{[project]} Frequently Asked Questions (FAQ).
Introduction

Frequently Asked Questions are intended to provide details for specific questions that may or may not be covered in the User Guide, Configuration, or Command reference. If you are unable to find details for your specific issue here, remember that the Issues List in GitHub is also a valuable resource.

What if I get the <quote>could not find WAL segment</quote> error?

The cause of this error can be a result of many different issues, some of which may be:

misconfigured archive_command misconfigured configuration files network or permissions issue third party product (e.g. S3, Swift or Minio) configuration issue large amount of WAL queueing to be archived

It is advisable to:

check the archive_command in check the configuration settings on each host (e.g. pg* settings are set on the repository host and repo* settings on the pg host) run the check command with {[dash]}-archive-timeout set to a higher value than in the configuration file (or default) to see if the WAL queue needs more time to clear. If the system is generating a lot of WAL, then consider configuring asynchronous archiving
How do I manually purge a backup set?

A full backup set can be expired using the {[dash]}-set option as explained in Command Reference: Expire.

How can I configure options independently for each command?

has the ability to set options independently in the configuration file for each command. Configure Cluster Stanza details this feature as well as option precedence.

For example, the process-max option can be optimized for each command:

[global] # used where not overridden process-max=2 [global:backup] # more cores for backup process-max=4 [global:restore] # all the cores for restore process-max=8 [global:archive-push] # more cores for archive-push process-max=3 [global:archive-get] # fewer cores for archive-get process-max=1
Can I use dots (periods) in my S3 bucket name?

RFC-2818 does not allow wildcards to match on a dot (.) so s3 bucket names must not contain dots. If there are dots in the S3 bucket name then an error such as unable to find hostname 'my.backup.bucket.s3.amazonaws.com' in certificate common name or subject alternative names will occur.

Where can I find packages for older versions of <backrest/>?

The apt.postgresql.org repository maintains an archive of older versions. Debian also maintains snapshots of all test builds.

Why does a backup attempt fail when <br-option>backup-standby=y</br-option> and the standby database is down?

Configuring backup from standby is generally intended to reduce load on the primary, so switching backups to the primary when the standby is down often defeats the point. Putting more load on the primary in a situation where there are already failures in the system is not recommended. Backups are not critical as long as you have one that is fairly recent -- the important thing is to keep up with WAL archiving. There is plenty of time to get a backup when the system is stable again.

If you really need a backup, the solution is to have more standbys or remove backup-standby. This can be overridden on the command line with --no-backup-standby, so there is no need to reconfigure for a one-off backup.

Should I setup my repository on a standby host?

No. When primary and standby databases are configured, the configuration files should be symmetric in order to seamlessly handle failovers. If they are not, the configurations will need to be changed on failover or further problems may result.

See the Dedicated Repository Host section of the User Guide for more information.

Time-based Point-in-Time Recovery does not appear to work, why?

The most common mistake when using time-based Point-in-Time Recovery is forgetting to choose a backup set that is before the target time. will attempt to discover a backup to play forward from the time specified by the --target= if the --set option is not specified. If a backup set cannot be found, then restore will default to the latest backup. However, if the latest backup is after the target time, then --target= is not considered valid by and is therefore ignored, resulting in WAL recovery to the latest time available.

To use the --set option, choose a backup set by running the info command and finding the backup with a timestamp stop that is before the target time. Then when running the restore, specify the option --set=BACKUP_LABEL where BACKUP_LABEL is the chosen backup set.

See the Point-in-Time Recovery section of the User Guide for more information.

What does the WAL archive suffix mean?

The suffix is the SHA1 checksum used to verify file integrity. There is no way to omit it.

Does it take longer to restore specific backup types (full, differential, incremental)?

The various backup types require the same amount of time to restore. Restore retrieves files based on the backup manifest, which may reference files from a previous backup in the case of incremental or differential backups. While there could be differences in time spent making a given backup (depending on backup type), database size determines restore time (disk I/O, network I/O, etc. being equal).