Hi!

Never met such a problem. Looks like corruption goes because unusual data or some weirdness.
There are certain SYS() functions you can use to manage memory buffers, but I don't sure it can help. You can also try to clean up memory buffers using SYS(1104) function each time after a prtion of work. Finally, there are certain cases that can spoil indexes. If I can take a look to this program...

>I am working with a team of programmers and we have encountered a problem when using cursors created by SQLEXEC into MS SQL server 7.0. This application is a batch process type of program it extracts data from our MRP system and populates an SQL database. During the process I use the SQLEXEC command to create cursors from lookup tables (they are very static) which are usually small cursors, less than 500 records, I then index them, and seek into them for data validation which is much faster than going back up to SQL every record...Speed is VERY important in this process. Also I pull a rather large cursor that is using a SQL view, it's about 68000 records, I need this to get static data for each record I am processing, so I pull the cursor and then index it so I can search into it. This is where it gets weird, I pull the small cursors, index them, and then I pull the large cursor, as soon as I pull the large one, the small cursors get random data corruption in some of the fields.
>The corruption looks like "sticks", if I do a asc(left()) of the field with the corruption I get a 0 back. Well asc(0) = NULL. If I reverse the order and pull the big cursor first and the small cursors last, it will not always corrupt the data...but sometimes It will. I have tested this on more than one machine, I have no Idea what this could be other than some sort of memory issue. If anyone knows any settings I could modify in FOX to help with this or ANYTHING I can do...I would be MOST grateful...hell I would dance at your wedding...