
<thead>
<row>
- <entry>User Name</entry>
- <entry>Password</entry>
- <entry>Detail</entry>
+ <entry>User Name</entry>
+ <entry>Password</entry>
+ <entry>Detail</entry>
</row>
</thead>
<tbody>
<row>
- <entry>repl</entry>
- <entry>repl</entry>
- <entry>PostgreSQL replication user</entry>
+ <entry>repl</entry>
+ <entry>repl</entry>
+ <entry>PostgreSQL replication user</entry>
</row>
<row>
- <entry>pgpool</entry>
- <entry>pgpool</entry>
- <entry>Pgpool-II health check (<xref linkend="GUC-HEALTH-CHECK-USER">) and replication delay check (<xref linkend="GUC-SR-CHECK-USER">) user</entry>
+ <entry>pgpool</entry>
+ <entry>pgpool</entry>
+ <entry>Pgpool-II health check (<xref linkend="GUC-HEALTH-CHECK-USER">) and replication delay check (<xref linkend="GUC-SR-CHECK-USER">) user</entry>
</row>
<row>
<entry>postgres</entry>
</programlisting>
<para>
- If you want to show "replication_state" and "replication_sync_state" column in
- <xref linkend="SQL-SHOW-POOL-NODES"> command result, role <literal>pgpool</literal>
- needs to be PostgreSQL super user or or in <literal>pg_monitor</literal> group
- (<productname>Pgpool-II</productname> 4.1 or later). Grant <literal>pg_monitor</literal>
- to <literal>pgpool</literal>:
+ To show <literal>replication_state</literal> and
+ <literal>replication_sync_state</literal> column in
+ <xref linkend="SQL-SHOW-POOL-NODES"> command result, role
+ <literal>pgpool</literal> needs to be PostgreSQL superuser or in
+ <literal>pg_monitor</literal> group (<productname>Pgpool-II</productname> 4.1 or later).
+ Grant <literal>pg_monitor</literal>
+ to <literal>pgpool</literal>:
</para>
<programlisting>
GRANT pg_monitor TO pgpool;
<note>
<para>
If you plan to use <xref linkend="guc-detach-false-primary">(<productname>Pgpool-II</productname> 4.0 or later),
- role "pgpool" needs to be <productname>PostgreSQL</productname> super user or
- or in "pg_monitor" group to use this feature.
+ role "pgpool" needs to be <productname>PostgreSQL</productname> superuser
+ or in <literal>pg_monitor</literal> group to use this feature.
</para>
</note>
<para>
enable <literal>scram-sha-256</literal> authentication method.
</para>
<programlisting>
-host all all samenet scram-sha-256
-host replication all samenet scram-sha-256
- </programlisting>
- </listitem>
-
- <listitem>
- <para>
- To use the automated failover and online recovery of <productname>Pgpool-II</productname>,
- it is requried to allow <emphasis>SSH public key authentication
- (passwordless SSH login)</emphasis> to all backend servers
- as <productname>Pgpool-II</productname> startup user
- (default is <literal>postgres</literal>. Pgpool-II 4.0 or before, default is <literal>root</literal>)
- and <productname>PostgreSQL</productname> startup user (default is <literal>postgres</literal>).
- </para>
- <para>
- First, set <literal>postgres</literal> user's passwrod.
- </para>
- <programlisting>
-[all servers]# passwd postgres
+host all pgpool samenet scram-sha-256
+host all postgres samenet scram-sha-256
+host replication repl samenet scram-sha-256
</programlisting>
- <para>
- Execute the following command on all servers
- to set up passwordless <literal>SSH</literal>.
- In this example, we assume that the generated key file name is
- <literal>id_rsa_pgpool</literal>.
- </para>
- <programlisting>
-[all servers]# mkdir ~/.ssh
-[all servers]# chmod 700 ~/.ssh
-[all servers]# cd ~/.ssh
-[all servers]# ssh-keygen -t rsa -f id_rsa_pgpool
-[all servers]# ssh-copy-id -i id_rsa_pgpool.pub postgres@server1
-[all servers]# ssh-copy-id -i id_rsa_pgpool.pub postgres@server2
-[all servers]# ssh-copy-id -i id_rsa_pgpool.pub postgres@server3
+ </sect3>
+ <sect3 id="example-cluster-before-starting-ssh">
+ <title>Setting up SSH public key authentication</title>
+ <para>
+ To use the automated failover and online recovery of <productname>Pgpool-II</productname>,
+ it is requried to configure <emphasis>SSH public key authentication
+ (passwordless SSH login)</emphasis> to all backend servers using
+ <literal>postgres</literal> user (the default user Pgpool-II is running as.
+ Pgpool-II 4.0 or before, the default user is <literal>root</literal>).
+ </para>
+ <para>
+ Execute the following command on all servers to generate a key pair using
+ the RSA algorithm. In this example, we assume that the generated key file
+ name is <literal>id_rsa_pgpool</literal>.
+ </para>
+ <programlisting>
[all servers]# su - postgres
[all servers]$ mkdir ~/.ssh
[all servers]$ chmod 700 ~/.ssh
[all servers]$ cd ~/.ssh
[all servers]$ ssh-keygen -t rsa -f id_rsa_pgpool
-[all servers]$ ssh-copy-id -i id_rsa_pgpool.pub postgres@server1
-[all servers]$ ssh-copy-id -i id_rsa_pgpool.pub postgres@server2
-[all servers]$ ssh-copy-id -i id_rsa_pgpool.pub postgres@server3
- </programlisting>
- <para>
- After setting SSH, make sure that you can run
- <command>ssh postgres@serverX -i ~/.ssh/id_rsa_pgpool</command> command
- as <productname>Pgpool-II</productname> startup user and
- <productname>PostgreSQL</productname> startup user to
- log in without entering a password.
- </para>
+ </programlisting>
+ <para>
+ Then add the public key <filename>id_rsa_pgpool.pub</filename> to
+ <filename>/var/lib/pgsql/.ssh/authorized_keys</filename> file
+ on each server.
+ </para>
+ <para>
+ After setting SSH, make sure that you can run
+ <command>ssh postgres@serverX -i ~/.ssh/id_rsa_pgpool</command> command
+ as <literal>postgres</literal> user to login to each server
+ without entering a password.
