Link Search Menu Expand Document

FTK Backups and Archive Policy

 

The Digital Archives program uses the software Forensic Toolkit (FTK) to process the majority of the Library’s born-digital acquisitions. For each collection a case is created in FTK and information regarding this case is stored in a PostGreSQL database. During processing, a backup of the case can be created which contains case information and a copy of the database files regarding that case. After processing, a case is archived. The case directory keeps the case information and the database files which are then deleted from the PostGresSQL database. Locally, we also include a compressed copy of the case’s evidence in our archive except for certain exceptions, noted below.

Backup of Case

  • Backups will be made daily when an archivist does significant work on a collection to preserve their work.

  • The previous backup will be deleted at this time.

  • There will always only be one backup per case in FTK unless a collection is particularly complex and Digital Preservation staff decides that multiple backups would be beneficial.

Archive of Case

  • A case will be archived when a collection has been processed. This means the archivist has finished the arrangement and description of the collection, the finding aid has been approved, and finding aid components have been exported.

  • After a case has been archived, all backups of that case will be deleted.

  • Compressed copies of the case evidence will be retained with the case. Exceptions to this rule include cases where the evidence exceeds 100 GB. In these cases, evidence will not be held with the archive, but rather accessed as needed from other backup sources.

  • An archive will be retained for a two year retention period and disposition will occur in quarterly increments. Exceptions to this rule include:

    • The collection contains media that could not be accessed, rendered, or otherwise fully arranged, described, and made accessible to researchers. An archive will be retained so that processing can be revisited at a later date.

    • Collections with lots of executables. We have found that arrangement and description of these collections often are revisited.

    • Exceptions will be stored in a separate directory and rationale provided in a README.txt file that will be stored with the archive.