>>>>>There are two problems:
>>>>>
>>>>>- CDX bloat. If you simply REINDEX, the new tags will be appended to the end of the CDX file. You need to DELETE TAG to remove the space, but then, REINDEX won't work.
>>>>>- REINDEX relies on the tag information stored in the header of the CDX. If this is corrupted, the index will be too or VFP won't be able to read the information. In this case you may have to delete the CDX to build the indeces.
>>>>
>>>>Thanks, I wasn't aware of the bloat issue. When corruption occurs, I replace the CDX with an empty one (stored in my DBC's User field) then issue reindex. This has corrected every corruption that I've run into.
>>>
>>>
>>>That's a great idea.
>>
>>Thanks, I was frustrated by DELETE TAG ALL messing up the relationships... Tried a few unsuccessful attempts before I got it working. THEN, I heard about SDT. We bought it, but never used it since I'd already got my DB maint procedures in place.
>
>
>SDT has some other great features, like updating DBCs at the client and fixing corruption.

True, I just haven't NEEDED it yet. <G> Besides, we now support both VFP and SQL Server from one code base. And, SDT didn't handle SQL Server when last I looked. I've gotten pretty proficient at writing update scripts for VFP and SQL to keep my DBCs in sync. Not that SDT wouldn't do a good job, I just haven't found/made the time to delve deeper into it. <sigh>

We just bought xCase and I'm in the process of learning how to use it.