+ </para>
- <note>
- If you failed to login using SSH public key authentication, please check the following:
- <itemizedlist>
- <listitem>
- Ensure the public key authentication option <literal>PubkeyAuthentication</literal> are allowed in <filename>/etc/ssh/sshd_config</filename>:
- </para>
- </listitem>
- </itemizedlist>
+ <note>
+ <para>
+ If you failed to login using SSH public key authentication, please check the following:
+ <itemizedlist>
+ <listitem>
+ <para>
+ Ensure the public key authentication option <literal>PubkeyAuthentication</literal> are allowed in <filename>/etc/ssh/sshd_config</filename>:
+ </para>
+ </listitem>
+ </itemizedlist>
+ <programlisting>
PubkeyAuthentication yes
- </programlisting>
- <itemizedlist>
- <listitem>
+ </programlisting>
+ <itemizedlist>
+ <listitem>
+ <para>
if the password authentication is disabled, you can fail to run
<command>ssh-copy-id</command>, temporarily add the following
configuration in <filename>/etc/ssh/sshd_config</filename>.
- </para>
- </listitem>
- </itemizedlist>
+ </para>
+ </listitem>
+ </itemizedlist>
+ <programlisting>
PasswordAuthentication yes
- </programlisting>
- <itemizedlist>
- <listitem>
- If SELinux is enabled, SSH public key authentication (passwordless SSH) may fail.
- You need to run the following command on all servers.
- </para>
- </listitem>
- </itemizedlist>
+ </programlisting>
+ <itemizedlist>
+ <listitem>
+ <para>
+ If SELinux is enabled, SSH public key authentication (passwordless SSH) may fail.
+ You need to run the following command on all servers.
+ </para>
+ </listitem>
+ </itemizedlist>
+ <programlisting>
[all servers]# su - postgres
[all servers]$ restorecon -Rv ~/.ssh
- </programlisting>
- </para>
- </note>
- </listitem>
-
- <listitem>
- <para>
- To allow <literal>repl</literal> user without specifying password for streaming
- replication and online recovery, and execute <application>pg_rewind</application>
- using <literal>postgres</literal>, we create the <filename>.pgpass</filename> file
- in <literal>postgres</literal> user's home directory and change the permission to
- <literal>600</literal> on each <productname>PostgreSQL</productname> server.
+ </programlisting>
</para>
- <programlisting>
+ </note>
+ </sect3>
+
+ <sect3 id="example-cluster-before-starting-pgpass">
+ <title>Creating .pgpass</title>
+ <para>
+ To allow <literal>repl</literal> user without specifying password for streaming
+ replication and online recovery, and execute <application>pg_rewind</application>
+ using <literal>postgres</literal>, we
+ Create the <filename>.pgpass</filename> file in <literal>postgres</literal>
+ user's home directory and change the permission to <literal>600</literal>
+ on each <productname>PostgreSQL</productname> server.
+ This file allows <literal>repl</literal> user and <literal>postgres</literal>
+ user without providing a password for streaming replication and failover.
+ </para>
+ <programlisting>
[all servers]# su - postgres
[all servers]$ vi /var/lib/pgsql/.pgpass
server1:5432:replication:repl:<repl user password>
server1:5432:postgres:postgres:<postgres user password>
server2:5432:postgres:postgres:<postgres user password>
server3:5432:postgres:postgres:<postgres user password>
-[all servers]$ chmod 600 /var/lib/pgsql/.pgpass
- </programlisting>
- </listitem>
+[all servers]$ chmod 600 /var/lib/pgsql/.pgpass
+ </programlisting>
+ </sect3>
- <listitem>
- <para>
- When connect to <productname>Pgpool-II</productname> and <productname>PostgreSQL</productname> servers, the target port must be accessible by enabling firewall management softwares. Following is an example for <systemitem>CentOS/RHEL7</systemitem>.
- </para>
- <programlisting>
+ <sect3 id="example-cluster-before-starting-firewall">
+ <title>Setting up firewall</title>
+ <para>
+ When connect to <productname>Pgpool-II</productname> and <productname>PostgreSQL</productname> servers, the target port must be accessible by enabling firewall management softwares. Following is an example for <systemitem>Rocky Linux 8/RHEL 8</systemitem>.
+ </para>
+ <programlisting>
[all servers]# firewall-cmd --permanent --zone=public --add-service=postgresql
[all servers]# firewall-cmd --permanent --zone=public --add-port=9999/tcp --add-port=9898/tcp --add-port=9000/tcp --add-port=9694/udp
[all servers]# firewall-cmd --reload
- </programlisting>
- </listitem>
- </itemizedlist>
+ </programlisting>
+ </sect3>
</sect2>
<sect2 id="example-cluster-pgpool-node-id">
</itemizedlist>
</sect2>
+ <sect2 id="pcp-authentication">
+ <title>PCP connection authentication</title>
+ <para>
+ To use PCP commands PCP user names and md5 encrypted passwords must be
+ declared in <filename>pcp.conf</filename> in format
+ "<literal>username:encrypted password</literal>".
+ </para>
+ <para>
+ In this example, we set PCP username to "<literal>pgpool</literal>"
+ and password to "<literal>pgpool_password</literal>".
+ Use <xref linkend="PG-MD5"> to create the encrypted password entry for
+ <literal>pgpool</literal> user as below:
+ </para>
+ <programlisting>
+[all servers]# echo 'pgpool:'`pg_md5 pgpool_password` >> /etc/pgpool-II/pcp.conf
+
+[all servers]# cat /etc/pgpool-II/pcp.conf
+# USERID:MD5PASSWD
+pgpool:4aa0cb9673e84b06d4c8a848c80eb5d0
+ </programlisting>
+ </sect2>
+
<sect2 id="example-cluster-pgpool-config">
<title><productname>Pgpool-II</productname> Configuration</title>
<para>
<sect3 id="example-cluster-pgpool-config-listen-addresses">
<title>listen_addresses</title>
<para>
- To allow Pgpool-II to accept all incoming connections, we set <varname>listen_addresses = '*'</varname>.
+ To allow Pgpool-II and PCP to accept all incoming connections, set the following
+ parameters to <literal>'*'</literal>.
