Backup, restore, and disaster recovery

How to back up your analytics subsystem and restore after a disaster.

An analytics subsystem backup consists of:
Analytics infrastructure configuration backup
Your infrastructure configuration backup contains information such as endpoints, offload configuration, remote backup site credentials.

The infrastructure configuration backup is also referred to as the config backup.

On Kubernetes and OpenShift®, the config backup consists of YAML file backups of your analytics CR and certificates.

Your config backup is needed for disaster recovery, where you must redeploy your analytics subsystem.

Analytics database backup
Your database backup contains all your analytics summary data and shared queries. If you have storage capacity, you can configure your analytics database backups to include all your API event data. When you enable database backups, local backups are created at 55 minutes past each hour. You can configure your database backups to be periodically copied to a remote SFTP server or S3 compatible object-store. The configuration of your database backups, such as object-store credentials, are defined in your analytics infrastructure configuration. Therefore, to restore an analytics database backup after a disaster, you must first restore your config backup.
Your analytics database backups are intended for use in two scenarios:
  • Disaster recovery: Your entire analytics subsystem fails. You redeploy the subsystem (use your config backup to ensure that the same config settings and certificates are defined) and then restore your most recent database backup.
  • Corrupted index recovery: One or more analytics indices are corrupted, so you restore just those indices from a backup that contains them.
Analytics database backups are not intended for long-term storage of your analytics data. If you require long-term storage, use the offload or export features.

Review the backup, restore, and disaster recovery documentation for your platform: