pgBackRest User Guide - RHEL & CentOS 7 / PostgreSQL 9.6

Introduction

This user guide is intended to be followed sequentially from beginning to end — each section depends on the last. For example, the Backup section relies on setup that is performed in the Quick Start section. Once pgBackRest is up and running then skipping around is possible but following the user guide in order is recommended the first time through.

Although the examples are targeted at RHEL/CentOS 7 and PostgreSQL 9.6, it should be fairly easy to apply this guide to any Unix distribution and PostgreSQL version. The only OS-specific commands are those to create, start, stop, and drop PostgreSQL clusters. The pgBackRest commands will be the same on any Unix system though the locations to install Perl libraries and executables may vary.

Configuration information and documentation for PostgreSQL can be found in the PostgreSQL Manual.

A somewhat novel approach is taken to documentation in this user guide. Each command is run on a virtual machine when the documentation is built from the XML source. This means you can have a high confidence that the commands work correctly in the order presented. Output is captured and displayed below the command when appropriate. If the output is not included it is because it was deemed not relevant or was considered a distraction from the narrative.

All commands are intended to be run as an unprivileged user that has sudo privileges for both the root and postgres users. It's also possible to run the commands directly as their respective users without modification and in that case the sudo commands can be stripped off.

Concepts

The following concepts are defined as they are relevant to pgBackRest, PostgreSQL, and this user guide.

2.1

Backup

A backup is a consistent copy of a database cluster that can be restored to recover from a hardware failure, to perform Point-In-Time Recovery, or to bring up a new standby.

Full Backup: pgBackRest copies the entire contents of the database cluster to the backup. The first backup of the database cluster is always a Full Backup. pgBackRest is always able to restore a full backup directly. The full backup does not depend on any files outside of the full backup for consistency.

Differential Backup: pgBackRest copies only those database cluster files that have changed since the last full backup. pgBackRest restores a differential backup by copying all of the files in the chosen differential backup and the appropriate unchanged files from the previous full backup. The advantage of a differential backup is that it requires less disk space than a full backup, however, the differential backup and the full backup must both be valid to restore the differential backup.

Incremental Backup: pgBackRest copies only those database cluster files that have changed since the last backup (which can be another incremental backup, a differential backup, or a full backup). As an incremental backup only includes those files changed since the prior backup, they are generally much smaller than full or differential backups. As with the differential backup, the incremental backup depends on other backups to be valid to restore the incremental backup. Since the incremental backup includes only those files since the last backup, all prior incremental backups back to the prior differential, the prior differential backup, and the prior full backup must all be valid to perform a restore of the incremental backup. If no differential backup exists then all prior incremental backups back to the prior full backup, which must exist, and the full backup itself must be valid to restore the incremental backup.

2.2

Restore

A restore is the act of copying a backup to a system where it will be started as a live database cluster. A restore requires the backup files and one or more WAL segments in order to work correctly.

2.3

Write Ahead Log (WAL)

WAL is the mechanism that PostgreSQL uses to ensure that no committed changes are lost. Transactions are written sequentially to the WAL and a transaction is considered to be committed when those writes are flushed to disk. Afterwards, a background process writes the changes into the main database cluster files (also known as the heap). In the event of a crash, the WAL is replayed to make the database consistent.

WAL is conceptually infinite but in practice is broken up into individual 16MB files called segments. WAL segments follow the naming convention 0000000100000A1E000000FE where the first 8 hexadecimal digits represent the timeline and the next 16 digits are the logical sequence number (LSN).

2.4

Encryption

Encryption is the process of converting data into a format that is unrecognizable unless the appropriate password (also referred to as passphrase) is provided.

pgBackRest will encrypt the repository based on a user-provided password, thereby preventing unauthorized access to data stored within the repository.

Upgrading pgBackRest

3.1

Upgrading pgBackRest from v1 to v2

Upgrading from v1 to v2 is fairly straight-forward. The repository format has not changed and all non-deprecated options from v1 are accepted, so for most installations it is simply a matter of installing the new version.

However, there are a few caveats:

The deprecated thread-max option is no longer valid. Use process-max instead.
The deprecated archive-max-mb option is no longer valid. This has been replaced with the archive-push-queue-max option which has different semantics.
The default for the backup-user option has changed from backrest to pgbackrest.
In v2.02 the default location of the pgBackRest configuration file has changed from /etc/pgbackrest.conf to /etc/pgbackrest/pgbackrest.conf. If /etc/pgbackrest/pgbackrest.conf does not exist, the /etc/pgbackrest.conf file will be loaded instead, if it exists.

Many option names have changed to improve consistency although the old names from v1 are still accepted. In general, db-* options have been renamed to pg-* and backup-*/retention-* options have been renamed to repo-* when appropriate.

PostgreSQL and repository options must be indexed when using the new names introduced in v2, e.g. pg1-host, pg1-path, repo1-path, repo1-type, etc. Only one repository is allowed currently but more flexibility is planned for v2.

Build

RHEL/CentOS 7 packages for pgBackRest are available from Crunchy Data or yum.postgresql.org, but it is also easy to download the source and install manually.

When building from source it is best to use a build host rather than building on production. Many of the tools required for the build should generally not be installed in production. pgBackRest consists of a single executable so it is easy to copy to a new host once it is built.

build ⇒ Download version 2.15 of pgBackRest

sudo wget -q -O - \
       https://github.com/pgbackrest/pgbackrest/archive/release/2.15.1.tar.gz | \
       sudo tar zx -C /root

build ⇒ Install build dependencies

sudo yum install build-essential gcc make openssl-devel libxml2-devel \
       perl-ExtUtils-Embed

pgBackRest supports 32-bit distributions that build Perl with 64-bit integer support.

build ⇒ Check for 64-bit integers

perl -V | grep USE_64_BIT_INT

                        USE_64_BIT_INT USE_ITHREADS USE_LARGE_FILES

The pgBackRest executable is written in C. This allows certain time-critical commands (like async archive-push/archive-get) to run more quickly.

build ⇒ Build pgBackRest package

(cd /root/pgbackrest-release-2.15/src && ./configure)

sudo make -s -C /root/pgbackrest-release-2.15/src

Installation

A new host named pg1 is created to contain the demo cluster and run pgBackRest examples.

pgBackRest needs to be installed from a package or installed manually as shown here.

pg-primary ⇒ Copy pgBackRest binary from build host

sudo scp build:/root/pgbackrest-release-2.15/src/pgbackrest /usr/bin

sudo chmod 755 /usr/bin/pgbackrest

pgBackRest contains embedded Perl which requires some additional modules.

pg-primary ⇒ Install required Perl packages

sudo yum install perl perl-Time-HiRes perl-Digest-SHA perl-DBD-Pg perl-JSON-PP

Finally, pgBackRest requires log and configuration directories and a configuration file.

pg-primary ⇒ Create pgBackRest configuration file and directories

sudo mkdir -p -m 770 /var/log/pgbackrest

sudo chown postgres:postgres /var/log/pgbackrest

sudo mkdir -p /etc/pgbackrest

sudo mkdir -p /etc/pgbackrest/conf.d

sudo touch /etc/pgbackrest/pgbackrest.conf

sudo chmod 640 /etc/pgbackrest/pgbackrest.conf

sudo chown postgres:postgres /etc/pgbackrest/pgbackrest.conf

pgBackRest should now be properly installed but it is best to check. If any dependencies were missed then you will get an error when running pgBackRest from the command line.

pg-primary ⇒ Make sure the installation worked

sudo -u postgres pgbackrest

pgBackRest 2.15 - General help

Usage:
    pgbackrest [options] [command]

Commands:
    archive-get     Get a WAL segment from the archive.
    archive-push    Push a WAL segment to the archive.
    backup          Backup a database cluster.
    check           Check the configuration.
    expire          Expire backups that exceed retention.
    help            Get help.
    info            Retrieve information about backups.
    restore         Restore a database cluster.
    stanza-create   Create the required stanza data.
    stanza-delete   Delete a stanza.
    stanza-upgrade  Upgrade a stanza.
    start           Allow pgBackRest processes to run.
    stop            Stop pgBackRest processes from running.
    version         Get version.

Use 'pgbackrest help [command]' for more information.

Quick Start

The Quick Start section will cover basic configuration of pgBackRest and PostgreSQL and introduce the backup, restore, and info commands.

6.1

Setup Demo Cluster

Creating the demo cluster is optional but is strongly recommended, especially for new users, since the example commands in the user guide reference the demo cluster; the examples assume the demo cluster is running on the default port (i.e. 5432). The cluster will not be started until a later section because there is still some configuration to do.

pg-primary ⇒ Create the demo cluster

sudo -u postgres /usr/pgsql-9.6/bin/initdb \
       -D /var/lib/pgsql/9.6/data -k -A peer

By default PostgreSQL will only accept local connections. The examples in this guide will require connections from other servers so listen_addresses is configured to listen on all interfaces. This may not be appropriate for secure installations.

pg-primary:/var/lib/pgsql/9.6/data/postgresql.conf ⇒ Set listen_addresses

listen_addresses = '*'

For demonstration purposes the log_line_prefix setting will be minimally configured. This keeps the log output as brief as possible to better illustrate important information.

pg-primary:/var/lib/pgsql/9.6/data/postgresql.conf ⇒ Set log_line_prefix

listen_addresses = '*'
log_line_prefix = ''

By default RHEL/CentOS 7 includes the day of the week in the log filename. This makes automating the user guide a bit more complicated so the log_filename is set to a constant.

pg-primary:/var/lib/pgsql/9.6/data/postgresql.conf ⇒ Set log_filename

listen_addresses = '*'
log_filename = 'postgresql.log'
log_line_prefix = ''

6.2

Configure Cluster Stanza

A stanza is the configuration for a PostgreSQL database cluster that defines where it is located, how it will be backed up, archiving options, etc. Most db servers will only have one Postgres database cluster and therefore one stanza, whereas backup servers will have a stanza for every database cluster that needs to be backed up.

It is tempting to name the stanza after the primary cluster but a better name describes the databases contained in the cluster. Because the stanza name will be used for the primary and all replicas it is more appropriate to choose a name that describes the actual function of the cluster, such as app or dw, rather than the local cluster name, such as main or prod.

The name 'demo' describes the purpose of this cluster accurately so that will also make a good stanza name.

pgBackRest needs to know where the base data directory for the PostgreSQL cluster is located. The path can be requested from PostgreSQL directly but in a recovery scenario the PostgreSQL process will not be available. During backups the value supplied to pgBackRest will be compared against the path that PostgreSQL is running on and they must be equal or the backup will return an error. Make sure that pg-path is exactly equal to data_directory in postgresql.conf.

By default RHEL/CentOS 7 stores clusters in /var/lib/pgsql/[version]/data so it is easy to determine the correct path for the data directory.

When creating the /etc/pgbackrest/pgbackrest.conf file, the database owner (usually postgres) must be granted read privileges.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure the PostgreSQL cluster data directory

[demo]
pg1-path=/var/lib/pgsql/9.6/data

pgBackRest configuration files follow the Windows INI convention. Sections are denoted by text in brackets and key/value pairs are contained in each section. Lines beginning with # are ignored and can be used as comments.

There are multiple ways the pgBackRest configuration files can be loaded:

config and config-include-path are default: the default config file will be loaded, if it exists, and *.conf files in the default config include path will be appended, if they exist.
config option is specified: only the specified config file will be loaded and is expected to exist.
config-include-path is specified: *.conf files in the config include path will be loaded and the path is required to exist. The default config file will be be loaded if it exists. If it is desireable to load only the files in the specified config include path, then the --no-config option can also be passed.
config and config-include-path are specified: using the user-specified values, the config file will be loaded and *.conf files in the config include path will be appended. The files are expected to exist.
config-path is specified: this setting will override the base path for the default location of the config file and/or the base path of the default config-include-path setting unless the config and/or config-incude-path option is explicitly set.

The files are concatenated as if they were one big file; order doesn't matter, but there is precedence based on sections. The precedence (highest to lowest) is:

[stanza:command]
[stanza]
[global:command]
[global]

NOTE:

--config, --config-include-path and --config-path are command-line only options.

pgBackRest can also be configured using environment variables as described in the command reference.

pg-primary ⇒ Configure log-path using the environment

sudo -u postgres bash -c ' \
       export PGBACKREST_LOG_PATH=/path/set/by/env && \
       pgbackrest --log-level-console=error help backup log-path'

pgBackRest 2.15 - 'backup' command - 'log-path' option help

Path where log files are stored.

The log path provides a location for pgBackRest to store log files. Note that
if log-level-file=off then no log path is required.

current: /path/set/by/env

default: /var/log/pgbackrest

6.3

Create the Repository

The repository is where pgBackRest stores backups and archives WAL segments.

It may be difficult to estimate in advance how much space you'll need. The best thing to do is take some backups then record the size of different types of backups (full/incr/diff) and measure the amount of WAL generated per day. This will give you a general idea of how much space you'll need, though of course requirements will likely change over time as your database evolves.

