Look at http://support.microsoft.com/default.aspx?scid=kb;en-us;168762. Basically, we take our tables when they are empty and copy off the .CDX files. This gives us the structure of the indexes without any data entries. Then, when we reindex, we copy the good, empty .CDX files over top of the live ones, then issue REINDEX to populate the empty .CDX files with the necessary data entries.

I'm sure a tool like Stonefield Data Toolkit is much more robust, but this has worked pretty well for us.

>What do you mean by a clean index?