The DR Series system design includes an online data integrity verification feature known as Data Check, which checks for potential or unexpected data inconsistencies in the data store associated with the internal system deduplication engine. Data Check performs a series of checks for unexpected data inconsistencies as early as possible in the data ingest and backup process.
Data Check checks and verifies data both during the write process and also the data already stored on the system disks. The design purpose is to detect potential issues early enough in the data management process so that original data can be used to backup and correct any potential data inconsistencies. Data Check reports data verification issues, but it is not intended nor designed to repair these issues itself.
Any data inconsistencies that are encountered are reported as DR Series system alerts, and these filesystem errors can be repaired using the Maintenance mode (for more information, see Maintenance.
The Data Check feature runs continuously except for when the DR Series system enters its Maintenance mode (it does not run while the system is in this mode). Data Check leaves the system in an Operational mode when it detects an error, at which point, it sends an alert and an event.
If an alert has already been sent, but has not been cleared (for example, when repairs occur during the Maintenance mode), no new event is sent. Similarly, for events, one is sent for the first detected data inconsistency, and then the total number of issues detected during the scan are listed in a new event.
If Data Check is enabled, it runs in the background as a low-priority process, and changes to an idle state when the other major DR Series system operations (data ingest, replication, and cleaner) are active.
|
NOTE: Unless otherwise noted, all later references to datacheck or Data Check in this guide are used interchangeably to represent the Data Check feature in the DR Series system. |