For this demonstration the repository will be stored on the same host as the PostgreSQL server. This is the simplest configuration and is useful in cases where traditional backup software is employed to backup the database host.

pg-primary ⇒ Create the pgBackRest repository

sudo mkdir -p /var/lib/pgbackrest

sudo chmod 750 /var/lib/pgbackrest

sudo chown postgres:postgres /var/lib/pgbackrest

The repository path must be configured so pgBackRest knows where to find it.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure the pgBackRest repository path

[demo]
pg1-path=/var/lib/pgsql/9.6/data

[global]
repo1-path=/var/lib/pgbackrest

6.4

Configure Archiving

Backing up a running PostgreSQL cluster requires WAL archiving to be enabled. Note that at least one WAL segment will be created during the backup process even if no explicit writes are made to the cluster.

pg-primary:/var/lib/pgsql/9.6/data/postgresql.conf ⇒ Configure archive settings

archive_command = 'pgbackrest --stanza=demo archive-push %p'
archive_mode = on
listen_addresses = '*'
log_filename = 'postgresql.log'
log_line_prefix = ''
max_wal_senders = 3
wal_level = replica

Setting wal_level to at least replica and increasing max_wal_senders is a good idea even if there are currently no replicas as this will allow them to be added later without restarting the primary cluster.

The PostgreSQL cluster must be restarted after making these changes and before performing a backup.

pg-primary ⇒ Restart the demo cluster

sudo systemctl restart postgresql-9.6.service

When archiving a WAL segment is expected to take more than 60 seconds (the default) to reach the pgBackRest repository, then the pgBackRest archive-timeout option should be increased. Note that this option is not the same as the PostgreSQL archive_timeout option which is used to force a WAL segment switch; useful for databases where there are long periods of inactivity. For more information on the PostgreSQL archive_timeout option, see PostgreSQL Write Ahead Log.

The archive-push command can be configured with its own options. For example, a lower compression level may be set to speed archiving without affecting the compression used for backups.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Config archive-push to use a lower compression level

[demo]
pg1-path=/var/lib/pgsql/9.6/data

[global]
repo1-path=/var/lib/pgbackrest

[global:archive-push]
compress-level=3

This configuration technique can be used for any command and can even target a specific stanza, e.g. demo:archive-push.

6.5

Configure Retention

pgBackRest expires backups based on retention options.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure retention to 2 full backups

[demo]
pg1-path=/var/lib/pgsql/9.6/data

[global]
repo1-path=/var/lib/pgbackrest
repo1-retention-full=2

[global:archive-push]
compress-level=3

More information about retention can be found in the Retention section.

6.6

Configure Repository Encryption

The repository will be configured with a cipher type and key to demonstrate encryption. Encryption is always performed client-side even if the repository type (e.g. S3 or other object store) supports encryption.

It is important to use a long, random passphrase for the cipher key. A good way to generate one is to run: openssl rand -base64 48.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure pgBackRest repository encryption

[demo]
pg1-path=/var/lib/pgsql/9.6/data

[global]
repo1-cipher-pass=zWaf6XtpjIVZC5444yXB+cgFDFl7MxGlgkZSaoPvTGirhPygu4jOKOXf9LO4vjfO
repo1-cipher-type=aes-256-cbc
repo1-path=/var/lib/pgbackrest
repo1-retention-full=2

[global:archive-push]
compress-level=3

Once the repository has been configured and the stanza created and checked, the repository encryption settings cannot be changed.

6.7

Create the Stanza

The stanza-create command must be run on the host where the repository is located to initialize the stanza. It is recommended that the check command be run after stanza-create to ensure archiving and backups are properly configured.

pg-primary ⇒ Create the stanza and check the configuration

sudo -u postgres pgbackrest --stanza=demo --log-level-console=info stanza-create

P00   INFO: stanza-create command begin 2.15: --log-level-console=info --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/var/lib/pgbackrest --stanza=demo

P00   INFO: stanza-create command end: completed successfully

6.8

Check the Configuration

The check command validates that pgBackRest and the archive_command setting are configured correctly for archiving and backups. It detects misconfigurations, particularly in archiving, that result in incomplete backups because required WAL segments did not reach the archive. The command can be run on the database or the repository host. The command may also be run on the standby host, however, since pg_switch_xlog()/pg_switch_wal() cannot be performed on the standby, the command will only test the repository configuration.

Note that pg_create_restore_point('pgBackRest Archive Check') and pg_switch_xlog()/pg_switch_wal() are called to force PostgreSQL to archive a WAL segment. Restore points are only supported in PostgreSQL >= 9.1 so for older versions the check command may fail if there has been no write activity since the last log rotation, therefore it is recommended that activity be generated by the user if there have been no writes since the last WAL switch before running the check command.

pg-primary ⇒ Check the configuration

sudo -u postgres pgbackrest --stanza=demo --log-level-console=info check

P00   INFO: check command begin 2.15: --log-level-console=info --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/var/lib/pgbackrest --stanza=demo

P00   INFO: WAL segment 000000010000000000000001 successfully stored in the archive at '/var/lib/pgbackrest/archive/demo/9.6-1/0000000100000000/000000010000000000000001-950df57b860fa827a4b4d20dbd0717976f07504e.gz'

P00   INFO: check command end: completed successfully

6.9

Perform a Backup

To perform a backup of the PostgreSQL cluster run pgBackRest with the backup command.

pg-primary ⇒ Backup the demo cluster

sudo -u postgres pgbackrest --stanza=demo \
       --log-level-console=info backup

P00   INFO: backup command begin 2.15: --log-level-console=info --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/var/lib/pgbackrest --repo1-retention-full=2 --stanza=demo

P00   WARN: no prior backup exists, incr backup has been changed to full

P00   INFO: execute non-exclusive pg_start_backup() with label "pgBackRest backup started at 2019-06-25 01:43:23": backup begins after the next regular checkpoint completes
P00   INFO: backup start archive = 000000010000000000000002, lsn = 0/2000028
       [filtered 864 lines of output]
P01   INFO: backup file /var/lib/pgsql/9.6/data/base/1/12312 (0B, 100%)
P01   INFO: backup file /var/lib/pgsql/9.6/data/base/1/12307 (0B, 100%)

P00   INFO: full backup size = 20.9MB

P00   INFO: execute non-exclusive pg_stop_backup() and wait for all WAL segments to archive
P00   INFO: backup stop archive = 000000010000000000000002, lsn = 0/2000130
       [filtered 4 lines of output]

By default pgBackRest will attempt to perform an incremental backup. However, an incremental backup must be based on a full backup and since no full backup existed pgBackRest ran a full backup instead.

The type option can be used to specify a full or differential backup.

pg-primary ⇒ Differential backup of the demo cluster

sudo -u postgres pgbackrest --stanza=demo --type=diff \
       --log-level-console=info backup

       [filtered 3 lines of output]
P00   INFO: backup start archive = 000000010000000000000003, lsn = 0/3000028
P01   INFO: backup file /var/lib/pgsql/9.6/data/global/pg_control (8KB, 100%) checksum db2865c02e0b7e6443868fd03edadaa3954a1d51

P00   INFO: diff backup size = 8KB

P00   INFO: execute non-exclusive pg_stop_backup() and wait for all WAL segments to archive
P00   INFO: backup stop archive = 000000010000000000000003, lsn = 0/30000F8
       [filtered 4 lines of output]

This time there was no warning because a full backup already existed. While incremental backups can be based on a full or differential backup, differential backups must be based on a full backup. A full backup can be performed by running the backup command with --type=full.

More information about the backup command can be found in the Backup section.

6.10

Schedule a Backup

Backups can be scheduled with utilities such as cron.

In the following example, two cron jobs are configured to run; full backups are scheduled for 6:30 AM every Sunday with differential backups scheduled for 6:30 AM Monday through Saturday. If this crontab is installed for the first time mid-week, then pgBackRest will run a full backup the first time the differential job is executed, followed the next day by a differential backup.

#m h   dom mon dow   command
30 06  *   *   0     pgbackrest --type=full --stanza=demo backup
30 06  *   *   1-6   pgbackrest --type=diff --stanza=demo backup

Once backups are scheduled it's important to configure retention so backups are expired on a regular schedule, see Retention.

6.11

Backup Information

Use the info command to get information about backups.

pg-primary ⇒ Get info for the demo cluster

sudo -u postgres pgbackrest info

stanza: demo
    status: ok
    cipher: aes-256-cbc

    db (current)
        wal archive min/max (9.6-1): 000000010000000000000002/000000010000000000000003

        full backup: 20190625-014323F

            timestamp start/stop: 2019-06-25 01:43:23 / 2019-06-25 01:43:35
            wal start/stop: 000000010000000000000002 / 000000010000000000000002
            database size: 20.9MB, backup size: 20.9MB
            repository size: 2.5MB, repository backup size: 2.5MB

        diff backup: 20190625-014323F_20190625-014336D

            timestamp start/stop: 2019-06-25 01:43:36 / 2019-06-25 01:43:39
            wal start/stop: 000000010000000000000003 / 000000010000000000000003
            database size: 20.9MB, backup size: 8.2KB
            repository size: 2.5MB, repository backup size: 421B
            backup reference list: 20190625-014323F

Each stanza has a separate section and it is possible to limit output to a single stanza with the --stanza option. The stanza 'status' gives a brief indication of the stanza's health. If this is 'ok' then pgBackRest is functioning normally. The 'wal archive min/max' shows the minimum and maximum WAL currently stored in the archive. Note that there may be gaps due to archive retention policies or other reasons.

The backups are displayed oldest to newest. The oldest backup will always be a full backup (indicated by an F at the end of the label) but the newest backup can be full, differential (ends with D), or incremental (ends with I).

The 'timestamp start/stop' defines the time period when the backup ran. The 'timestamp stop' can be used to determine the backup to use when performing Point-In-Time Recovery. More information about Point-In-Time Recovery can be found in the Point-In-Time Recovery section.

The 'wal start/stop' defines the WAL range that is required to make the database consistent when restoring. The backup command will ensure that this WAL range is in the archive before completing.

The 'database size' is the full uncompressed size of the database while 'backup size' is the amount of data actually backed up (these will be the same for full backups). The 'repository size' includes all the files from this backup and any referenced backups that are required to restore the database while 'repository backup size' includes only the files in this backup (these will also be the same for full backups). Repository sizes reflect compressed file sizes if compression is enabled in pgBackRest or the filesystem.

The 'backup reference list' contains the additional backups that are required to restore this backup.

6.12

Restore a Backup

Backups can protect you from a number of disaster scenarios, the most common of which are hardware failure and data corruption. The easiest way to simulate data corruption is to remove an important PostgreSQL cluster file.

pg-primary ⇒ Stop the demo cluster and delete the pg_control file

sudo systemctl stop postgresql-9.6.service

sudo -u postgres rm /var/lib/pgsql/9.6/data/global/pg_control

Starting the cluster without this important file will result in an error.

pg-primary ⇒ Attempt to start the corrupted demo cluster

sudo systemctl start postgresql-9.6.service

sudo systemctl status postgresql-9.6.service

       [filtered 8 lines of output]
Jun 25 01:43:41 pg-primary systemd[1]: Starting PostgreSQL 9.6 database server...
Jun 25 01:43:41 pg-primary systemd[1]: postgresql-9.6.service: main process exited, code=exited, status=2/INVALIDARGUMENT

Jun 25 01:43:41 pg-primary systemd[1]: Failed to start PostgreSQL 9.6 database server.

Jun 25 01:43:41 pg-primary systemd[1]: Unit postgresql-9.6.service entered failed state.
Jun 25 01:43:41 pg-primary systemd[1]: postgresql-9.6.service failed.

To restore a backup of the PostgreSQL cluster run pgBackRest with the restore command. The cluster needs to be stopped (in this case it is already stopped) and all files must be removed from the PostgreSQL data directory.

pg-primary ⇒ Remove old files from demo cluster

sudo -u postgres find /var/lib/pgsql/9.6/data -mindepth 1 -delete

pg-primary ⇒ Restore the demo cluster and start PostgreSQL

sudo -u postgres pgbackrest --stanza=demo restore

sudo systemctl start postgresql-9.6.service

This time the cluster started successfully since the restore replaced the missing pg_control file.

More information about the restore command can be found in the Restore section.

Backup

The Backup section introduces additional backup command features.

