Hi!

"sometimes after a writeback the lines are in reverse order" - in reverse order because index incorrect, or they're in reverse order because they're written physically in that order? If they're written physically in the reverse order, I guess you can fix this by 3-d buffering or tableupdate() just after each new record adding. Record pointer in VFP for new records in the buffered table is negatve startign from -1. Than -2. Than -3... See if this can cause a problem in the program.

HTH.

>I have a large, 5 MB, VFP 6.0 EXE that has just begun to have a real problem. One of the tables in the system is work order lines. When working on the workorder we copy the lines to a cursor because a considerable amount of work is done reguarding change orders, etc.
>
>The problem comes in when writing the lines from the cursor back to the table. The table is indexed on the workorder key which is the same on all lines for a given workorder. It seems that in about 1 in about 50 or so writebacks something causes the table pointer to get out of whack. The table buffering is set to 5 - optimistic table buffering. I have been able to correct the problem by closing the table, using the FLUSH command, and reopening it just before the writeback however users report that sometimes after a writeback the lines are in reverse order.
>
>The problem seems to happen only on lower grade PCs - no name stuff but it has happend at at least 3 different shops on about 5-7 different computers. The part of the program causing the problem has not been changed in years so the only thing I can associate with the problem is that the size of the entire exe has gone over 5 MB because of other additions to the program. These additions are in other areas that should have no effect on the problem area. No modifications have been made to the table itself. I reindexed all tables and ran validate database - no errors.
>
>Anyone got any ideas?