</para>
<programlisting>
listen_addresses = '*'
+pcp_listen_addresses = '*'
</programlisting>
</sect3>
backend_hostname0 = 'server1'
backend_port0 = 5432
backend_weight0 = 1
-backend_data_directory0 = '/var/lib/pgsql/15/data'
+backend_data_directory0 = '/var/lib/pgsql/16/data'
backend_flag0 = 'ALLOW_TO_FAILOVER'
backend_hostname1 = 'server2'
backend_port1 = 5432
backend_weight1 = 1
-backend_data_directory1 = '/var/lib/pgsql/15/data'
+backend_data_directory1 = '/var/lib/pgsql/16/data'
backend_flag1 = 'ALLOW_TO_FAILOVER'
backend_hostname2 = 'server3'
backend_port2 = 5432
backend_weight2 = 1
-backend_data_directory2 = '/var/lib/pgsql/15/data'
+backend_data_directory2 = '/var/lib/pgsql/16/data'
backend_flag2 = 'ALLOW_TO_FAILOVER'
</programlisting>
<para>
- To show "replication_state" and "replication_sync_state" column in <xref linkend="SQL-SHOW-POOL-NODES">
- command result, <xref linkend="GUC-BACKEND-APPLICATION-NAME"> parameter is required.
- Here we specify each backend's hostname in these parameters. (<productname>Pgpool-II</productname> 4.1 or later)
+ To show <literal>replication_state</literal> and
+ <literal>replication_sync_state</literal> column in
+ <xref linkend="SQL-SHOW-POOL-NODES"> command result,
+ <xref linkend="GUC-BACKEND-APPLICATION-NAME"> parameter is required.
+ Here we specify each backend's hostname in these parameters
+ (<productname>Pgpool-II</productname> 4.1 or later).
+ Make sure that the value set in <varname>backend_application_nameX</varname>
+ matches the value set in <varname>application_name</varname>
+ of <varname>primary_conninfo</varname>.
</para>
<programlisting>
...
<sect3 id="example-cluster-pgpool-config-failover">
<title>Failover configuration</title>
<para>
- Specify failover.sh script to be executed after failover in <varname>failover_command</varname>
- parameter.
- If we use 3 PostgreSQL servers, we need to specify follow_primary_command to run after failover on the primary node failover.
- In case of two PostgreSQL servers, follow_primary_command setting is not necessary.
+ Specify the script that will be executed when failover occurs in
+ <xref linkend="GUC-FAILOVER-COMMAND">. When using three or more
+ PostgreSQL servers, it's required to specify
+ <xref linkend="GUC-FOLLOW-PRIMARY-COMMAND"> to synchronize the standby
+ with the new primary. In case of two PostgreSQL servers, the setting of
+ <xref linkend="GUC-FOLLOW-PRIMARY-COMMAND"> is not required.
</para>
<para>
<productname>Pgpool-II</productname> replaces the following special characters with the backend specific
information while executing the scripts.
- See <xref linkend="GUC-FAILOVER-COMMAND"> for more details about each character.
+ See <xref linkend="GUC-FAILOVER-COMMAND"> and <xref linkend="GUC-FOLLOW-PRIMARY-COMMAND"> for more details about each character.
</para>
<programlisting>
failover_command = '/etc/pgpool-II/failover.sh %d %h %p %D %m %H %M %P %r %R %N %S'
-follow_primary_command = '/etc/pgpool-II/follow_primary.sh %d %h %p %D %m %H %M %P %r %R
+follow_primary_command = '/etc/pgpool-II/follow_primary.sh %d %h %p %D %m %H %M %P %r %R'
</programlisting>
<note>
<para>
</para>
</note>
<para>
- Sample scripts <ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/failover.sh.sample;hb=refs/heads/V4_4_STABLE">failover.sh</ulink>
- and <ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/follow_primary.sh.sample;hb=refs/heads/V4_4_STABLE">follow_primary.sh</ulink>
+ Sample scripts <ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/failover.sh.sample;hb=refs/heads/V4_5_STABLE">failover.sh</ulink>
+ and <ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/follow_primary.sh.sample;hb=refs/heads/V4_5_STABLE">follow_primary.sh</ulink>
are installed in <filename>/etc/pgpool-II/</filename>. Create failover scripts using these sample files.
</para>
<programlisting>
[all servers]# chown postgres:postgres /etc/pgpool-II/{failover.sh,follow_primary.sh}
</programlisting>
<para>
- Basically, it should work if you change <emphasis>PGHOME</emphasis> according to PostgreSQL installation directory.
+ Basically, it should work if you change <emphasis>PGHOME</emphasis> according to
+ PostgreSQL installation directory.
</para>
<programlisting>
[all servers]# vi /etc/pgpool-II/failover.sh
...
-PGHOME=/usr/pgsql-15
+PGHOME=/usr/pgsql-16
...
[all servers]# vi /etc/pgpool-II/follow_primary.sh
...
-PGHOME=/usr/pgsql-15
+PGHOME=/usr/pgsql-16
...
</programlisting>
<para>
- Since user authentication is required to use the <literal>PCP</literal> command in
- <varname>follow_primary_command</varname> script,
- we need to specify user name and md5 encrypted password in <filename>pcp.conf</filename>
- in format "<literal>username:encrypted password</literal>".
- </para>
- <para>
- if <literal>pgpool</literal> user is specified in <varname>PCP_USER</varname> in <filename>follow_primary.sh</filename>,
+ Make sure the entry of the PCP user specified in <varname>PCP_USER</varname> in
+ <filename>follow_primary.sh</filename> is created in <filename>pcp.conf</filename>.
+ In this example, we have created in <xref linkend="PCP-AUTHENTICATION">
</para>
<programlisting>
# cat /etc/pgpool-II/follow_primary.sh
...
</programlisting>
<para>
- then we use <xref linkend="PG-MD5"> to create the encrypted password entry for <literal>pgpool</literal> user as below:
+ Since <filename>follow_primary.sh</filename> script must execute PCP command without
+ entering a password, we need to create <filename>.pcppass</filename> in
+ <literal>postgres</literal> user's home directory on each server
+ (the home directory of the user Pgpool-II is running as).
+ The format of <filename>.pcppass</filename> is
+ "<literal>hostname:port:username:password</literal>".
</para>
- <programlisting>
-[all servers]# echo 'pgpool:'`pg_md5 PCP password` >> /etc/pgpool-II/pcp.conf
- </programlisting>
<para>
- Since <filename>follow_primary.sh</filename> script must execute PCP command without entering a
- password, we need to create <filename>.pcppass</filename> in the home directory of
- <productname>Pgpool-II</productname> startup user (postgres user) on each server.