7.1

Fast Start Option

By default pgBackRest will wait for the next regularly scheduled checkpoint before starting a backup. Depending on the checkpoint_timeout and checkpoint_segments settings in PostgreSQL it may be quite some time before a checkpoint completes and the backup can begin.

pg-primary ⇒ Incremental backup of the demo cluster with the regularly scheduled checkpoint

sudo -u postgres pgbackrest --stanza=demo --type=incr \
       --log-level-console=info backup

P00   INFO: backup command begin 2.15: --log-level-console=info --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/var/lib/pgbackrest --repo1-retention-full=2 --stanza=demo --type=incr
P00   INFO: last backup label = 20190625-014323F_20190625-014336D, version = 2.15

P00   INFO: execute non-exclusive pg_start_backup() with label "pgBackRest backup started at 2019-06-25 01:43:55": backup begins after the next regular checkpoint completes

P00   INFO: backup start archive = 000000020000000000000005, lsn = 0/5000028
P00   WARN: a timeline switch has occurred since the last backup, enabling delta checksum
       [filtered 9 lines of output]

When --start-fast is passed on the command-line or start-fast=y is set in /etc/pgbackrest/pgbackrest.conf an immediate checkpoint is requested and the backup will start more quickly. This is convenient for testing and for ad-hoc backups. For instance, if a backup is being taken at the beginning of a release window it makes no sense to wait for a checkpoint. Since regularly scheduled backups generally only happen once per day it is unlikely that enabling the start-fast in /etc/pgbackrest/pgbackrest.conf will negatively affect performance, however for high-volume transactional systems you may want to pass --start-fast on the command-line instead. Alternately, it is possible to override the setting in the configuration file by passing --no-start-fast on the command-line.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Enable the start-fast option

pg-primary ⇒ Incremental backup of the demo cluster with an immediate checkpoint

sudo -u postgres pgbackrest --stanza=demo --type=incr \
       --log-level-console=info backup

P00   INFO: backup command begin 2.15: --log-level-console=info --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/var/lib/pgbackrest --repo1-retention-full=2 --stanza=demo --start-fast --type=incr
P00   INFO: last backup label = 20190625-014323F_20190625-014355I, version = 2.15

P00   INFO: execute non-exclusive pg_start_backup() with label "pgBackRest backup started at 2019-06-25 01:44:01": backup begins after the requested immediate checkpoint completes

P00   INFO: backup start archive = 000000020000000000000006, lsn = 0/6000028
P01   INFO: backup file /var/lib/pgsql/9.6/data/global/pg_control (8KB, 100%) checksum faa4883811ea5e218d29407b0aed84a17a3b44e3
       [filtered 7 lines of output]

7.2

Archive Timeout

During an online backup pgBackRest waits for WAL segments that are required for backup consistency to be archived. This wait time is governed by the pgBackRest archive-timeout option which defaults to 60 seconds. If archiving an individual segment is known to take longer then this option should be increased.

Monitoring

Monitoring is an important part of any production system. There are many tools available and pgBackRest can be monitored on any of them with a little work.

pgBackRest can output information about the repository in JSON format which includes a list of all backups for each stanza and WAL archive info. A script is needed to extract information in a format that the monitoring system can understand.

pg-primary ⇒ Get pgBackRest info in JSON format

sudo -u postgres pgbackrest --output=json info

[
    {
        "archive" : [
            {
                "database" : {
                    "id" : 1
                },
                "id" : "9.6-1",
                "max" : "000000020000000000000006",
                "min" : "000000010000000000000002"
            }
        ],
        "backup" : [
            {
                "archive" : {
                    "start" : "000000010000000000000002",
                    "stop" : "000000010000000000000002"
                },
                "backrest" : {
                    "format" : 5,
                    "version" : "2.15"
                },
                "database" : {
                    "id" : 1
                },
                "info" : {
                    "delta" : 21952964,
                    "repository" : {
                        "delta" : 2581349,
                        "size" : 2581349
                    },
                    "size" : 21952964
                },
                "label" : "20190625-014323F",
                "prior" : null,
                "reference" : null,
                "timestamp" : {
                    "start" : 1561427003,
                    "stop" : 1561427015
                },
                "type" : "full"
            },
            {
                "archive" : {
                    "start" : "000000010000000000000003",
                    "stop" : "000000010000000000000003"
                },
                "backrest" : {
                    "format" : 5,
                    "version" : "2.15"
                },
                "database" : {
                    "id" : 1
                },
                "info" : {
                    "delta" : 8421,
                    "repository" : {
                        "delta" : 421,
                        "size" : 2581333
                    },
                    "size" : 21952964
                },
                "label" : "20190625-014323F_20190625-014336D",
                "prior" : "20190625-014323F",
                "reference" : [
                    "20190625-014323F"
                ],
                "timestamp" : {
                    "start" : 1561427016,
                    "stop" : 1561427019
                },
                "type" : "diff"
            },
            {
                "archive" : {
                    "start" : "000000020000000000000005",
                    "stop" : "000000020000000000000005"
                },
                "backrest" : {
                    "format" : 5,
                    "version" : "2.15"
                },
                "database" : {
                    "id" : 1
                },
                "info" : {
                    "delta" : 9341,
                    "repository" : {
                        "delta" : 837,
                        "size" : 2581477
                    },
                    "size" : 21953532
                },
                "label" : "20190625-014323F_20190625-014355I",
                "prior" : "20190625-014323F_20190625-014336D",
                "reference" : [
                    "20190625-014323F"
                ],
                "timestamp" : {
                    "start" : 1561427035,
                    "stop" : 1561427039
                },
                "type" : "incr"
            },
            {
                "archive" : {
                    "start" : "000000020000000000000006",
                    "stop" : "000000020000000000000006"
                },
                "backrest" : {
                    "format" : 5,
                    "version" : "2.15"
                },
                "database" : {
                    "id" : 1
                },
                "info" : {
                    "delta" : 8421,
                    "repository" : {
                        "delta" : 421,
                        "size" : 2581477
                    },
                    "size" : 21953532
                },
                "label" : "20190625-014323F_20190625-014401I",
                "prior" : "20190625-014323F_20190625-014355I",
                "reference" : [
                    "20190625-014323F",
                    "20190625-014323F_20190625-014355I"
                ],
                "timestamp" : {
                    "start" : 1561427041,
                    "stop" : 1561427044
                },
                "type" : "incr"
            }
        ],
        "cipher" : "aes-256-cbc",
        "db" : [
            {
                "id" : 1,
                "system-id" : 6706277820808323406,
                "version" : "9.6"
            }
        ],
        "name" : "demo",
        "status" : {
            "code" : 0,
            "message" : "ok"
        }
    }
]

8.1

In PostgreSQL

The PostgreSQL COPY command allows pgBackRest info to be loaded into a table. The following example wraps that logic in a function that can be used to perform real-time queries.

pg-primary ⇒ Load pgBackRest info function for PostgreSQL

sudo -u postgres cat \
       /var/lib/pgsql/pgbackrest/doc/example/pgsql-pgbackrest-info.sql

-- An example of monitoring pgBackRest from within PostgreSQL
--
-- Use copy to export data from the pgBackRest info command into the jsonb
-- type so it can be queried directly by PostgreSQL.

-- Create monitor schema
create schema monitor;

-- Get pgBackRest info in JSON format
create function monitor.pgbackrest_info()
    returns jsonb AS $$
declare
    data jsonb;
begin
    -- Create a temp table to hold the JSON data
    create temp table temp_pgbackrest_data (data jsonb);

    -- Copy data into the table directory from the pgBackRest into command
    copy temp_pgbackrest_data (data)
        from program
            'pgbackrest --output=json info | tr ''\n'' '' ''' (format text);

    select temp_pgbackrest_data.data
      into data
      from temp_pgbackrest_data;

    drop table temp_pgbackrest_data;

    return data;
end $$ language plpgsql;

sudo -u postgres psql -f \
       /var/lib/pgsql/pgbackrest/doc/example/pgsql-pgbackrest-info.sql

Now the monitor.pgbackrest_info() function can be used to determine the last successful backup time and archived WAL for a stanza.

pg-primary ⇒ Query last successful backup time and archived WAL

sudo -u postgres cat \
       /var/lib/pgsql/pgbackrest/doc/example/pgsql-pgbackrest-query.sql

-- Get last successful backup for each stanza
--
-- Requires the monitor.pgbackrest_info function.
with stanza as
(
    select data->'name' as name,
           data->'backup'->(
               jsonb_array_length(data->'backup') - 1) as last_backup,
           data->'archive'->(
               jsonb_array_length(data->'archive') - 1) as current_archive
      from jsonb_array_elements(monitor.pgbackrest_info()) as data
)
select name,
       to_timestamp(
           (last_backup->'timestamp'->>'stop')::numeric) as last_successful_backup,
       current_archive->>'max' as last_archived_wal
  from stanza;

sudo -u postgres psql -f \
       /var/lib/pgsql/pgbackrest/doc/example/pgsql-pgbackrest-query.sql

  name  | last_successful_backup |    last_archived_wal
--------+------------------------+--------------------------
 "demo" | 2019-06-25 01:44:04+00 | 000000020000000000000006
(1 row)

Retention

Generally it is best to retain as many backups as possible to provide a greater window for Point-in-Time Recovery, but practical concerns such as disk space must also be considered. Retention options remove older backups once they are no longer needed.

9.1

Full Backup Retention

Set repo1-retention-full to the number of full backups required. New backups must be completed before expiration will occur — that means if repo1-retention-full=2 then there will be three full backups stored before the oldest one is expired.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure repo1-retention-full

Backup repo1-retention-full=2 but currently there is only one full backup so the next full backup to run will not expire any full backups.

pg-primary ⇒ Perform a full backup

sudo -u postgres pgbackrest --stanza=demo --type=full \
       --log-level-console=detail backup

       [filtered 874 lines of output]
P00   INFO: backup command end: completed successfully
P00   INFO: expire command begin

P00 DETAIL: archive retention on backup 20190625-014323F, archiveId = 9.6-1, start = 000000010000000000000002

P00 DETAIL: no archive to remove, archiveId = 9.6-1
P00   INFO: expire command end: completed successfully

Archive is expired because WAL segments were generated before the oldest backup. These are not useful for recovery — only WAL segments generated after a backup can be used to recover that backup.

pg-primary ⇒ Perform a full backup

sudo -u postgres pgbackrest --stanza=demo --type=full \
       --log-level-console=info backup

       [filtered 873 lines of output]
P00   INFO: backup command end: completed successfully
P00   INFO: expire command begin

P00   INFO: expire full backup set: 20190625-014323F, 20190625-014323F_20190625-014336D, 20190625-014323F_20190625-014355I, 20190625-014323F_20190625-014401I

P00   INFO: remove expired backup 20190625-014323F_20190625-014401I
P00   INFO: remove expired backup 20190625-014323F_20190625-014355I
       [filtered 2 lines of output]

The 20190625-014323F full backup is expired and archive retention is based on the 20190625-014409F which is now the oldest full backup.

9.2

Differential Backup Retention

Set repo1-retention-diff to the number of differential backups required. Differentials only rely on the prior full backup so it is possible to create a rolling set of differentials for the last day or more. This allows quick restores to recent points-in-time but reduces overall space consumption.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure repo1-retention-diff

Backup repo1-retention-diff=1 so two differentials will need to be performed before one is expired. An incremental backup is added to demonstrate incremental expiration. Incremental backups cannot be expired independently — they are always expired with their related full or differential backup.

pg-primary ⇒ Perform differential and incremental backups

sudo -u postgres pgbackrest --stanza=demo --type=diff backup

sudo -u postgres pgbackrest --stanza=demo --type=incr backup

Now performing a differential backup will expire the previous differential and incremental backups leaving only one differential backup.

pg-primary ⇒ Perform a differential backup

sudo -u postgres pgbackrest --stanza=demo --type=diff \
       --log-level-console=info backup

       [filtered 9 lines of output]
P00   INFO: backup command end: completed successfully
P00   INFO: expire command begin

P00   INFO: expire diff backup set: 20190625-014423F_20190625-014437D, 20190625-014423F_20190625-014442I

P00   INFO: remove expired backup 20190625-014423F_20190625-014442I
P00   INFO: remove expired backup 20190625-014423F_20190625-014437D

9.3

Archive Retention

Although pgBackRest automatically removes archived WAL segments when expiring backups (the default expires WAL for full backups based on the repo1-retention-full option), it may be useful to expire archive more aggressively to save disk space. Note that full backups are treated as differential backups for the purpose of differential archive retention.

