When you clone a partition from a source disk with read errors due to bad sectors to a new disk, Windows will cry and report bad sectors on the new disk.

To begin from the scratch, if you need to back up or clone partitions that use proprietary NTFS file systems, your options are to go back to DD or to use ntfsclone. This unimpressive but useful program is part of ntfsprogs package a multi utility archive for NTFS formatted disks. Most Linux distributions include the package, but if not, then building the tool from scratch is not much of a task.

Ntfsclone’s controls and functionality are similar of dd. For e.g.

ntfsclone - - save - image
- -output myimage.
img /dev/hda1

will save the first partition on the first disk to a file called myimage.img, and the following command will write the contents back to disk.

ntfsclone - - restore – image
- - overwrite
/dev/hda1/ myimage.img

If your original NTFS partition contains errors, ntfsclone will cancel the operation. Therefore, you need to use the - - rescue parameter with ntfsclone to keep reading in spite of errors, thus rescuing the partition.

Unless you are certain that the bad clusters were just an error, you should do a full disk surface scan using your hard disk manufacturer’s tools or Windows native chkdsk /r after doing this. This would help in marking those bad sectors and Windows should not report errors.

There are some disadvantages though:

1. should not be used by people not knowing what they are doing
2. it would be difficult to implement incremental backups