> Software > Databases > Azure

Microsoft Azure Database for PostgreSQL

You can use the Commvault software to back up and restore PostgreSQL data on the Azure Database for PostgreSQL Single Servers and Azure Database for PostgreSQL Flexible Servers.

Note

  • The recommended maximum size for an Azure PostgreSQL database is 2 TB. Daily backups are not supported for databases larger than 2 TB. For information about handling larger databases, see Azure Best practices for Azure Database.

  • Commvault supports pgvector extension for Azure PostgreSQL. The destination instance should also have pgvector extension in case of Out-of-Place restore.

To perform PostgreSQL cloud backups from replica, consider the following recommendation:

  1. For PostgreSQL master-replica configurations:

    1. Use the master node for backups to avoid potential conflicts with the recovery process on replicas, which might interfere with tools such as pg_dump. Configuring backups on the master node introduces some overhead, as the Commvault access node connects directly to the master database to perform pg_dump.

    2. If using a replica for backups, ensure to configure max_standby_archive_delay and max_standby_streaming_delay settings to allow sufficient time for the Commvault triggered pg_dump to complete without being canceled due to WAL replay conflicts. For more information, see PostgreSQL Hot Standby Conflicts.

  2. Schedule backups during off-peak or quiet hours

    • Schedule backups during off-peak or quiet hours to ensure faster execution and minimize the risk of connection interruptions or resource contention.

Backups

Backups You Can Perform

  • Full backups

When You Can Perform Backups

  • On a schedule: The server plan that you assign manages scheduled backups

  • On demand: You can perform on-demand backups at any time

Data You Can Back Up

  • PostgreSQL user databases

    Note

    • For Azure database for PostgreSQL Flexible Servers, by default, the pg_cron extension is enabled and the schema that is named as cron is created in the postgres database where the pg_cron metadata is saved. Since the Azure Server administrator will not have the privilege to dump/drop/restore the schema, the postgres database backups fail. Hence, the postgres database backup will be skipped from the backups of Azure PostgreSQL Flexible Servers.

    • The following databases will be skipped from the backups of:

      • Azure PostgreSQL Flexible Servers - azure_sys, azure_maintenance, template0, template1, postgres

      • Azure single servers - azure_sys, azure_maintenance, template0, template1

    • If cron schema is set to a different database, you must create a new subclient in order to skip the database from backups.

Restores

Backups You Can Use for Restores

  • The most recent backup: For example, restore the most recent backup to its original location

  • A backup from a specific date: For example, restore data to a point in time before it became unusable

  • Backups from a date range: For example, restore data that was accidentally deleted

Destinations You Can Restore To

  • In-place: Restore data to the location that it was backed up from

  • Out-of-place: Restore data to the following destinations:

    • Cross-cloud: You can restore PostgreSQL databases from one cloud platform (such as Amazon Web Services and Google Cloud Platform) to a different cloud platform. The source computer and destination computer should have the same operating system.

    • An instance in the same platform or a different platform (such as Alibaba Cloud, Amazon Web Services, Azure, and Google Cloud Platform) across regions

    • An on-premises instance

    • A disk location

×

Loading...