Expiring archive will never remove WAL segments that are required to make a backup consistent. However, since Point-in-Time-Recovery (PITR) only works on a continuous WAL stream, care should be taken when aggressively expiring archive outside of the normal backup expiration process.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure repo1-retention-diff

pg-primary ⇒ Perform differential backup

sudo -u postgres pgbackrest --stanza=demo --type=diff \
       --log-level-console=info backup

       [filtered 8 lines of output]
P00   INFO: execute non-exclusive pg_stop_backup() and wait for all WAL segments to archive
P00   INFO: backup stop archive = 000000020000000000000011, lsn = 0/110000F8

P00   INFO: new backup label = 20190625-014423F_20190625-014453D

P00   INFO: backup command end: completed successfully
P00   INFO: expire command begin

pg-primary ⇒ Expire archive

sudo -u postgres pgbackrest --stanza=demo --log-level-console=detail \
       --repo1-retention-archive-type=diff --repo1-retention-archive=1 expire

P00   INFO: expire command begin 2.15: --log-level-console=detail --log-level-stderr=off --no-log-timestamp --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/var/lib/pgbackrest --repo1-retention-archive=1 --repo1-retention-archive-type=diff --repo1-retention-diff=2 --repo1-retention-full=2 --stanza=demo
P00 DETAIL: archive retention on backup 20190625-014409F, archiveId = 9.6-1, start = 000000020000000000000008, stop = 000000020000000000000008
P00 DETAIL: archive retention on backup 20190625-014423F, archiveId = 9.6-1, start = 00000002000000000000000A, stop = 00000002000000000000000A

P00 DETAIL: archive retention on backup 20190625-014423F_20190625-014447D, archiveId = 9.6-1, start = 00000002000000000000000E, stop = 00000002000000000000000E

P00 DETAIL: archive retention on backup 20190625-014423F_20190625-014453D, archiveId = 9.6-1, start = 000000020000000000000011

P00 DETAIL: remove archive: archiveId = 9.6-1, start = 000000020000000000000009, stop = 000000020000000000000009
P00 DETAIL: remove archive: archiveId = 9.6-1, start = 00000002000000000000000B, stop = 00000002000000000000000D
P00 DETAIL: remove archive: archiveId = 9.6-1, start = 00000002000000000000000F, stop = 000000020000000000000010

P00   INFO: expire command end: completed successfully

The 20190625-014423F_20190625-014447D differential backup has archived WAL segments that must be retained to make the older backups consistent even though they cannot be played any further forward with PITR. WAL segments generated after 20190625-014423F_20190625-014447D but before 20190625-014423F_20190625-014453D are removed. WAL segments generated after the new backup 20190625-014423F_20190625-014453D remain and can be used for PITR.

Since full backups are considered differential backups for the purpose of differential archive retention, if a full backup is now performed with the same settings, only the archive for that full backup is retained for PITR.

Restore

The Restore section introduces additional restore command features.

10.1

Delta Option

Restore a Backup in Quick Start required the database cluster directory to be cleaned before the restore could be performed. The delta option allows pgBackRest to automatically determine which files in the database cluster directory can be preserved and which ones need to be restored from the backup — it also removes files not present in the backup manifest so it will dispose of divergent changes. This is accomplished by calculating a SHA-1 cryptographic hash for each file in the database cluster directory. If the SHA-1 hash does not match the hash stored in the backup then that file will be restored. This operation is very efficient when combined with the process-max option. Since the PostgreSQL server is shut down during the restore, a larger number of processes can be used than might be desirable during a backup when the PostgreSQL server is running.

pg-primary ⇒ Stop the demo cluster, perform delta restore

sudo systemctl stop postgresql-9.6.service

sudo -u postgres pgbackrest --stanza=demo --delta \
       --log-level-console=detail restore

       [filtered 891 lines of output]
P01 DETAIL: restore file /var/lib/pgsql/9.6/data/base/1/12307 - exists and is zero size (0B, 100%)
P00   INFO: write /var/lib/pgsql/9.6/data/recovery.conf

P00   INFO: restore global/pg_control (performed last to ensure aborted restores cannot be started)

P00   INFO: restore command end: completed successfully

pg-primary ⇒ Restart PostgreSQL

sudo systemctl start postgresql-9.6.service

10.2

Restore Selected Databases

There may be cases where it is desirable to selectively restore specific databases from a cluster backup. This could be done for performance reasons or to move selected databases to a machine that does not have enough space to restore the entire cluster backup.

To demonstrate this feature two databases are created: test1 and test2. A fresh backup is run so pgBackRest is aware of the new databases.

pg-primary ⇒ Create two test databases and perform a backup

sudo -u postgres psql -c "create database test1;"

CREATE DATABASE

sudo -u postgres psql -c "create database test2;"

CREATE DATABASE

sudo -u postgres pgbackrest --stanza=demo --type=incr backup

Each test database will be seeded with tables and data to demonstrate that recovery works with selective restore.

pg-primary ⇒ Create a test table in each database

sudo -u postgres psql -c "create table test1_table (id int); \
       insert into test1_table (id) values (1);" test1

INSERT 0 1

sudo -u postgres psql -c "create table test2_table (id int); \
       insert into test2_table (id) values (2);" test2

INSERT 0 1

One of the main reasons to use selective restore is to save space. The size of the test1 database is shown here so it can be compared with the disk utilization after a selective restore.

pg-primary ⇒ Show space used by test1 database

sudo -u postgres du -sh /var/lib/pgsql/9.6/data/base/24576

7.0M	/var/lib/pgsql/9.6/data/base/24576

Stop the cluster and restore only the test2 database. Built-in databases (template0, template1, and postgres) are always restored.

pg-primary ⇒ Restore from last backup including only the test2 database

sudo systemctl stop postgresql-9.6.service

sudo -u postgres pgbackrest --stanza=demo --delta \
       --db-include=test2 restore

sudo systemctl start postgresql-9.6.service

Once recovery is complete the test2 database will contain all previously created tables and data.

pg-primary ⇒ Demonstrate that the test2 database was recovered

sudo -u postgres psql -c "select * from test2_table;" test2

 id
----
  2
(1 row)

The test1 database, despite successful recovery, is not accessible. This is because the entire database was restored as sparse, zeroed files. PostgreSQL can successfully apply WAL on the zeroed files but the database as a whole will not be valid because key files contain no data. This is purposeful to prevent the database from being accidentally used when it might contain partial data that was applied during WAL replay.

pg-primary ⇒ Attempting to connect to the test1 database will produce an error

sudo -u postgres psql -c "select * from test1_table;" test1

psql: FATAL:  relation mapping file "base/24576/pg_filenode.map" contains invalid data

Since the test1 database is restored with sparse, zeroed files it will only require as much space as the amount of WAL that is written during recovery. While the amount of WAL generated during a backup and applied during recovery can be significant it will generally be a small fraction of the total database size, especially for large databases where this feature is most likely to be useful.

It is clear that the test1 database uses far less disk space during the selective restore than it would have if the entire database had been restored.

pg-primary ⇒ Show space used by test1 database after recovery

sudo -u postgres du -sh /var/lib/pgsql/9.6/data/base/24576

152K	/var/lib/pgsql/9.6/data/base/24576

At this point the only action that can be taken on the invalid test1 database is drop database. pgBackRest does not automatically drop the database since this cannot be done until recovery is complete and the cluster is accessible.

pg-primary ⇒ Drop the test1 database

sudo -u postgres psql -c "drop database test1;"

DROP DATABASE

Now that the invalid test1 database has been dropped only the test2 and built-in databases remain.

pg-primary ⇒ List remaining databases

sudo -u postgres psql -c "select oid, datname from pg_database order by oid;"

  oid  |  datname
-------+-----------
     1 | template1
 12469 | template0
 12470 | postgres

 24577 | test2

(4 rows)

Point-in-Time Recovery

Restore a Backup in Quick Start performed default recovery, which is to play all the way to the end of the WAL stream. In the case of a hardware failure this is usually the best choice but for data corruption scenarios (whether machine or human in origin) Point-in-Time Recovery (PITR) is often more appropriate.

Point-in-Time Recovery (PITR) allows the WAL to be played from the last backup to a specified time, transaction id, or recovery point. For common recovery scenarios time-based recovery is arguably the most useful. A typical recovery scenario is to restore a table that was accidentally dropped or data that was accidentally deleted. Recovering a dropped table is more dramatic so that's the example given here but deleted data would be recovered in exactly the same way.

pg-primary ⇒ Backup the demo cluster and create a table with very important data

sudo -u postgres pgbackrest --stanza=demo --type=diff backup

sudo -u postgres psql -c "begin; \
       create table important_table (message text); \
       insert into important_table values ('Important Data'); \
       commit; \
       select * from important_table;"

    message
----------------

 Important Data

(1 row)

It is important to represent the time as reckoned by PostgreSQL and to include timezone offsets. This reduces the possibility of unintended timezone conversions and an unexpected recovery result.

pg-primary ⇒ Get the time from PostgreSQL

sudo -u postgres psql -Atc "select current_timestamp"

2019-06-25 01:45:33.362422+00

Now that the time has been recorded the table is dropped. In practice finding the exact time that the table was dropped is a lot harder than in this example. It may not be possible to find the exact time, but some forensic work should be able to get you close.

pg-primary ⇒ Drop the important table

sudo -u postgres psql -c "begin; \
       drop table important_table; \
       commit; \
       select * from important_table;"

ERROR:  relation "important_table" does not exist

LINE 1: ...le important_table;     commit;     select * from important_...
                                                             ^

Now the restore can be performed with time-based recovery to bring back the missing table.

pg-primary ⇒ Stop PostgreSQL, restore the demo cluster to 2019-06-25 01:45:33.362422+00, and display recovery.conf

sudo systemctl stop postgresql-9.6.service

sudo -u postgres pgbackrest --stanza=demo --delta \
       --type=time "--target=2019-06-25 01:45:33.362422+00" \
       --target-action=promote restore

sudo -u postgres cat /var/lib/pgsql/9.6/data/recovery.conf

restore_command = 'pgbackrest --stanza=demo archive-get %f "%p"'

recovery_target_time = '2019-06-25 01:45:33.362422+00'

recovery_target_action = 'promote'

The recovery.conf file has been automatically generated by pgBackRest so PostgreSQL can be started immediately. Once PostgreSQL has finished recovery the table will exist again and can be queried.

pg-primary ⇒ Start PostgreSQL and check that the important table exists

sudo systemctl start postgresql-9.6.service

sudo -u postgres psql -c "select * from important_table"

    message
----------------

 Important Data

(1 row)

The PostgreSQL log also contains valuable information. It will indicate the time and transaction where the recovery stopped and also give the time of the last transaction to be applied.

pg-primary ⇒ Examine the PostgreSQL log output

sudo -u postgres cat /var/lib/pgsql/9.6/data/pg_log/postgresql.log

LOG:  database system was interrupted; last known up at 2019-06-25 01:45:26 UTC

LOG:  starting point-in-time recovery to 2019-06-25 01:45:33.362422+00

LOG:  restored log file "00000004.history" from archive
LOG:  restored log file "000000040000000000000016" from archive
LOG:  redo starts at 0/16000028
LOG:  consistent recovery state reached at 0/160000F8
LOG:  restored log file "000000040000000000000017" from archive

LOG:  recovery stopping before commit of transaction 639, time 2019-06-25 01:45:33.807831+00

LOG:  redo done at 0/17020CA8

LOG:  last completed transaction was at log time 2019-06-25 01:45:32.937199+00

LOG:  selected new timeline ID: 5
LOG:  archive recovery complete
       [filtered 3 lines of output]

This example was rigged to give the correct result. If a backup after the required time is chosen then PostgreSQL will not be able to recover the lost table. PostgreSQL can only play forward, not backward. To demonstrate this the important table must be dropped (again).

pg-primary ⇒ Drop the important table (again)

sudo -u postgres psql -c "begin; \
       drop table important_table; \
       commit; \
       select * from important_table;"

ERROR:  relation "important_table" does not exist

LINE 1: ...le important_table;     commit;     select * from important_...
                                                             ^

Now take a new backup and attempt recovery from the new backup.

pg-primary ⇒ Perform a backup then attempt recovery from that backup

sudo -u postgres pgbackrest --stanza=demo --type=incr backup

sudo systemctl stop postgresql-9.6.service

sudo -u postgres pgbackrest --stanza=demo --delta \
       --type=time "--target=2019-06-25 01:45:33.362422+00" --target-action=promote restore

sudo systemctl start postgresql-9.6.service

sudo -u postgres psql -c "select * from important_table"

ERROR:  relation "important_table" does not exist

LINE 1: select * from important_table
                      ^

Looking at the log output it's not obvious that recovery failed to restore the table. The key is to look for the presence of the recovery stopping before... and last completed transaction... log messages. If they are not present then the recovery to the specified point-in-time was not successful.

pg-primary ⇒ Examine the PostgreSQL log output to discover the recovery was not successful

sudo -u postgres cat /var/lib/pgsql/9.6/data/pg_log/postgresql.log

LOG:  database system was interrupted; last known up at 2019-06-25 01:45:43 UTC

LOG:  starting point-in-time recovery to 2019-06-25 01:45:33.362422+00

LOG:  restored log file "00000005.history" from archive
LOG:  restored log file "000000050000000000000018" from archive
LOG:  redo starts at 0/18000028

LOG:  consistent recovery state reached at 0/180000F8

LOG:  redo done at 0/180000F8
LOG:  restored log file "000000050000000000000018" from archive
       [filtered 7 lines of output]

Using an earlier backup will allow PostgreSQL to play forward to the correct time. The info command can be used to find the next to last backup.

