T 4.30 Loss of database integrity/consistency

A loss of database integrity or consistency means that data are still present in a database but have become partly corrupted or unintelligible. As a result, the data cannot be correctly processed any more. This database integrity and consistency can be impaired in a variety of ways, for example, through inadvertent data manipulations (e.g. unintentional modifications to data), inadequate checks of the synchronisation of transactions, and deliberate intrusions.

The following consequences can arise as a result:

• Certain tasks which rely on the correctness of the data in the database can no longer be performed fully or at all.

• The information in the database as a whole is corrupted.

• Considerable time and effort are required to restore the integrity and consistency of the database.

Depending on the factor responsible for the loss of database integrity/consistency, it can be difficult or even impossible to determine exactly which data were modified. This can lead to further financial losses and security risks.

Example:

To save space and time, a database file is created in the /tmp file system on a Unix server. This file is deleted subsequently during a nightly cron job, as a result of which the entire database becomes useless.