Backup concept for server applications

Hosting and operation packages operated in //SEIBERT/MEDIA-managed data centers benefit from our two-tier backup concept. So you get a carefree setup and fail-safe systems.

This concept was primarily designed for Atlassian server applications. If you are interested in a backup concept for your data center applications, we will be happy to advise you individually.


Kontakt




What data is backed up?

Within our backup concept, all app data is backed up. In a typical Atlassian application, this could be a database (PostgreSQL), file attachments in Confluence, ticket attachments in Jira, or repositories in Bitbucket.

The configuration data of customer systems - for example, web server settings - is not captured by classic backups. Instead, they are covered by the global configuration management system of //SEIBERT/MEDIA.

This data is therefore stored in a git repository, which is stored in our own Bitbucket instance. Sensitive access data is encrypted there and only accessible to a limited group of people. Bitbucket itself is again subject to the same security mechanisms as those described above.






Two-level backup system - How to backup the data



Snapshots within the customer system

  • Customer systems use ZFS as a file system. Here, snapshots of the application data and the database on block-level are generated hourly.

  • By default, these snapshots are kept for the following periods:

    • 24 hourly snapshots

    • 7 daily snapshots
    • 5 weekly snapshots
  • Local snapshots of this kind do not help in the event of total system failure. Still, they do protect against short-term corruption of data, which can occur, for example, if projects or areas are accidentally deleted or changed.






Backups on outsourced data center

The backups are made to an outsourced data center at the provider Hetzner:

  • Here the application data and the database are backed up (binary format, no dump).
  • A local ZFS snapshot of the customer's system serves as a data source to create a consistent backup.
  • There is a daily backup. By default, these backups are kept for the following periods:
    • 7 daily backups
    • 4 weekly backups
    • 3 monthly backups
  • In the case of total system failure, these backups can be restored.
  • Our backups are tested every 60 days through recovery tests, whereby different failure and recovery scenarios are carried out.
  • The physical separation ensures a further reduction of risks (e.g., natural disasters), but European data protection guidelines are still being observed.




Backup-Server

Encryption

All data is transferred encrypted from the customer system to the backup server (SSH) and stored therein encrypted form.

Redundancy

The backup servers usually have three hard disks. If one hard disk fails, it can be replaced without loss of data.

Separation between customers

The backup spaces of different customers are separated on the file system level. Backup clients also have strictly regulated system rights.

Round Robin Backups

Today's backup is stored on server $A, tomorrow's backup on server $B; then, it continues on server $A. This distribution further increases reliability.


More about operation packages:

  • No labels

This content was last updated on 04/12/2022.

This content hasn't been updated in a while. That doesn't have to be a problem. Some of our pages live for years without becoming obsolete.

Old content can be incorrect, misleading or outdated. Please get in contact with us via a form on this page, our live chat or via email with content@seibert.group if you are in doubt, have a question, suggestion, or want changes from us.