pg-primary ⇒ Get backup info for the demo cluster

sudo -u postgres pgbackrest info

stanza: demo
    status: ok
    cipher: aes-256-cbc

    db (current)
        wal archive min/max (9.6-1): 000000020000000000000008/000000050000000000000018

        full backup: 20190625-014409F
            timestamp start/stop: 2019-06-25 01:44:09 / 2019-06-25 01:44:22
            wal start/stop: 000000020000000000000008 / 000000020000000000000008
            database size: 20.9MB, backup size: 20.9MB
            repository size: 2.5MB, repository backup size: 2.5MB

        full backup: 20190625-014423F
            timestamp start/stop: 2019-06-25 01:44:23 / 2019-06-25 01:44:35
            wal start/stop: 00000002000000000000000A / 00000002000000000000000A
            database size: 20.9MB, backup size: 20.9MB
            repository size: 2.5MB, repository backup size: 2.5MB

        diff backup: 20190625-014423F_20190625-014453D
            timestamp start/stop: 2019-06-25 01:44:53 / 2019-06-25 01:44:56
            wal start/stop: 000000020000000000000011 / 000000020000000000000011
            database size: 20.9MB, backup size: 97.6KB
            repository size: 2.5MB, repository backup size: 11.8KB
            backup reference list: 20190625-014423F

        incr backup: 20190625-014423F_20190625-014504I
            timestamp start/stop: 2019-06-25 01:45:04 / 2019-06-25 01:45:14
            wal start/stop: 000000030000000000000013 / 000000030000000000000013
            database size: 34.6MB, backup size: 13.7MB
            repository size: 4MB, repository backup size: 1.6MB
            backup reference list: 20190625-014423F, 20190625-014423F_20190625-014453D

        diff backup: 20190625-014423F_20190625-014525D

            timestamp start/stop: 2019-06-25 01:45:25 / 2019-06-25 01:45:32
            wal start/stop: 000000040000000000000016 / 000000040000000000000016
            database size: 27.8MB, backup size: 7.0MB
            repository size: 3.3MB, repository backup size: 846.8KB
            backup reference list: 20190625-014423F

        incr backup: 20190625-014423F_20190625-014542I
            timestamp start/stop: 2019-06-25 01:45:42 / 2019-06-25 01:45:47
            wal start/stop: 000000050000000000000018 / 000000050000000000000018
            database size: 27.7MB, backup size: 2MB
            repository size: 3.3MB, repository backup size: 211KB

            backup reference list: 20190625-014423F, 20190625-014423F_20190625-014525D

The default behavior for restore is to use the last backup but an earlier backup can be specified with the --set option.

pg-primary ⇒ Stop PostgreSQL, restore from the selected backup, and start PostgreSQL

sudo systemctl stop postgresql-9.6.service

sudo -u postgres pgbackrest --stanza=demo --delta \
       --type=time "--target=2019-06-25 01:45:33.362422+00" \
       --set=20190625-014423F_20190625-014525D --target-action=promote restore

sudo systemctl start postgresql-9.6.service

sudo -u postgres psql -c "select * from important_table"

    message
----------------

 Important Data

(1 row)

Now the the log output will contain the expected recovery stopping before... and last completed transaction... messages showing that the recovery was successful.

pg-primary ⇒ Examine the PostgreSQL log output for log messages indicating success

sudo -u postgres cat /var/lib/pgsql/9.6/data/pg_log/postgresql.log

LOG:  database system was interrupted; last known up at 2019-06-25 01:45:26 UTC

LOG:  starting point-in-time recovery to 2019-06-25 01:45:33.362422+00

LOG:  restored log file "00000004.history" from archive
LOG:  restored log file "000000040000000000000016" from archive
LOG:  redo starts at 0/16000028
LOG:  consistent recovery state reached at 0/160000F8
LOG:  restored log file "000000040000000000000017" from archive

LOG:  recovery stopping before commit of transaction 639, time 2019-06-25 01:45:33.807831+00

LOG:  redo done at 0/17020CA8

LOG:  last completed transaction was at log time 2019-06-25 01:45:32.937199+00

LOG:  restored log file "00000005.history" from archive
LOG:  restored log file "00000006.history" from archive
       [filtered 5 lines of output]

S3-Compatible Object Store Support

pg-primary ⇒ Install packages required for S3-compatible object store support

sudo yum install perl-XML-LibXML perl-IO-Socket-SSL

pgBackRest supports locating repositories in S3-compatible object stores. The bucket used to store the repository must be created in advance — pgBackRest will not do it automatically. The repository can be located in the bucket root (/) but it's usually best to place it in a subpath so object store logs or other data can also be stored in the bucket without conflicts.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure S3

[demo]
pg1-path=/var/lib/pgsql/9.6/data

[global]
process-max=4
repo1-cipher-pass=zWaf6XtpjIVZC5444yXB+cgFDFl7MxGlgkZSaoPvTGirhPygu4jOKOXf9LO4vjfO
repo1-cipher-type=aes-256-cbc
repo1-path=/demo-repo
repo1-retention-diff=2
repo1-retention-full=2
repo1-s3-bucket=demo-bucket
repo1-s3-ca-file=/etc/pki/tls/certs/ca-bundle.crt
repo1-s3-endpoint=s3.us-east-1.amazonaws.com
repo1-s3-key=accessKey1
repo1-s3-key-secret=verySecretKey1
repo1-s3-region=us-east-1
repo1-type=s3
start-fast=y

[global:archive-push]
compress-level=3

NOTE:

The region and endpoint will need to be configured to where the bucket is located. The values given here are for the us-east-1 region.

A role should be created to run pgBackRest and the bucket permissions should be set as restrictively as possible. This sample Amazon S3 policy will restrict all reads and writes to the bucket and repository path.

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "s3:ListBucket"
            ],
            "Resource": [
                "arn:aws:s3:::demo-bucket"
            ],
            "Condition": {
                "StringEquals": {
                    "s3:prefix": [
                        "",
                        "demo-repo"
                    ],
                    "s3:delimiter": [
                        "/"
                    ]
                }
            }
        },
        {
            "Effect": "Allow",
            "Action": [
                "s3:ListBucket"
            ],
            "Resource": [
                "arn:aws:s3:::demo-bucket"
            ],
            "Condition": {
                "StringLike": {
                    "s3:prefix": [
                        "demo-repo/*"
                    ]
                }
            }
        },
        {
            "Effect": "Allow",
            "Action": [
                "s3:PutObject",
                "s3:GetObject",
                "s3:DeleteObject"
            ],
            "Resource": [
                "arn:aws:s3:::demo-bucket/demo-repo/*"
            ]
        }
    ]
}

Commands are run exactly as if the repository were stored on a local disk.

pg-primary ⇒ Create the stanza

sudo -u postgres pgbackrest --stanza=demo --log-level-console=info stanza-create

P00   INFO: stanza-create command begin 2.15: --log-level-console=info --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/demo-repo --repo1-s3-bucket=demo-bucket --repo1-s3-ca-file=/etc/pki/tls/certs/ca-bundle.crt --repo1-s3-endpoint=s3.us-east-1.amazonaws.com --repo1-s3-key= --repo1-s3-key-secret= --repo1-s3-region=us-east-1 --repo1-type=s3 --stanza=demo

P00   INFO: stanza-create command end: completed successfully

File creation time in object stores is relatively slow so commands benefit by increasing process-max to parallelize file creation.

pg-primary ⇒ Backup the demo cluster

sudo -u postgres pgbackrest --stanza=demo \
       --log-level-console=info backup

P00   INFO: backup command begin 2.15: --log-level-console=info --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --process-max=4 --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/demo-repo --repo1-retention-diff=2 --repo1-retention-full=2 --repo1-s3-bucket=demo-bucket --repo1-s3-ca-file=/etc/pki/tls/certs/ca-bundle.crt --repo1-s3-endpoint=s3.us-east-1.amazonaws.com --repo1-s3-key= --repo1-s3-key-secret= --repo1-s3-region=us-east-1 --repo1-type=s3 --stanza=demo --start-fast

P00   WARN: no prior backup exists, incr backup has been changed to full

P00   INFO: execute non-exclusive pg_start_backup() with label "pgBackRest backup started at 2019-06-25 01:46:14": backup begins after the requested immediate checkpoint completes
P00   INFO: backup start archive = 000000070000000000000018, lsn = 0/18000028
       [filtered 1136 lines of output]
P04   INFO: backup file /var/lib/pgsql/9.6/data/base/1/12312 (0B, 100%)
P01   INFO: backup file /var/lib/pgsql/9.6/data/base/1/12307 (0B, 100%)

P00   INFO: full backup size = 27.7MB

P00   INFO: execute non-exclusive pg_stop_backup() and wait for all WAL segments to archive
P00   INFO: backup stop archive = 000000070000000000000018, lsn = 0/18000130
       [filtered 5 lines of output]

Delete a Stanza

The stanza-delete command removes data in the repository associated with a stanza.

WARNING:

Use this command with caution — it will permanently remove all backups and archives from the pgBackRest repository for the specified stanza.

To delete a stanza:

Shut down the PostgreSQL cluster associated with the stanza (or use --force to override).
Run the stop command on the repository host.
Run the stanza-delete command on the repository host.

Once the command successfully completes, it is the responsibility of the user to remove the stanza from all pgBackRest configuration files.

pg-primary ⇒ Stop PostgreSQL cluster to be removed

sudo systemctl stop postgresql-9.6.service

pg-primary ⇒ Stop pgBackRest for the stanza

sudo -u postgres pgbackrest --stanza=demo --log-level-console=info stop

P00   INFO: stop command begin 2.15: --log-level-console=info --log-level-stderr=off --no-log-timestamp --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/demo-repo --repo1-s3-bucket=demo-bucket --repo1-s3-ca-file=/etc/pki/tls/certs/ca-bundle.crt --repo1-s3-endpoint=s3.us-east-1.amazonaws.com --repo1-s3-key= --repo1-s3-key-secret= --repo1-s3-region=us-east-1 --repo1-type=s3 --stanza=demo

P00   INFO: stop command end: completed successfully

pg-primary ⇒ Delete the stanza

sudo -u postgres pgbackrest --stanza=demo --log-level-console=info stanza-delete

P00   INFO: stanza-delete command begin 2.15: --log-level-console=info --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --repo1-cipher-pass= --repo1-cipher-type=aes-256-cbc --repo1-path=/demo-repo --repo1-s3-bucket=demo-bucket --repo1-s3-ca-file=/etc/pki/tls/certs/ca-bundle.crt --repo1-s3-endpoint=s3.us-east-1.amazonaws.com --repo1-s3-key= --repo1-s3-key-secret= --repo1-s3-region=us-east-1 --repo1-type=s3 --stanza=demo

P00   INFO: stanza-delete command end: completed successfully

Dedicated Repository Host

The configuration described in Quickstart is suitable for simple installations but for enterprise configurations it is more typical to have a dedicated repository host where the backups and WAL archive files are stored. This separates the backups and WAL archive from the database server so database host failures have less impact. It is still a good idea to employ traditional backup software to backup the repository host.

14.1

Installation

A new host named repository is created to store the cluster backups.

The pgbackrest user is created to own the pgBackRest repository. Any user can own the repository but it is best not to use postgres (if it exists) to avoid confusion.

repository ⇒ Create pgbackrest user

sudo groupadd pgbackrest

sudo adduser -gpgbackrest -n pgbackrest

pgBackRest needs to be installed from a package or installed manually as shown here.

repository ⇒ Copy pgBackRest binary from build host

sudo scp build:/root/pgbackrest-release-2.15/src/pgbackrest /usr/bin

sudo chmod 755 /usr/bin/pgbackrest

pgBackRest contains embedded Perl which requires some additional modules.

repository ⇒ Install required Perl packages

sudo yum install perl perl-Time-HiRes perl-Digest-SHA perl-DBD-Pg perl-JSON-PP

Finally, pgBackRest requires log and configuration directories and a configuration file.

repository ⇒ Create pgBackRest configuration file and directories

sudo mkdir -p -m 770 /var/log/pgbackrest

sudo chown pgbackrest:pgbackrest /var/log/pgbackrest

sudo mkdir -p /etc/pgbackrest

sudo mkdir -p /etc/pgbackrest/conf.d

sudo touch /etc/pgbackrest/pgbackrest.conf

sudo chmod 640 /etc/pgbackrest/pgbackrest.conf

sudo chown pgbackrest:pgbackrest /etc/pgbackrest/pgbackrest.conf

repository ⇒ Create the pgBackRest repository

sudo mkdir -p /var/lib/pgbackrest

sudo chmod 750 /var/lib/pgbackrest

sudo chown pgbackrest:pgbackrest /var/lib/pgbackrest

14.2

Setup Passwordless SSH

pgBackRest requires passwordless SSH to enable communication between the hosts.

repository ⇒ Create repository host key pair

sudo -u pgbackrest mkdir -m 750 /home/pgbackrest/.ssh

sudo -u pgbackrest ssh-keygen -f /home/pgbackrest/.ssh/id_rsa \
       -t rsa -b 4096 -N ""

pg-primary ⇒ Create pg-primary host key pair

sudo -u postgres mkdir -m 750 -p /var/lib/pgsql/.ssh

sudo -u postgres ssh-keygen -f /var/lib/pgsql/.ssh/id_rsa \
       -t rsa -b 4096 -N ""

Exchange keys between repository and pg-primary.

repository ⇒ Copy pg-primary public key to repository

(echo -n 'no-agent-forwarding,no-X11-forwarding,no-port-forwarding,' && \
       echo -n 'command="/usr/bin/pgbackrest ${SSH_ORIGINAL_COMMAND#* }" ' && \
       sudo ssh root@pg-primary cat /var/lib/pgsql/.ssh/id_rsa.pub) | \
       sudo -u pgbackrest tee -a /home/pgbackrest/.ssh/authorized_keys

pg-primary ⇒ Copy repository public key to pg-primary

(echo -n 'no-agent-forwarding,no-X11-forwarding,no-port-forwarding,' && \
       echo -n 'command="/usr/bin/pgbackrest ${SSH_ORIGINAL_COMMAND#* }" ' && \
       sudo ssh root@repository cat /home/pgbackrest/.ssh/id_rsa.pub) | \
       sudo -u postgres tee -a /var/lib/pgsql/.ssh/authorized_keys

Test that connections can be made from repository to pg-primary and vice versa.

repository ⇒ Test connection from repository to pg-primary

sudo -u pgbackrest ssh postgres@pg-primary

pg-primary ⇒ Test connection from pg-primary to repository

sudo -u postgres ssh pgbackrest@repository

NOTE:

ssh has been configured to only allow pgBackRest to be run via passwordless ssh. This enhances security in the event that one of the service accounts is hijacked.

