I made a SWAG that the type of corruption was coincidental with respect to data being written back to the disk. The Client32 settings I gave have a significant effect on saving data back to the disk.

I would also ask them if there has been a change in the version of Client32 being used. They should be using the most current version. There were problems with the versions between 2.5 and 3.3, exclusively.

Later versions of the client did correct some of this. However, if the previous client was not uninstalled before the newer version was installed, the previous [bad] settings could remain.

>Mark,
>
>Some of these issues have been corrected in later versions of the client.
>
>My question is, what are "corrupted delete flags" and why them in particular and not the usual header/reccount mismatch?