Resources
Backup and restore procedures
A backup is the copying and archiving of data so it can be used to restore the original in the event of data loss - whether by accidental deletion or by corruption during writing, reading, storage, transmission or processing.
The purpose of backups is to recover data after its loss. Data deletion refers to users - accidentally or intentionally - removing data within an application such as a FileMaker database. Data corruption refers to errors that occur during writing, reading, storage, transmission or processing, introducing unintended changes to the original data.
Our hosting environment employs a number of technologies that automate backup routines at several levels: the application program, the operating system, and the server.
At the application level
FileMaker Server can define backup schedules to be performed on chosen databases via its Admin Console, and can define where those backups are saved. When a scheduled backup runs, FileMaker Server pauses the databases and then copies them to the defined backup folder - which can be on another drive or in any specified location.
Backups are preserved to ensure the integrity and protection of data, and at this level they are scheduled to occur:
- Daily, at midnight - the last 7 copies are kept and synced to AWS S3 storage at 1:40 am each day, so the last 7 midnight backups are held off-machine in AWS S3.
- Hourly, at 20 minutes past the hour from 6:20 am to 11:20 pm - the last 6 retained, on-machine only.
- Four-hourly, at 6:40 am, 10:40 am, 2:40 pm, 6:40 pm and 10:40 pm - the last 4 retained, on-machine only.
If a longer off-machine history is required, this can easily be changed using AWS S3 bucket settings and an adjustment to the copy script. If the hourly or four-hourly backups should be held off-machine, that can be scripted and scheduled, as can additional weekly or monthly backups to AWS S3. Please note that additional backups increase storage usage, which attracts additional cost.
Restore
Restoring a backup from AWS S3 is carried out through the AWS Management Console: Services → S3 → select the backup to be restored → select its checkbox → click Download and save the backup to a local computer. It can then be pushed back up to the server.
Any individual record, or selection of records, can be restored by import. The entire database can be "rolled back" to a point in time by replacing the current database with a backup copy on the server. The key is to inform The FM Studio of the need for restoration as soon as possible.