14.3

Configuration

The repository host must be configured with the pg-primary host/user and database path. The primary will be configured as db1 to allow a standby to be added later.

repository:/etc/pgbackrest/pgbackrest.conf ⇒ Configure pg1-host/pg1-host-user and pg1-path

[demo]
pg1-host=pg-primary
pg1-path=/var/lib/pgsql/9.6/data

[global]
repo1-path=/var/lib/pgbackrest
repo1-retention-full=2
start-fast=y

The database host must be configured with the repository host/user. The default for the repo1-host-user option is pgbackrest. If the postgres user does restores on the repository host it is best not to also allow the postgres user to perform backups. However, the postgres user can read the repository directly if it is in the same group as the pgbackrest user.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure repo1-host/repo1-host-user

[demo]
pg1-path=/var/lib/pgsql/9.6/data

[global]
log-level-file=detail
repo1-host=repository

Commands are run the same as on a single host configuration except that some commands such as backup and expire are run from the repository host instead of the database host.

Create the stanza in the new repository.

repository ⇒ Create the stanza

sudo -u pgbackrest pgbackrest --stanza=demo stanza-create

Check that the configuration is correct on both the database and repository hosts. More information about the check command can be found in Check the Configuration.

pg-primary ⇒ Check the configuration

sudo -u postgres pgbackrest --stanza=demo check

repository ⇒ Check the configuration

sudo -u pgbackrest pgbackrest --stanza=demo check

14.4

Perform a Backup

To perform a backup of the PostgreSQL cluster run pgBackRest with the backup command on the repository host.

repository ⇒ Backup the demo cluster

sudo -u pgbackrest pgbackrest --stanza=demo backup

P00   WARN: no prior backup exists, incr backup has been changed to full

Since a new repository was created on the repository host the warning about the incremental backup changing to a full backup was emitted.

14.5

Restore a Backup

To perform a restore of the PostgreSQL cluster run pgBackRest with the restore command on the database host.

pg-primary ⇒ Stop the demo cluster, restore, and restart PostgreSQL

sudo systemctl stop postgresql-9.6.service

sudo -u postgres pgbackrest --stanza=demo --delta restore

sudo systemctl start postgresql-9.6.service

A new backup must be performed due to the timeline switch.

repository ⇒ Backup the demo cluster

sudo -u pgbackrest pgbackrest --stanza=demo backup

Parallel Backup / Restore

pgBackRest offers parallel processing to improve performance of compression and transfer. The number of processes to be used for this feature is set using the --process-max option.

It is usually best not to use more than 25% of available CPUs for the backup command. Backups don't have to run that fast as long as they are performed regularly and the backup process should not impact database performance, if at all possible.

The restore command can and should use all available CPUs because during a restore the PostgreSQL cluster is shut down and there is generally no other important work being done on the host. If the host contains multiple clusters then that should be considered when setting restore parallelism.

repository ⇒ Perform a backup with single process

sudo -u pgbackrest pgbackrest --stanza=demo --type=full backup

repository:/etc/pgbackrest/pgbackrest.conf ⇒ Configure pgBackRest to use multiple backup processes

[demo]
pg1-host=pg-primary
pg1-path=/var/lib/pgsql/9.6/data

[global]
process-max=3
repo1-path=/var/lib/pgbackrest
repo1-retention-full=2
start-fast=y

repository ⇒ Perform a backup with multiple processes

sudo -u pgbackrest pgbackrest --stanza=demo --type=full backup

repository ⇒ Get backup info for the demo cluster

sudo -u pgbackrest pgbackrest info

stanza: demo
    status: ok
    cipher: none

    db (current)
        wal archive min/max (9.6-1): 00000008000000000000001E/00000008000000000000001F

        full backup: 20190625-014744F

            timestamp start/stop: 2019-06-25 01:47:44 / 2019-06-25 01:47:59

            wal start/stop: 00000008000000000000001E / 00000008000000000000001E
            database size: 27.8MB, backup size: 27.8MB
            repository size: 3.2MB, repository backup size: 3.2MB

        full backup: 20190625-014801F

            timestamp start/stop: 2019-06-25 01:48:01 / 2019-06-25 01:48:10

            wal start/stop: 00000008000000000000001F / 00000008000000000000001F
            database size: 27.8MB, backup size: 27.8MB
            repository size: 3.2MB, repository backup size: 3.2MB

The performance of the last backup should be improved by using multiple processes. For very small backups the difference may not be very apparent, but as the size of the database increases so will time savings.

Starting and Stopping

Sometimes it is useful to prevent pgBackRest from running on a system. For example, when failing over from a primary to a standby it's best to prevent pgBackRest from running on the old primary in case PostgreSQL gets restarted or can't be completely killed. This will also prevent pgBackRest from running on cron.

pg-primary ⇒ Stop the pgBackRest services

sudo -u postgres pgbackrest stop

New pgBackRest processes will no longer run.

repository ⇒ Attempt a backup

sudo -u pgbackrest pgbackrest --stanza=demo backup

P00  ERROR: [062]: raised from remote process on 'pg-primary': stop file exists for all stanzas

Specify the --force option to terminate any pgBackRest process that are currently running. If pgBackRest is already stopped then stopping again will generate a warning.

pg-primary ⇒ Stop the pgBackRest services again

sudo -u postgres pgbackrest stop

P00   WARN: stop file already exists for all stanzas

Start pgBackRest processes again with the start command.

pg-primary ⇒ Start the pgBackRest services

sudo -u postgres pgbackrest start

It is also possible to stop pgBackRest for a single stanza.

pg-primary ⇒ Stop pgBackRest services for the demo stanza

sudo -u postgres pgbackrest --stanza=demo stop

New pgBackRest processes for the specified stanza will no longer run.

repository ⇒ Attempt a backup

sudo -u pgbackrest pgbackrest --stanza=demo backup

P00  ERROR: [062]: raised from remote process on 'pg-primary': stop file exists for stanza demo

The stanza must also be specified when starting the pgBackRest processes for a single stanza.

pg-primary ⇒ Start the pgBackRest services for the demo stanza

sudo -u postgres pgbackrest --stanza=demo start

Replication

Replication allows multiple copies of a PostgreSQL cluster (called standbys) to be created from a single primary. The standbys are useful for balancing reads and to provide redundancy in case the primary host fails.

17.1

Installation

A new host named pg-standby is created to run the standby.

pgBackRest needs to be installed from a package or installed manually as shown here.

pg-standby ⇒ Copy pgBackRest binary from build host

sudo scp build:/root/pgbackrest-release-2.15/src/pgbackrest /usr/bin

sudo chmod 755 /usr/bin/pgbackrest

pgBackRest contains embedded Perl which requires some additional modules.

pg-standby ⇒ Install required Perl packages

sudo yum install perl perl-Time-HiRes perl-Digest-SHA perl-DBD-Pg perl-JSON-PP

Finally, pgBackRest requires log and configuration directories and a configuration file.

pg-standby ⇒ Create pgBackRest configuration file and directories

sudo mkdir -p -m 770 /var/log/pgbackrest

sudo chown postgres:postgres /var/log/pgbackrest

sudo mkdir -p /etc/pgbackrest

sudo mkdir -p /etc/pgbackrest/conf.d

sudo touch /etc/pgbackrest/pgbackrest.conf

sudo chmod 640 /etc/pgbackrest/pgbackrest.conf

sudo chown postgres:postgres /etc/pgbackrest/pgbackrest.conf

17.2

Setup Passwordless SSH

pgBackRest requires passwordless SSH to enable communication between the hosts.

pg-standby ⇒ Create pg-standby host key pair

sudo -u postgres mkdir -m 750 -p /var/lib/pgsql/.ssh

sudo -u postgres ssh-keygen -f /var/lib/pgsql/.ssh/id_rsa \
       -t rsa -b 4096 -N ""

Exchange keys between repository and pg-standby.

repository ⇒ Copy pg-standby public key to repository

(echo -n 'no-agent-forwarding,no-X11-forwarding,no-port-forwarding,' && \
       echo -n 'command="/usr/bin/pgbackrest ${SSH_ORIGINAL_COMMAND#* }" ' && \
       sudo ssh root@pg-standby cat /var/lib/pgsql/.ssh/id_rsa.pub) | \
       sudo -u pgbackrest tee -a /home/pgbackrest/.ssh/authorized_keys

pg-standby ⇒ Copy repository public key to pg-standby

(echo -n 'no-agent-forwarding,no-X11-forwarding,no-port-forwarding,' && \
       echo -n 'command="/usr/bin/pgbackrest ${SSH_ORIGINAL_COMMAND#* }" ' && \
       sudo ssh root@repository cat /home/pgbackrest/.ssh/id_rsa.pub) | \
       sudo -u postgres tee -a /var/lib/pgsql/.ssh/authorized_keys

Test that connections can be made from repository to pg-standby and vice versa.

repository ⇒ Test connection from repository to pg-standby

sudo -u pgbackrest ssh postgres@pg-standby

pg-standby ⇒ Test connection from pg-standby to repository

sudo -u postgres ssh pgbackrest@repository

17.3

Hot Standby

A hot standby performs replication using the WAL archive and allows read-only queries.

pgBackRest configuration is very similar to pg-primary except that the standby_mode setting will be enabled to keep the cluster in recovery mode when the end of the WAL stream has been reached.

pg-standby:/etc/pgbackrest/pgbackrest.conf ⇒ Configure pgBackRest on the standby

[demo]
pg1-path=/var/lib/pgsql/9.6/data
recovery-option=standby_mode=on

[global]
log-level-file=detail
repo1-host=repository

Create the path where PostgreSQL will be restored.

pg-standby ⇒ Create PostgreSQL path

sudo -u postgres mkdir -p -m 700 /var/lib/pgsql/9.6/data

Now the standby can be created with the restore command.

pg-standby ⇒ Restore the demo standby cluster

sudo -u postgres pgbackrest --stanza=demo restore

sudo -u postgres cat /var/lib/pgsql/9.6/data/recovery.conf

standby_mode = 'on'
restore_command = 'pgbackrest --stanza=demo archive-get %f "%p"'

NOTE:

The standby_mode setting has been written into the recovery.conf file. Configuring recovery settings in pgBackRest means that the recovery.conf file does not need to be stored elsewhere since it will be properly recreated with each restore. The --type=preserve option can be used with the restore to leave the existing recovery.conf file in place if that behavior is preferred.