+ In this example, we assume that the PCP user is <literal>pgpool</literal>
+ and the password is <literal>pgpool_password</literal>.
</para>
<programlisting>
[all servers]# su - postgres
-[all servers]$ echo 'localhost:9898:pgpool:<pgpool user password>' > ~/.pcppass
+[all servers]$ echo 'localhost:9898:pgpool:pgpool_password' > ~/.pcppass
[all servers]$ chmod 600 ~/.pcppass
</programlisting>
<note>
<sect3 id="example-cluster-pgpool-config-online-recovery">
<title>Pgpool-II Online Recovery Configurations</title>
<para>
- Next, in order to perform online recovery with <productname>Pgpool-II</productname> we specify
- the <productname>PostgreSQL</productname> user name and online recovery command
- <command>recovery_1st_stage</command>.
+ Next, configure the required parameters to perform online recovery.
Because <emphasis>Superuser</emphasis> privilege in <productname>PostgreSQL</productname>
- is required for performing online recovery, we specify <literal>postgres</literal> user in <xref linkend="GUC-RECOVERY-USER">.
- Then, we create <filename>recovery_1st_stage</filename> and <filename>pgpool_remote_start</filename>
- in database cluster directory of <productname>PostgreSQL</productname> primary server (server1), and add execute permission.
-
+ is required for performing online recovery, we specify <literal>postgres</literal>
+ user in <xref linkend="GUC-RECOVERY-USER">. In this example, we leave
+ <xref linkend="GUC-RECOVERY-PASSWORD"> empty, and create the entry in
+ <xref linkend="GUC-POOL-PASSWD">. See <xref linkend="example-cluster-pgpool-config-auth">
+ for how to create the entry in <xref linkend="GUC-POOL-PASSWD">.
</para>
<programlisting>
recovery_user = 'postgres'
recovery_1st_stage_command = 'recovery_1st_stage'
</programlisting>
<para>
- Online recovery sample scripts<ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/recovery_1st_stage.sample;hb=refs/heads/V4_4_STABLE">recovery_1st_stage</ulink>
- and <ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/pgpool_remote_start.sample;hb=refs/heads/V4_4_STABLE">pgpool_remote_start</ulink>
+ Then, we create <filename>recovery_1st_stage</filename> and <filename>pgpool_remote_start</filename>
+ in database cluster directory of <productname>PostgreSQL</productname>
+ primary server (server1).
+ </para>
+ <para>
+ The sample scripts of online recovery <ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/recovery_1st_stage.sample;hb=refs/heads/V4_5_STABLE">recovery_1st_stage</ulink>
+ and <ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/pgpool_remote_start.sample;hb=refs/heads/V4_5_STABLE">pgpool_remote_start</ulink>
are installed in <filename>/etc/pgpool-II/</filename>. Copy these files to the data directory of the primary server (server1).
</para>
<programlisting>
-[server1]# cp -p /etc/pgpool-II/sample_scripts/recovery_1st_stage.sample /var/lib/pgsql/15/data/recovery_1st_stage
-[server1]# cp -p /etc/pgpool-II/sample_scripts/pgpool_remote_start.sample /var/lib/pgsql/15/data/pgpool_remote_start
-[server1]# chown postgres:postgres /var/lib/pgsql/15/data/{recovery_1st_stage,pgpool_remote_start}
+[server1]# cp -p /etc/pgpool-II/sample_scripts/recovery_1st_stage.sample /var/lib/pgsql/16/data/recovery_1st_stage
+[server1]# cp -p /etc/pgpool-II/sample_scripts/pgpool_remote_start.sample /var/lib/pgsql/16/data/pgpool_remote_start
+[server1]# chown postgres:postgres /var/lib/pgsql/16/data/{recovery_1st_stage,pgpool_remote_start}
</programlisting>
<para>
Basically, it should work if you change <emphasis>PGHOME</emphasis> according to PostgreSQL installation directory.
</para>
<programlisting>
-[server1]# vi /var/lib/pgsql/15/data/recovery_1st_stage
+[server1]# vi /var/lib/pgsql/16/data/recovery_1st_stage
...
-PGHOME=/usr/pgsql-15
+PGHOME=/usr/pgsql-16
...
-[server1]# vi /var/lib/pgsql/15/data/pgpool_remote_start
+[server1]# vi /var/lib/pgsql/16/data/pgpool_remote_start
...
-PGHOME=/usr/pgsql-15
+PGHOME=/usr/pgsql-16
...
</programlisting>
enable_pool_hba = on
</programlisting>
<para>
- The format of <filename>pool_hba.conf</filename> file follows very closely PostgreSQL's
- <filename>pg_hba.conf</filename> format. Set <literal>pgpool</literal> and <literal>postgres</literal> user's authentication method to <literal>scram-sha-256</literal>.
+ The format of <filename>pool_hba.conf</filename> file follows
+ PostgreSQL's <filename>pg_hba.conf</filename> format very closely.
+ Set <literal>pgpool</literal> and <literal>postgres</literal> user's
+ authentication method to <literal>scram-sha-256</literal>. In this example,
+ it is assumed that the application connecting to <productname>Pgpool-II</productname>
+ is in the same subnet.
</para>
<programlisting>
-host all pgpool 0.0.0.0/0 scram-sha-256
-host all postgres 0.0.0.0/0 scram-sha-256
+host all pgpool samenet scram-sha-256
+host all postgres samenet scram-sha-256
</programlisting>
<note>
<para>
</note>
<para>
The default password file name for authentication is <xref linkend="GUC-POOL-PASSWD">.
- To use <literal>scram-sha-256</literal> authentication, the decryption key to decrypt the passwords
- is required. We create the <literal>.pgpoolkey</literal> file in <productname>Pgpool-II</productname>
- start user <literal>postgres</literal>'s (<productname>Pgpool-II</productname> 4.1 or later) home directory.
