|
| 1 | +--- |
| 2 | +title: Backup and restore – Hyperscale (Citus) - Azure Database for PostgreSQL |
| 3 | +description: Protecting data from accidental corruption or deletion |
| 4 | +author: jonels-msft |
| 5 | +ms.author: jonels |
| 6 | +ms.service: postgresql |
| 7 | +ms.subservice: hyperscale-citus |
| 8 | +ms.topic: conceptual |
| 9 | +ms.date: 04/14/2020 |
| 10 | +--- |
| 11 | + |
| 12 | +# Backup and restore in Azure Database for PostgreSQL - Hyperscale (Citus) |
| 13 | + |
| 14 | +Azure Database for PostgreSQL – Hyperscale (Citus) automatically creates |
| 15 | +backups of each node and stores them in locally redundant storage. Backups can |
| 16 | +be used to restore your Hyperscale (Citus) cluster to a point-in-time. Backup |
| 17 | +and restore are an essential part of any business continuity strategy because |
| 18 | +they protect your data from accidental corruption or deletion. |
| 19 | + |
| 20 | +## Backups |
| 21 | + |
| 22 | +Azure Database for PostgreSQL takes snapshot backups of the data files and the |
| 23 | +transaction log. Snapshot backups are taken at least once a day. These backups |
| 24 | +allow you to restore a server to any point-in-time within backup retention |
| 25 | +period. Currently all clusters have backup retention period is 35 days. All |
| 26 | +backups are encrypted using AES 256-bit encryption. |
| 27 | + |
| 28 | +In Azure regoins that support availability zones backup snapshots are stored in |
| 29 | +3 availability zones. It means that you would be able to restore your |
| 30 | +Hyperscale (Citus) cluster from backup as long as at least one availability |
| 31 | +zone is online. |
| 32 | + |
| 33 | +These backup files cannot be exported. The backups can only be used for restore |
| 34 | +operations in Azure Database for PostgreSQL. |
| 35 | + |
| 36 | +### Backup storage cost |
| 37 | + |
| 38 | +See Azure Database for PostgreSQL -Hyperscale (Citus) [pricing |
| 39 | +page](https://azure.microsoft.com/en-us/pricing/details/postgresql/hyperscale-citus/) |
| 40 | +for current backup storage pricing. |
| 41 | + |
| 42 | +## Restore |
| 43 | + |
| 44 | +In Azure Database for PostgreSQL, performing a restore creates a new Hyperscale |
| 45 | +(Citus) cluster from the original nodes’ backups. |
| 46 | + |
| 47 | +> [!IMPORTANT] |
| 48 | +> Deleted Hyperscale (Citus) clusters cannot be restored. If you delete the |
| 49 | +> cluster, all nodes that belong to the cluster are deleted and cannot be |
| 50 | +> recovered. To protect cluster resources, post deployment, from accidental |
| 51 | +> deletion or unexpected changes, administrators can leverage [management |
| 52 | +> locks](/azure/azure-resource-manager/management/lock-resources). |
| 53 | +
|
| 54 | +### Point-in-time restore (PITR) |
| 55 | + |
| 56 | +You can perform a restore to any point in time within 35 days. A new cluster is |
| 57 | +created in the same Azure region, subscription, and resource group as the |
| 58 | +original one. It is created with the original cluster's configuration for the |
| 59 | +number of node, number of vCores, storage size, user roles, and server |
| 60 | +parameters. |
| 61 | + |
| 62 | +Point-in-time restore is useful in multiple scenarios. For example, when a user |
| 63 | +accidentally deletes data, drops an important table or database, or if an |
| 64 | +application accidentally overwrites good data with bad data due to an |
| 65 | +application defect. |
| 66 | + |
| 67 | +> [!IMPORTANT] |
| 68 | +> You will need to open a support request to perform point-in-time restore of |
| 69 | +> your Hyperscale (Citus) cluster. |
| 70 | +
|
| 71 | +### Post-restore tasks |
| 72 | + |
| 73 | +After a restore from either recovery mechanism, you should perform the |
| 74 | +following tasks to get your users and applications back up and running: |
| 75 | + |
| 76 | +• If the new server is meant to replace the original server, redirect clients |
| 77 | + and client applications to the new server |
| 78 | +• Ensure appropriate server-level firewall and VNet rules are in place for |
| 79 | + users to connect. These rules are not copied over from the original server. |
| 80 | +• Ensure appropriate logins and database level permissions are in place |
| 81 | +• Configure alerts, as appropriate |
| 82 | + |
| 83 | +## Next steps |
| 84 | + |
| 85 | +• Learn how to restore using [the Azure |
| 86 | + portal](/azure/postgresql/howto-restore-server-portal). |
| 87 | +• Learn how to restore using [the Azure |
| 88 | + CLI](/azure/postgresql/howto-restore-server-cli). |
| 89 | +• To learn more about business continuity, see the [business continuity |
| 90 | + overview](/azure/postgresql/concepts-business-continuity). |
0 commit comments