Trusted
WirelessNZ: Mate, I dont have the same level of confidence you have in the application. …
Actually I didn’t express a view on the application one way or the other. The CRC is done by the disk controller
at a hardware (sector) level before the read sector is handed on to the OS/application, the CRC checks
whether the CRChecksum generated when the sector is read is the same as the polynomial checksum
generated by the controller & recorded on the disk when the sector was written. The checksum itself is not
generated or checked by the OS/Application – rather the CRC status reported by the controller is checked.
There may be a faint possibility that the CRC status interpretation by the OS/Application is incorrect; it seems
unlikely that Office 2007 or the backup software fiddle at a low-level like a disk editor with the bits within a sector.
Does the absence of an Event log entry give confidence that the sector where the CRC problem was
reported is in fact OK without thorough scanning? Does the Maxtor Shared Network Attached Storage II have
provision to facilitate scanning/diagnostics? The fact that a volume shadow copy (on a different physical part of
the disk) does not show a CRC problem unfortunately says nothing about the current status of the sector
where the CRC problem was reported.