- (<productname>Pgpool-II</productname> 4.0 or before, by default <productname>Pgpool-II</productname>
- is started as <literal>root</literal>)
+ To use <literal>scram-sha-256</literal> authentication, the decryption key to
+ decrypt the passwords is required. We create the <literal>.pgpoolkey</literal>
+ file in <literal>postgres</literal> user's home directory
+ (the user Pgpool-II is running as. <productname>Pgpool-II</productname> 4.0 or before,
+ <productname>Pgpool-II</productname> is running as <literal>root</literal> by default)
<programlisting>
[all servers]# su - postgres
[all servers]$ echo 'some string' > ~/.pgpoolkey
use_watchdog = on
</programlisting>
<para>
- Specify virtual IP address that accepts connections from clients on
- <literal>server1</literal>, <literal>server2</literal>, <literal>server3</literal>.
- Ensure that the IP address set to virtual IP isn't used yet.
+ Set virtual IP address to <xref linkend="GUC-DELEGATE-IP">.
+ Ensure that the IP address isn't used yet.
</para>
<programlisting>
-delegate_ip = '192.168.137.150'
+delegate_ip = '192.168.100.50'
</programlisting>
<para>
The network interface used in this example is "enp0s8".
Since root privilege is required to execute <varname>if_up/down_cmd</varname> or
<varname>arping_cmd</varname> command, use setuid on these command or allow
- <productname>Pgpool-II</productname> startup user, <literal>postgres</literal> user (Pgpool-II 4.1 or later) to run <command>sudo</command> command without a password.
+ <literal>postgres</literal> user (the user Pgpool-II is running as) to run
+ <command>sudo</command> command without a password.
</para>
<note>
<para>
<note>
<para>
If "Defaults requiretty" is set in the <filename>/etc/sudoers</filename>,
- please ensure that the <productname>pgpool</productname> startup user can execute the <command>if_up_cmd</command>, <command>if_down_cmd</command> and <command>arping_cmd</command> command without a tty.
+ please ensure that the user that <productname>Pgpool-II</productname> is running as can execute the <command>if_up_cmd</command>, <command>if_down_cmd</command> and <command>arping_cmd</command> command without a tty.
</para>
</note>
<para>
pgpool_port2 = 9999
</programlisting>
<para>
- Specify the method of lifecheck <xref linkend="guc-wd-lifecheck-method">
+ Configure the method of lifecheck <xref linkend="guc-wd-lifecheck-method">
and the lifecheck interval <xref linkend="guc-wd-interval">.
Here, we use <literal>heartbeat</literal> method to perform watchdog lifecheck.
</para>
heartbeat_device2 = ''
</programlisting>
<para>
- If the <xref linkend="guc-wd-lifecheck-method"> is set to <literal>heartbeat</literal>,
+ If <xref linkend="guc-wd-lifecheck-method"> is set to <literal>heartbeat</literal>,
specify the time to detect a fault <xref linkend="guc-wd-heartbeat-deadtime"> and
the interval to send heartbeat signals <xref linkend="guc-wd-heartbeat-deadtime">.
</para>
</programlisting>
<para>
- When <literal>Watchdog</literal> process is abnormally terminated, the virtual IP may be "up" on both of the old and new active pgpool nodes.
- To prevent this, configure <xref linkend="guc-wd-escalation-command"> to bring down the virtual IP on other pgpool nodes before bringing up the virtual IP on the new active pgpool node.
+ This setting is optional.
+ When <literal>Watchdog</literal> process is abnormally terminated,
+ the virtual IP may be "up" on both of the old and new active pgpool nodes.
+ To prevent this, configure <xref linkend="guc-wd-escalation-command">
+ to bring down the virtual IP on other Pgpool-II nodes before
+ bringing up the virtual IP on the new leader Pgpool-II node.
</para>
<programlisting>
wd_escalation_command = '/etc/pgpool-II/escalation.sh'
</programlisting>
<para>
- The sample script <ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/escalation.sh.sample;hb=refs/heads/V4_4_STABLE">escalation.sh</ulink> is installed in <filename>/etc/pgpool-II/</filename>.
+ The sample script <ulink url="https://git.postgresql.org/gitweb/?p=pgpool2.git;a=blob_plain;f=src/sample/scripts/escalation.sh.sample;hb=refs/heads/V4_5_STABLE">escalation.sh</ulink> is installed in <filename>/etc/pgpool-II/</filename>.
</para>
<programlisting>
[all servers]# cp -p /etc/pgpool-II/sample_scripts/escalation.sh.sample /etc/pgpool-II/escalation.sh
<para>
Basically, it should work if you change the following variables according to your environment.
- PGPOOL is tha array of the hostname that running Pgpool-II.
- VIP is the virtual IP address that you set as delegate_ip.
+ PGPOOLS is a list of hostnames where Pgpool-II is running.
+ VIP is the virtual IP address that is set to <xref linkend="guc-delegate-ip">.
DEVICE is the network interface for the virtual IP.
</para>
<programlisting>
[all servers]# vi /etc/pgpool-II/escalation.sh
...
PGPOOLS=(server1 server2 server3)
-VIP=192.168.137.150
+VIP=192.168.100.50
DEVICE=enp0s8
...
</programlisting>
</note>
<note>
<para>
- If use_watchdog = on, please make sure the pgpool node number is specified
- in <filename>pgpool_node_id</filename> file.
+ If <varname>use_watchdog = on</varname>, please make sure the pgpool
+ node number is specified in <filename>pgpool_node_id</filename> file.
See <xref linkend="example-cluster-pgpool-node-id"> for details.
</para>
</note>
</programlisting>
<para>
- The configuration of <filename>pgpool.conf</filename> on server1 is completed. Copy the <filename>pgpool.conf</filename>
- to other <productname>Pgpool-II</productname> nodes (server2 and server3).
+ The configuration of <filename>pgpool.conf</filename> on server1 is completed.
+ Copy the <filename>pgpool.conf</filename> to other
+ <productname>Pgpool-II</productname> nodes (server2 and server3).
</para>
<programlisting>
[server1]# scp -p /etc/pgpool-II/pgpool.conf root@server2:/etc/pgpool-II/pgpool.conf
</sect3>
</sect2>
- <sect2 id="example-cluster-start-stop">
- <title>Starting/Stopping Pgpool-II</title>
- <para>
- Next we start <productname>Pgpool-II</productname>. Before starting
- <productname>Pgpool-II</productname>, please start
- <productname>PostgreSQL</productname> servers first.
- Also, when stopping <productname>PostgreSQL</productname>, it is necessary to
- stop Pgpool-II first.
