Chapter 9. Maintaining Units148 3ware Serial ATA RAID Controller User GuideHow Errors Are HandledVerification makes use of the same error checking and error repair techniquesused during ordinary use of drives configured through 3ware RAIDcontrollers.When verification encounters an error, the controller typically retries thecommand. If there are cable CRC errors, there may be multiple retriesincluding downgrade of the UDMA mode. If the error persists and isunrepairable (e.g., ECC errors), an error notification is issued to indicate theproblem. (See AEN “0026 Drive ECC error reported” on page 239.)If the disk drive is part of a redundant unit that is in a redundant state (notdegraded or rebuilding), then Dynamic Sector Repair automatically rewritesthe redundant data to the error location to force the drive to reallocate theerror location. A notification of repair is posted to the alarms list. The result isa restoration of drive and data integrity; the primary and redundant data areagain both valid.If the unit is not redundant, it is recommended that you perform a file-systemcheck to correct the issue. Under Windows, you can do this by right-clickingon the Drive and choosing Properties; then on the Tools tab, click Check Now.If the errors persist and cannot be overwritten from a backup copy, perform afinal backup of files that have changed since your last backup. You will needto replace the defective drive, recreate the array, and reinstall the data.Starting a Verify ManuallyVerification of units can be done automatically, on a schedule, or can bestarted manually, as described below. (See “Setting Auto Verify for a Unit” onpage 109 and “Scheduling Background Tasks” on page 156.)You can manually start a verify through either 3DM or 3BM.Note: If the unit has not previously been initialized and you manually select VerifyUnit in either 3BM or 3DM, the initialization process starts.