The hot_standby setting must be enabled before starting PostgreSQL to allow read-only connections on pg-standby. Otherwise, connection attempts will be refused. The rest of the configuration is in case the standby is promoted to a primary.

pg-standby:/var/lib/pgsql/9.6/data/postgresql.conf ⇒ Configure PostgreSQL

archive_command = 'pgbackrest --stanza=demo archive-push %p'
archive_mode = on
hot_standby = on
log_filename = 'postgresql.log'
log_line_prefix = ''
max_wal_senders = 3
wal_level = replica

pg-standby ⇒ Start PostgreSQL

sudo systemctl start postgresql-9.6.service

The PostgreSQL log gives valuable information about the recovery. Note especially that the cluster has entered standby mode and is ready to accept read-only connections.

pg-standby ⇒ Examine the PostgreSQL log output for log messages indicating success

sudo -u postgres cat /var/lib/pgsql/9.6/data/pg_log/postgresql.log

LOG:  database system was interrupted; last known up at 2019-06-25 01:48:02 UTC

LOG:  entering standby mode

LOG:  restored log file "00000008.history" from archive
LOG:  restored log file "00000008000000000000001F" from archive
LOG:  redo starts at 0/1F000028
LOG:  consistent recovery state reached at 0/1F000130

LOG:  database system is ready to accept read only connections

An easy way to test that replication is properly configured is to create a table on pg-primary.

pg-primary ⇒ Create a new table on the primary

sudo -u postgres psql -c " \
       begin; \
       create table replicated_table (message text); \
       insert into replicated_table values ('Important Data'); \
       commit; \
       select * from replicated_table";

    message
----------------

 Important Data

(1 row)

And then query the same table on pg-standby.

pg-standby ⇒ Query new table on the standby

sudo -u postgres psql -c "select * from replicated_table;"

ERROR:  relation "replicated_table" does not exist

LINE 1: select * from replicated_table;
                      ^

So, what went wrong? Since PostgreSQL is pulling WAL segments from the archive to perform replication, changes won't be seen on the standby until the WAL segment that contains those changes is pushed from pg-primary.

This can be done manually by calling pg_switch_xlog() which pushes the current WAL segment to the archive (a new WAL segment is created to contain further changes).

pg-primary ⇒ Call pg_switch_xlog()

sudo -u postgres psql -c "select *, current_timestamp from pg_switch_xlog()";

 pg_switch_xlog |              now
----------------+-------------------------------
 0/20022268     | 2019-06-25 01:48:58.731098+00
(1 row)

Now after a short delay the table will appear on pg-standby.

pg-standby ⇒ Now the new table exists on the standby (may require a few retries)

sudo -u postgres psql -c " \
       select *, current_timestamp from replicated_table"

    message     |              now
----------------+-------------------------------

 Important Data | 2019-06-25 01:49:00.677035+00

(1 row)

Check the standby configuration for access to the repository.

pg-standby ⇒ Check the configuration

sudo -u postgres pgbackrest --stanza=demo --log-level-console=info check

P00   INFO: check command begin 2.15: --log-level-console=info --log-level-file=detail --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --repo1-host=repository --stanza=demo

P00   INFO: switch xlog cannot be performed on the standby, all other checks passed successfully

P00   INFO: check command end: completed successfully

17.4

Streaming Replication

Instead of relying solely on the WAL archive, streaming replication makes a direct connection to the primary and applies changes as soon as they are made on the primary. This results in much less lag between the primary and standby.

Streaming replication requires a user with the replication privilege.

pg-primary ⇒ Create replication user

sudo -u postgres psql -c " \
       create user replicator password 'jw8s0F4' replication";

CREATE ROLE

The pg_hba.conf file must be updated to allow the standby to connect as the replication user. Be sure to replace the IP address below with the actual IP address of your pg-primary. A reload will be required after modifying the pg_hba.conf file.

pg-primary ⇒ Create pg_hba.conf entry for replication user

sudo -u postgres sh -c 'echo \
       "host    replication     replicator      172.17.0.6/32           md5" \
       >> /var/lib/pgsql/9.6/data/pg_hba.conf'

sudo systemctl reload postgresql-9.6.service

The standby needs to know how to contact the primary so the primary_conninfo setting will be configured in pgBackRest.

pg-standby:/etc/pgbackrest/pgbackrest.conf ⇒ Set primary_conninfo

[demo]
pg1-path=/var/lib/pgsql/9.6/data
recovery-option=standby_mode=on
recovery-option=primary_conninfo=host=172.17.0.4 port=5432 user=replicator

[global]
log-level-file=detail
repo1-host=repository

It is possible to configure a password in the primary_conninfo setting but using a .pgpass file is more flexible and secure.

pg-standby ⇒ Configure the replication password in the .pgpass file.

sudo -u postgres sh -c 'echo \
       "172.17.0.4:*:replication:replicator:jw8s0F4" \
       >> /var/lib/pgsql/.pgpass'

sudo -u postgres chmod 600 /var/lib/pgsql/.pgpass

Now the standby can be created with the restore command.

pg-standby ⇒ Stop PostgreSQL and restore the demo standby cluster

sudo systemctl stop postgresql-9.6.service

sudo -u postgres pgbackrest --stanza=demo --delta restore

sudo -u postgres cat /var/lib/pgsql/9.6/data/recovery.conf

primary_conninfo = 'host=172.17.0.4 port=5432 user=replicator'
standby_mode = 'on'
restore_command = 'pgbackrest --stanza=demo archive-get %f "%p"'

By default RHEL/CentOS 7 stores the postgresql.conf file in the PostgreSQL data directory. That means the change made to postgresql.conf was overwritten by the last restore and the hot_standby setting must be enabled again. Other solutions to this problem are to store the postgresql.conf file elsewhere or to enable the hot_standby setting on the pg-primary host where it will be ignored.

pg-standby:/var/lib/pgsql/9.6/data/postgresql.conf ⇒ Enable hot_standby

archive_command = 'pgbackrest --stanza=demo archive-push %p'
archive_mode = on
hot_standby = on
log_filename = 'postgresql.log'
log_line_prefix = ''
max_wal_senders = 3
wal_level = replica

pg-standby ⇒ Start PostgreSQL

sudo systemctl start postgresql-9.6.service

The PostgreSQL log will confirm that streaming replication has started.

pg-standby ⇒ Examine the PostgreSQL log output for log messages indicating success

sudo -u postgres cat /var/lib/pgsql/9.6/data/pg_log/postgresql.log

       [filtered 6 lines of output]
LOG:  database system is ready to accept read only connections
LOG:  restored log file "000000080000000000000020" from archive

LOG:  started streaming WAL from primary at 0/21000000 on timeline 8

Now when a table is created on pg-primary it will appear on pg-standby quickly and without the need to call pg_switch_xlog().

pg-primary ⇒ Create a new table on the primary

sudo -u postgres psql -c " \
       begin; \
       create table stream_table (message text); \
       insert into stream_table values ('Important Data'); \
       commit; \
       select *, current_timestamp from stream_table";

    message     |              now
----------------+-------------------------------

 Important Data | 2019-06-25 01:49:15.642895+00

(1 row)

pg-standby ⇒ Query table on the standby

sudo -u postgres psql -c " \
       select *, current_timestamp from stream_table"

    message     |              now
----------------+-------------------------------

 Important Data | 2019-06-25 01:49:16.383919+00

(1 row)

Asynchronous Archiving

Asynchronous archiving is enabled with the archive-async option. This option enables asynchronous operation for both the archive-push and archive-get commands.

A spool path is required. The commands will store transient data here but each command works quite a bit differently so spool path usage is described in detail in each section.

pg-primary ⇒ Create the spool directory

sudo mkdir -p -m 750 /var/spool/pgbackrest

sudo chown postgres:postgres /var/spool/pgbackrest

pg-standby ⇒ Create the spool directory

sudo mkdir -p -m 750 /var/spool/pgbackrest

sudo chown postgres:postgres /var/spool/pgbackrest

The spool path must be configured and asynchronous archiving enabled. Asynchronous archiving automatically confers some benefit by reducing the number of connections made to remote storage, but setting process-max can drastically improve performance by parallelizing operations. Be sure not to set process-max so high that it affects normal database operations.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Configure the spool path and asynchronous archiving

[demo]
pg1-path=/var/lib/pgsql/9.6/data

[global]
archive-async=y
log-level-file=detail
repo1-host=repository
spool-path=/var/spool/pgbackrest

[global:archive-get]
process-max=2

[global:archive-push]
process-max=2

pg-standby:/etc/pgbackrest/pgbackrest.conf ⇒ Configure the spool path and asynchronous archiving

[demo]
pg1-path=/var/lib/pgsql/9.6/data
recovery-option=standby_mode=on
recovery-option=primary_conninfo=host=172.17.0.4 port=5432 user=replicator

[global]
archive-async=y
log-level-file=detail
repo1-host=repository
spool-path=/var/spool/pgbackrest

[global:archive-get]
process-max=2

[global:archive-push]
process-max=2

NOTE:

process-max is configured using command sections so that the option is not used by backup and restore. This also allows different values for archive-push and archive-get.

For demonstration purposes streaming replication will be broken to force PostgreSQL to get WAL using the restore_command.

pg-primary ⇒ Break streaming replication by changing the replication password

sudo -u postgres psql -c "alter user replicator password 'bogus'"

ALTER ROLE

pg-standby ⇒ Restart standby to break connection

sudo systemctl restart postgresql-9.6.service

18.1

Archive Push

The asynchronous archive-push command offloads WAL archiving to a separate process (or processes) to improve throughput. It works by looking ahead to see which WAL segments are ready to be archived beyond the request that PostgreSQL is currently making via the archive_command. WAL segments are transferred to the archive directly from the pg_xlog/pg_wal directory and success is only returned by the archive_command when the WAL segment has been safely stored in the archive.

The spool path holds the current status of WAL archiving. Status files written into the spool directory are typically zero length and should consume a minimal amount of space (a few MB at most) and very little IO. All the information in this directory can be recreated so it is not necessary to preserve the spool directory if the cluster is moved to new hardware.

IMPORTANT:

In the original implementation of asynchronous archiving, WAL segments were copied to the spool directory before compression and transfer. The new implementation copies WAL directly from the pg_xlog directory. If asynchronous archiving was utilized in v1.12 or prior, read the v1.13 release notes carefully before upgrading.

The [stanza]-archive-push-async.log file can be used to monitor the activity of the asynchronous process. A good way to test this is to quickly push a number of WAL segments.

pg-primary ⇒ Test parallel asynchronous archiving

sudo -u postgres psql -c " \
       select pg_create_restore_point('test async push'); select pg_switch_xlog(); \
       select pg_create_restore_point('test async push'); select pg_switch_xlog(); \
       select pg_create_restore_point('test async push'); select pg_switch_xlog(); \
       select pg_create_restore_point('test async push'); select pg_switch_xlog(); \
       select pg_create_restore_point('test async push'); select pg_switch_xlog();"

sudo -u postgres pgbackrest --stanza=demo --log-level-console=info check

P00   INFO: check command begin 2.15: --log-level-console=info --log-level-file=detail --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --repo1-host=repository --stanza=demo

P00   INFO: WAL segment 000000080000000000000026 successfully stored in the archive at '/var/lib/pgbackrest/archive/demo/9.6-1/0000000800000000/000000080000000000000026-693b8d4fb09c1d2fd269f7da1b523d976bc61442.gz'

P00   INFO: check command end: completed successfully

Now the log file will contain parallel, asynchronous activity.

pg-primary ⇒ Check results in the log

sudo -u postgres cat /var/log/pgbackrest/demo-archive-push-async.log

-------------------PROCESS START-------------------
P00   INFO: archive-push-async command begin 2.15: [/var/lib/pgsql/9.6/data/pg_xlog] --log-level-console=off --log-level-file=detail --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --process-max=2 --repo1-host=repository --spool-path=/var/spool/pgbackrest --stanza=demo

P00   INFO: push 1 WAL file(s) to archive: 000000080000000000000021
P01 DETAIL: pushed WAL file '000000080000000000000021' to the archive

P00   INFO: archive-push-async command end: completed successfully

-------------------PROCESS START-------------------
P00   INFO: archive-push-async command begin 2.15: [/var/lib/pgsql/9.6/data/pg_xlog] --log-level-console=off --log-level-file=detail --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --process-max=2 --repo1-host=repository --spool-path=/var/spool/pgbackrest --stanza=demo

P00   INFO: push 4 WAL file(s) to archive: 000000080000000000000022...000000080000000000000025
P01 DETAIL: pushed WAL file '000000080000000000000022' to the archive
P02 DETAIL: pushed WAL file '000000080000000000000023' to the archive
P01 DETAIL: pushed WAL file '000000080000000000000024' to the archive
P02 DETAIL: pushed WAL file '000000080000000000000025' to the archive

P00   INFO: archive-push-async command end: completed successfully

-------------------PROCESS START-------------------
P00   INFO: archive-push-async command begin 2.15: [/var/lib/pgsql/9.6/data/pg_xlog] --log-level-console=off --log-level-file=detail --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --process-max=2 --repo1-host=repository --spool-path=/var/spool/pgbackrest --stanza=demo

P00   INFO: push 1 WAL file(s) to archive: 000000080000000000000026
P01 DETAIL: pushed WAL file '000000080000000000000026' to the archive

P00   INFO: archive-push-async command end: completed successfully

18.2

Archive Get

The asynchronous archive-get command maintains a local queue of WAL to improve throughput. If a WAL segment is not found in the queue it is fetched from the repository along with enough consecutive WAL to fill the queue. The maximum size of the queue is defined by archive-get-queue-max. Whenever the queue is less than half full more WAL will be fetched to fill it.