- </para>
- <itemizedlist>
- <listitem>
- <para>
- Starting <productname>Pgpool-II</productname>
- </para>
- <para>
- In section <link linkend="EXAMPLE-CLUSTER-PRE-SETUP">Before Starting</link>,
- we already set the auto-start of <productname>Pgpool-II</productname>. To start
- <productname>Pgpool-II</productname>, restart the whole system or execute the following command.
- </para>
- <programlisting>
-# systemctl start pgpool.service
- </programlisting>
- </listitem>
- <listitem>
- <para>
- Stopping <productname>Pgpool-II</productname>
- </para>
- <programlisting>
-# systemctl stop pgpool.service
- </programlisting>
- </listitem>
- </itemizedlist>
- </sect2>
-
<sect2 id="example-cluster-verify">
<title>How to use</title>
<para>
Let's start to use <productname>Pgpool-II</productname>.
- First, we start the primary <productname>PostgreSQL</productname>.
</para>
- <programlisting>
+ <sect3 id="example-cluster-verify-starting-stopping">
+ <title>Starting/Stopping Pgpool-II</title>
+ <itemizedlist>
+ <listitem>
+ <para>
+ Starting <productname>Pgpool-II</productname>
+ </para>
+ <para>
+ First, let's start <productname>Pgpool-II</productname>.
+ </para>
+ <para>
+ Before starting <productname>Pgpool-II</productname>,
+ the PostgreSQL primary server must be already running.
+ If PostgreSQL primary server is not running, start it first
+ using the following command.
+ </para>
+ <programlisting>
[server1]# su - postgres
-[server1]$ /usr/pgsql-15/bin/pg_ctl start -D $PGDATA
- </programlisting>
- <para>
- Then let's start <productname>Pgpool-II</productname> on <literal>server1</literal>,
- <literal>server2</literal>, <literal>server3</literal> by using the following command.
- </para>
- <programlisting>
-# systemctl start pgpool.service
- </programlisting>
+[server1]$ /usr/pgsql-16/bin/pg_ctl start -D $PGDATA
+ </programlisting>
+ <para>
+ Start <productname>Pgpool-II</productname> on <literal>server1</literal>,
+ <literal>server2</literal>, <literal>server3</literal> by using the following command.
+ </para>
+ <programlisting>
+[all servers]# systemctl start pgpool.service
+ </programlisting>
+ </listitem>
+ <listitem>
+ <para>
+ Stopping <productname>Pgpool-II</productname>
+ </para>
+ <para>
+ When stopping <productname>PostgreSQL</productname>,
+ <productname>Pgpool-II</productname> must be stopped first.
+ </para>
+ <programlisting>
+[all servers]# systemctl stop pgpool.service
+ </programlisting>
+ </listitem>
+ </itemizedlist>
+ </sect3>
<sect3 id="example-cluster-verify-standby">
- <title>Set up PostgreSQL standby server</title>
+ <title>Setting up PostgreSQL standby server</title>
<para>
First, we should set up <productname>PostgreSQL</productname> standby server by
- using <productname>Pgpool-II</productname> online recovery functionality. Ensure
- that <filename>recovery_1st_stage</filename> and <filename>pgpool_remote_start</filename>
- scripts used by <command>pcp_recovery_node</command> command are in database
- cluster directory of <productname>PostgreSQL</productname> primary server (<literal>server1</literal>).
+ using <productname>Pgpool-II</productname> online recovery functionality.
+ </para>
+ <para>
+ Connect to Pgpool-II via virtual IP to check the status of backend nodes.
+ As shown in the result, primary server is running on <literal>server1</literal>,
+ standby servers on <literal>server2</literal> and <literal>server3</literal>
+ are in "down" status.
</para>
<programlisting>
-# pcp_recovery_node -h 192.168.137.150 -p 9898 -U pgpool -n 1
+[any server]# psql -h 192.168.100.50 -p 9999 -U pgpool postgres -c "show pool_nodes"
+Password for user pgpool:
+ node_id | hostname | port | status | pg_status | lb_weight | role | pg_role | select_cnt | load_balance_node | replication_delay | replication_state | replication_sync_state | last_status_change
+---------+----------+------+--------+-----------+-----------+---------+---------+------------+-------------------+-------------------+-------------------+------------------------+---------------------
+ 0 | server1 | 5432 | up | up | 0.333333 | primary | primary | 0 | true | 0 | | | 2023-11-10 15:30:14
+ 1 | server2 | 5432 | down | down | 0.333333 | standby | unknown | 0 | false | 0 | | | 2023-11-10 15:30:14
+ 2 | server3 | 5432 | down | down | 0.333333 | standby | unknown | 0 | false | 0 | | | 2023-11-10 15:30:14
+(3 rows)
+ </programlisting>
+ <para>
+ Before running <xref linkend="pcp-recovery-node"> command,
+ ensure that <filename>recovery_1st_stage</filename> and
+ <filename>pgpool_remote_start</filename> scripts exist in the
+ data directory of <productname>PostgreSQL</productname>
+ primary server (<literal>server1</literal>).
+ </para>
+ <programlisting>
+[any server]# pcp_recovery_node -h 192.168.100.50 -p 9898 -U pgpool -n 1 -W
Password:
pcp_recovery_node -- Command Successful
-# pcp_recovery_node -h 192.168.137.150 -p 9898 -U pgpool -n 2
+[any server]# pcp_recovery_node -h 192.168.100.50 -p 9898 -U pgpool -n 2 -W
Password:
pcp_recovery_node -- Command Successful
</programlisting>
<para>
After executing <command>pcp_recovery_node</command> command,
- verify that <literal>server2</literal> and <literal>server3</literal>
- are started as <productname>PostgreSQL</productname> standby server.
+ verify that <productname>PostgreSQL</productname> standby servers
+ are running on <literal>server2</literal> and <literal>server3</literal>.
</para>
<programlisting>
-# psql -h 192.168.137.150 -p 9999 -U pgpool postgres -c "show pool_nodes"
+[any server]# psql -h 192.168.100.50 -p 9999 -U pgpool postgres -c "show pool_nodes"
Password for user pgpool
node_id | hostname | port | status | pg_status | lb_weight | role | pg_role | select_cnt | load_balance_node | replication_delay | replication_state | replication_sync_state | last_status_change
---------+----------+------+--------+-----------+-----------+---------+---------+------------+-------------------+-------------------+-------------------+------------------------+---------------------
- 0 | server1 | 5432 | up | up | 0.333333 | primary | primary | 0 | false | 0 | | | 2021-10-19 07:00:57
- 1 | server2 | 5432 | up | up | 0.333333 | standby | standby | 0 | true | 0 | streaming | async | 2021-10-19 07:00:57
- 2 | server3 | 5432 | up | up | 0.333333 | standby | standby | 0 | false | 0 | streaming | async | 2021-10-19 07:00:57
+ 0 | server1 | 5432 | up | up | 0.333333 | primary | primary | 0 | false | 0 | | | 2023-11-10 15:30:14
+ 1 | server2 | 5432 | up | up | 0.333333 | standby | standby | 0 | true | 0 | streaming | async | 2023-11-10 16:32:33
+ 2 | server3 | 5432 | up | up | 0.333333 | standby | standby | 0 | false | 0 | streaming | async | 2023-11-10 16:33:08
(3 rows)
</programlisting>
</sect3>
<sect3 id="example-cluster-verify-watchdog">
- <title>Switching active/standby watchdog</title>
+ <title>Switching leader/standby watchdog</title>
<para>
- Confirm the watchdog status by using <command>pcp_watchdog_info</command>. The <command>Pgpool-II</command> server which is started first run as <literal>LEADER</literal>.
+ Confirm the watchdog status by using <command>pcp_watchdog_info</command>. The <command>Pgpool-II</command> server which is started first runs as <literal>LEADER</literal>.
</para>
<programlisting>
-# pcp_watchdog_info -h 192.168.137.150 -p 9898 -U pgpool
+[any server]# pcp_watchdog_info -h 192.168.100.50 -p 9898 -U pgpool -W
Password:
3 3 YES server1:9999 Linux server1 server1
-server1:9999 Linux server1 server1 9999 9000 4 LEADER 0 MEMBER #The Pgpool-II server started first became "LEADER".
-server2:9999 Linux server2 server2 9999 9000 7 STANDBY 0 MEMBER #run as standby
-server3:9999 Linux server3 server3 9999 9000 7 STANDBY 0 MEMBER #run as standby
+server1:9999 Linux server1 server1 9999 9000 4 LEADER 0 MEMBER # The Pgpool-II server started first becames "LEADER".
+server2:9999 Linux server2 server2 9999 9000 7 STANDBY 0 MEMBER # running as STANDBY
+server3:9999 Linux server3 server3 9999 9000 7 STANDBY 0 MEMBER # running as STANDBY
</programlisting>
<para>
- Stop active server <literal>server1</literal>, then <literal>server2</literal> or
- <literal>server3</literal> will be promoted to active server. To stop
- <literal>server1</literal>, we can stop <productname>Pgpool-II</productname>
- service or shutdown the whole system. Here, we stop <productname>Pgpool-II</productname> service.
+ If the <literal>LEADER</literal> <productname>Pgpool-II</productname>
+ on <literal>server1</literal> goes down, standby
+ <productname>Pgpool-II</productname> on <literal>server2</literal> or
+ <literal>server3</literal> will become the new <literal>LEADER</literal>.
+ </para>
+ <para>
+ To verify this behavior, you may stop <productname>Pgpool-II</productname>
+ service or shutdown the whole system. Here, we stop
+ <productname>Pgpool-II</productname> service.
</para>
<programlisting>
[server1]# systemctl stop pgpool.service
-# pcp_watchdog_info -p 9898 -h 192.168.137.150 -U pgpool
+[server1]# pcp_watchdog_info -p 9898 -h 192.168.100.50 -U pgpool -W
Password:
3 3 YES server2:9999 Linux server2 server2
-server2:9999 Linux server2 server2 9999 9000 4 LEADER 0 MEMBER #server2 is promoted to LEADER
-server1:9999 Linux server1 server1 9999 9000 10 SHUTDOWN 0 MEMBER #server1 is stopped
-server3:9999 Linux server3 server3 9999 9000 7 STANDBY 0 MEMBER #server3 runs as STANDBY
+server2:9999 Linux server2 server2 9999 9000 4 LEADER 0 MEMBER # server2 becomes LEADER
+server1:9999 Linux server1 server1 9999 9000 10 SHUTDOWN 0 MEMBER # server1 is stopped
+server3:9999 Linux server3 server3 9999 9000 7 STANDBY 0 MEMBER # server3 is running as a STANDBY
</programlisting>
<para>
- Start <productname>Pgpool-II</productname> (<literal>server1</literal>) which we have stopped again,
- and verify that <literal>server1</literal> runs as a standby.
+ Restart the stopped <productname>Pgpool-II</productname>
+ on <literal>server1</literal> and verify that it is running
+ as a <literal>STANDBY</literal>.
</para>
<programlisting>
[server1]# systemctl start pgpool.service
-[server1]# pcp_watchdog_info -p 9898 -h 192.168.137.150 -U pgpool
+[server1]# pcp_watchdog_info -p 9898 -h 192.168.100.50 -U pgpool -W
Password:
3 3 YES server2:9999 Linux server2 server2
<sect3 id="example-cluster-verify-failover">
<title>Failover</title>
<para>
- First, use <command>psql</command> to connect to <productname>PostgreSQL</productname> via virtual IP,
+ First, use <command>psql</command> to connect to
+ <productname>PostgreSQL</productname> via virtual IP,
and verify the backend information.
</para>
<programlisting>
-# psql -h 192.168.137.150 -p 9999 -U pgpool postgres -c "show pool_nodes"
+# psql -h 192.168.100.50 -p 9999 -U pgpool postgres -c "show pool_nodes"
Password for user pgpool:
node_id | hostname | port | status | pg_status | lb_weight | role | pg_role | select_cnt | load_balance_node | replication_delay | replication_state | replication_sync_state | last_status_change
---------+----------+------+--------+-----------+-----------+---------+---------+------------+-------------------+-------------------+-------------------+------------------------+---------------------
- 0 | server1 | 5432 | up | up | 0.333333 | primary | primary | 0 | false | 0 | | | 2021-10-19 07:08:14
- 1 | server2 | 5432 | up | up | 0.333333 | standby | standby | 0 | false | 0 | streaming | async | 2021-10-19 07:08:14
- 2 | server3 | 5432 | up | up | 0.333333 | standby | standby | 0 | true | 0 | streaming | async | 2021-10-19 07:08:14
+ 0 | server1 | 5432 | up | up | 0.333333 | primary | primary | 0 | false | 0 | | | 2023-11-10 15:30:14
+ 1 | server2 | 5432 | up | up | 0.333333 | standby | standby | 0 | false | 0 | streaming | async | 2023-11-10 16:32:33
+ 2 | server3 | 5432 | up | up | 0.333333 | standby | standby | 0 | true | 0 | streaming | async | 2023-11-10 16:33:08
(3 rows)
</programlisting>
<para>
- Next, stop primary <productname>PostgreSQL</productname> server
- <literal>server1</literal>, and verify automatic failover.