Asynchronous operation is most useful in environments that generate a lot of WAL or have a high latency connection to the repository storage (i.e., S3 or other object stores). In the case of a high latency connection it may be a good idea to increase process-max.

The [stanza]-archive-get-async.log file can be used to monitor the activity of the asynchronous process.

pg-standby ⇒ Check results in the log

sudo -u postgres cat /var/log/pgbackrest/demo-archive-get-async.log

-------------------PROCESS START-------------------
P00   INFO: archive-get-async command begin 2.15: [00000008000000000000001F, 000000080000000000000020, 000000080000000000000021, 000000080000000000000022, 000000080000000000000023, 000000080000000000000024, 000000080000000000000025, 000000080000000000000026] --log-level-console=off --log-level-file=detail --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --process-max=2 --repo1-host=repository --spool-path=/var/spool/pgbackrest --stanza=demo
P00   INFO: get 8 WAL file(s) from archive: 00000008000000000000001F...000000080000000000000026

P01 DETAIL: found 00000008000000000000001F in the archive
P02 DETAIL: found 000000080000000000000020 in the archive

P01 DETAIL: unable to find 000000080000000000000021 in the archive
P02 DETAIL: unable to find 000000080000000000000022 in the archive
       [filtered 20 lines of output]
P00   INFO: archive-get-async command begin 2.15: [000000080000000000000021, 000000080000000000000022, 000000080000000000000023, 000000080000000000000024, 000000080000000000000025, 000000080000000000000026, 000000080000000000000027, 000000080000000000000028] --log-level-console=off --log-level-file=detail --log-level-stderr=off --no-log-timestamp --pg1-path=/var/lib/pgsql/9.6/data --process-max=2 --repo1-host=repository --spool-path=/var/spool/pgbackrest --stanza=demo
P00   INFO: get 8 WAL file(s) from archive: 000000080000000000000021...000000080000000000000028

P02 DETAIL: found 000000080000000000000022 in the archive
P01 DETAIL: found 000000080000000000000021 in the archive
P01 DETAIL: found 000000080000000000000024 in the archive
P02 DETAIL: found 000000080000000000000023 in the archive
P01 DETAIL: found 000000080000000000000025 in the archive
P02 DETAIL: found 000000080000000000000026 in the archive

P01 DETAIL: unable to find 000000080000000000000027 in the archive
P02 DETAIL: unable to find 000000080000000000000028 in the archive
       [filtered 13 lines of output]

pg-primary ⇒ Fix streaming replication by changing the replication password

sudo -u postgres psql -c "alter user replicator password 'jw8s0F4'"

ALTER ROLE

Backup from a Standby

pgBackRest can perform backups on a standby instead of the primary. Standby backups require the pg-standby host to be configured and the backup-standby option enabled. If more than one standby is configured then the first running standby found will be used for the backup.

repository:/etc/pgbackrest/pgbackrest.conf ⇒ Configure pg2-host/pg2-host-user and pg2-path

[demo]
pg1-host=pg-primary
pg1-path=/var/lib/pgsql/9.6/data
pg2-host=pg-standby
pg2-path=/var/lib/pgsql/9.6/data

[global]
backup-standby=y
process-max=3
repo1-path=/var/lib/pgbackrest
repo1-retention-full=2
start-fast=y

Both the primary and standby databases are required to perform the backup, though the vast majority of the files will be copied from the standby to reduce load on the primary. The database hosts can be configured in any order. pgBackRest will automatically determine which is the primary and which is the standby.

repository ⇒ Backup the demo cluster from pg2

sudo -u pgbackrest pgbackrest --stanza=demo --log-level-console=detail backup

       [filtered 2 lines of output]
P00   INFO: execute non-exclusive pg_start_backup() with label "pgBackRest backup started at 2019-06-25 01:49:35": backup begins after the requested immediate checkpoint completes
P00   INFO: backup start archive = 000000080000000000000028, lsn = 0/28000028

P00   INFO: wait for replay on the standby to reach 0/28000028
P00   INFO: replay on the standby reached 0/280003C0, checkpoint 0/28000060
P01   INFO: backup file pg-primary:/var/lib/pgsql/9.6/data/global/pg_control (8KB, 0%) checksum 60d5a83dacc30005f648638b382a9cac1c98a83c
P01   INFO: backup file pg-primary:/var/lib/pgsql/9.6/data/pg_log/postgresql.log (5.5KB, 0%) checksum 1b8b1c4e74a42018e132de85a9d0f72c4917b2b0
P01   INFO: backup file pg-primary:/var/lib/pgsql/9.6/data/pg_hba.conf (4.1KB, 0%) checksum 1c0bca2f7f44d354aaaefb2c21fd1290b7379690

P02   INFO: backup file pg-standby:/var/lib/pgsql/9.6/data/base/12470/2608 (440KB, 21%) checksum 596500d5404a89dd47a1c8fc7474b90bb3c32121
P03   INFO: backup file pg-standby:/var/lib/pgsql/9.6/data/base/12470/1249 (360KB, 37%) checksum faf11ef681c050046d598703e8776eca3812ea7a
       [filtered 38 lines of output]

This incremental backup shows that most of the files are copied from the pg-standby host and only a few are copied from the pg-primary host.

pgBackRest creates a standby backup that is identical to a backup performed on the primary. It does this by starting/stopping the backup on the pg-primary host, copying only files that are replicated from the pg-standby host, then copying the remaining few files from the pg-primary host. This means that logs and statistics from the primary database will be included in the backup.

Upgrading PostgreSQL

Immediately after upgrading PostgreSQL to a newer major version, the pg-path for all pgBackRest configurations must be set to the new database location and the stanza-upgrade run on the repository host. If the database is offline use the --no-online option.

The following instructions are not meant to be a comprehensive guide for upgrading PostgreSQL, rather they outline the general process for upgrading a primary and standby with the intent of demonstrating the steps required to reconfigure pgBackRest. It is recommended that a backup be taken prior to upgrading.

pg-primary ⇒ Stop old cluster

sudo systemctl stop postgresql-9.6.service

Stop the old cluster on the standby since it will be restored from the newly upgraded cluster.

pg-standby ⇒ Stop old cluster

sudo systemctl stop postgresql-9.6.service

Create the new cluster and perform upgrade.

pg-primary ⇒ Create new cluster and perform the upgrade

sudo -u postgres /usr/pgsql-10/bin/initdb \
       -D /var/lib/pgsql/10/data -k -A peer

sudo -u postgres sh -c 'cd /var/lib/pgsql && \
       /usr/pgsql-10/bin/pg_upgrade \
       --old-bindir=/usr/pgsql-9.6/bin \
       --new-bindir=/usr/pgsql-10/bin \
       --old-datadir=/var/lib/pgsql/9.6/data \
       --new-datadir=/var/lib/pgsql/10/data \
       --old-options=" -c config_file=/var/lib/pgsql/9.6/data/postgresql.conf" \
       --new-options=" -c config_file=/var/lib/pgsql/10/data/postgresql.conf"'

       [filtered 70 lines of output]
Checking for hash indexes                                   ok

Upgrade Complete

----------------
Optimizer statistics are not transferred by pg_upgrade so,
       [filtered 4 lines of output]

Configure the new cluster settings and port.

pg-primary:/var/lib/pgsql/10/data/postgresql.conf ⇒ Configure PostgreSQL

archive_command = 'pgbackrest --stanza=demo archive-push %p'
archive_mode = on
listen_addresses = '*'
log_line_prefix = ''
max_wal_senders = 3
port = 5432
wal_level = replica

Update the pgBackRest configuration on all systems to point to the new cluster.

pg-primary:/etc/pgbackrest/pgbackrest.conf ⇒ Upgrade the pg1-path

[demo]
pg1-path=/var/lib/pgsql/10/data

[global]
archive-async=y
log-level-file=detail
repo1-host=repository
spool-path=/var/spool/pgbackrest

[global:archive-get]
process-max=2

[global:archive-push]
process-max=2

pg-standby:/etc/pgbackrest/pgbackrest.conf ⇒ Upgrade the pg-path

[demo]
pg1-path=/var/lib/pgsql/10/data
recovery-option=standby_mode=on
recovery-option=primary_conninfo=host=172.17.0.4 port=5432 user=replicator

[global]
archive-async=y
log-level-file=detail
repo1-host=repository
spool-path=/var/spool/pgbackrest

[global:archive-get]
process-max=2

[global:archive-push]
process-max=2

repository:/etc/pgbackrest/pgbackrest.conf ⇒ Upgrade pg1-path and pg2-path, disable backup from standby

[demo]
pg1-host=pg-primary
pg1-path=/var/lib/pgsql/10/data
pg2-host=pg-standby
pg2-path=/var/lib/pgsql/10/data

[global]
backup-standby=n
process-max=3
repo1-path=/var/lib/pgbackrest
repo1-retention-full=2
start-fast=y

pg-primary ⇒ Copy hba configuration

sudo cp /var/lib/pgsql/9.6/data/pg_hba.conf \
       /var/lib/pgsql/10/data/pg_hba.conf

Before starting the new cluster, the stanza-upgrade command must be run on the server where the pgBackRest repository is located.

repository ⇒ Upgrade the stanza

sudo -u pgbackrest pgbackrest --stanza=demo --no-online \
       --log-level-console=info stanza-upgrade

P00   INFO: stanza-upgrade command begin 2.15: --no-backup-standby --log-level-console=info --log-level-stderr=off --no-log-timestamp --no-online --pg1-host=pg-primary --pg2-host=pg-standby --pg1-path=/var/lib/pgsql/10/data --pg2-path=/var/lib/pgsql/10/data --repo1-path=/var/lib/pgbackrest --stanza=demo

P00   INFO: stanza-upgrade command end: completed successfully

Start the new cluster and confirm it is successfully installed.

pg-primary ⇒ Start new cluster

sudo systemctl start postgresql-10.service

Test configuration using the check command.

pg-primary ⇒ Check configuration

sudo -u postgres systemctl status postgresql-10.service

● postgresql-10.service - PostgreSQL 10 database server
   Loaded: loaded (/usr/lib/systemd/system/postgresql-10.service; disabled; vendor preset: disabled)
   Active: active (running) since Tue 2019-06-25 01:49:57 UTC; 421ms ago
     Docs: https://www.postgresql.org/docs/10/static/
  Process: 3986 ExecStartPre=/usr/pgsql-10/bin/postgresql-10-check-db-dir ${PGDATA} (code=exited, status=0/SUCCESS)
 Main PID: 3991 (postmaster)
   CGroup: /docker/22adc956e711d039bce132beac84beb0782d62eebef745e5325d50ee2fffb506/system.slice/postgresql-10.service
           ├─3991 /usr/pgsql-10/bin/postmaster -D /var/lib/pgsql/10/data/
           ├─3992 postgres: logger process
           ├─3994 postgres: checkpointer process
           ├─3995 postgres: writer process
           ├─3996 postgres: wal writer process
           ├─3997 postgres: autovacuum launcher process
           ├─3998 postgres: archiver process
           ├─3999 postgres: stats collector process
           └─4000 postgres: bgworker: logical replication launcher

sudo -u postgres pgbackrest --stanza=demo check

Remove the old cluster.

pg-primary ⇒ Remove old cluster

sudo rm -rf /var/lib/pgsql/9.6/data

Install the new PostgreSQL binaries on the standby and create the cluster.

pg-standby ⇒ Remove old cluster and create the new cluster

sudo rm -rf /var/lib/pgsql/9.6/data

sudo -u postgres mkdir -p -m 700 /usr/pgsql-10/bin

Run the check on the repository host. The warning regarding the standby being down is expected since the standby cluster is down. Running this command demonstrates that the repository server is aware of the standby and is configured properly for the primary server.

repository ⇒ Check configuration

sudo -u pgbackrest pgbackrest --stanza=demo check

Run a full backup on the new cluster and then restore the standby from the backup. The backup type will automatically be changed to full if incr or diff is requested.

repository ⇒ Run a full backup

sudo -u pgbackrest pgbackrest --stanza=demo --type=full backup

pg-standby ⇒ Restore the demo standby cluster

sudo -u postgres pgbackrest --stanza=demo restore

pg-standby:/var/lib/pgsql/10/data/postgresql.conf ⇒ Configure PostgreSQL

hot_standby = on

pg-standby ⇒ Start PostgreSQL and check the pgBackRest configuration

sudo systemctl start postgresql-10.service

sudo -u postgres pgbackrest --stanza=demo check

Backup from standby can be enabled now that the standby is restored.

repository:/etc/pgbackrest/pgbackrest.conf ⇒ Reenable backup from standby

[demo]
pg1-host=pg-primary
pg1-path=/var/lib/pgsql/10/data
pg2-host=pg-standby
pg2-path=/var/lib/pgsql/10/data

[global]
backup-standby=y
process-max=3
repo1-path=/var/lib/pgbackrest
repo1-retention-full=2
start-fast=y