+ Next, stop the primary <productname>PostgreSQL</productname> server
+ on <literal>server1</literal> and verify that failover is performed
+ automatically.
</para>
<programlisting>
-[server1]$ pg_ctl -D /var/lib/pgsql/15/data -m immediate stop
+[server1]$ pg_ctl -D /var/lib/pgsql/16/data -m immediate stop
</programlisting>
<para>
- After stopping <productname>PostgreSQL</productname> on <literal>server1</literal>,
- failover occurs and <productname>PostgreSQL</productname> on
- <literal>server2</literal> becomes new primary DB.
+ After stopping <productname>PostgreSQL</productname> on
+ <literal>server1</literal>, failover occurs.
+ <productname>PostgreSQL</productname> on
+ <literal>server2</literal> becomes the new primary and
+ the standby server on <literal>server3</literal>
+ is configured as a standby of the new primary.
</para>
<programlisting>
-# psql -h 192.168.137.150 -p 9999 -U pgpool postgres -c "show pool_nodes"
+# psql -h 192.168.100.50 -p 9999 -U pgpool postgres -c "show pool_nodes"
Password for user pgpool:
node_id | hostname | port | status | pg_status | lb_weight | role | pg_role | select_cnt | load_balance_node | replication_delay | replication_state | replication_sync_state | last_status_change
---------+----------+------+--------+-----------+-----------+---------+---------+------------+-------------------+-------------------+-------------------+------------------------+---------------------
- 0 | server1 | 5432 | down | down | 0.333333 | standby | unknown | 0 | false | 0 | | | 2021-10-19 07:10:01
- 1 | server2 | 5432 | up | up | 0.333333 | primary | primary | 0 | false | 0 | | | 2021-10-19 07:10:01
- 2 | server3 | 5432 | up | up | 0.333333 | standby | standby | 0 | true | 0 | streaming | async | 2021-10-19 07:10:03
+ 0 | server1 | 5432 | down | down | 0.333333 | standby | unknown | 0 | false | 0 | | | 2023-11-10 17:05:40
+ 1 | server2 | 5432 | up | up | 0.333333 | primary | primary | 0 | false | 0 | | | 2023-11-10 17:05:40
+ 2 | server3 | 5432 | up | up | 0.333333 | standby | standby | 0 | true | 0 | streaming | async | 2023-11-10 17:05:51
(3 rows)
</programlisting>
<para>
- <literal>server3</literal> is running as standby of new primary <literal>server2</literal>.
+ <literal>server3</literal> is running as a standby of new primary <literal>server2</literal>.
</para>
<programlisting>
usesysid | 16385
usename | repl
application_name | server3
-client_addr | 192.168.137.103
+client_addr | 192.168.100.53
client_hostname |
client_port | 40916
-backend_start | 2021-10-19 07:10:03.067241+00
+backend_start | 2023-11-10 17:10:03.067241+00
backend_xmin |
state | streaming
sent_lsn | 0/12000260
replay_lag |
sync_priority | 0
sync_state | async
-reply_time | 2021-10-19 07:11:53.886477+00
+reply_time | 2023-11-10 17:17:23.886477+00
</programlisting>
</sect3>
<sect3 id="example-cluster-verify-online-recovery">
<title>Online Recovery</title>
<para>
- Here, we use <productname>Pgpool-II</productname> online recovery functionality to
- restore <literal>server1</literal> (old primary server) as a standby. Before
- restoring the old primary server, please ensure that
- <filename>recovery_1st_stage</filename> and <filename>pgpool_remote_start</filename> scripts
- exist in database cluster directory of current primary server <literal>server2</literal>.
+ Here, we use <productname>Pgpool-II</productname> online recovery
+ feature to restore the former primary on <literal>server1</literal>
+ as a standby.
</para>
<programlisting>
-# pcp_recovery_node -h 192.168.137.150 -p 9898 -U pgpool -n 0
+[any server]# pcp_recovery_node -h 192.168.100.50 -p 9898 -U pgpool -n 0 -W
Password:
pcp_recovery_node -- Command Successful
</programlisting>
<para>
- Then verify that <literal>server1</literal> is started as a standby.
+ Then verify that PostgreSQL on <literal>server1</literal> is
+ running as a standby.
</para>
<programlisting>
-# psql -h 192.168.137.150 -p 9999 -U pgpool postgres -c "show pool_nodes"
+[any server]# psql -h 192.168.100.50 -p 9999 -U pgpool postgres -c "show pool_nodes"
Password for user pgpool:
node_id | hostname | port | status | lb_weight | role | select_cnt | load_balance_node | replication_delay | replication_state | replication_sync_state | last_status_change
---------+----------+------+--------+-----------+-----------+---------+---------+------------+-------------------+-------------------+-------------------+------------------------+---------------------
- 0 | server1 | 5432 | up | up | 0.333333 | standby | standby | 0 | true | 0 | streaming | async | 2021-10-19 07:14:06
- 1 | server2 | 5432 | up | up | 0.333333 | primary | primary | 0 | false | 0 | | | 2021-10-19 07:10:01
- 2 | server3 | 5432 | up | up | 0.333333 | standby | standby | 0 | false | 0 | streaming | async | 2021-10-19 07:10:03
+ 0 | server1 | 5432 | up | up | 0.333333 | standby | standby | 0 | true | 0 | streaming | async | 2023-11-10 17:22:03
+ 1 | server2 | 5432 | up | up | 0.333333 | primary | primary | 0 | false | 0 | | | 2023-11-10 17:05:40
+ 2 | server3 | 5432 | up | up | 0.333333 | standby | standby | 0 | false | 0 | streaming | async | 2023-11-10 17:05:51
(3 rows)
</programlisting>
</sect3>
projects / pgpool2.git / commitdiff