>>>Are you aware that this happened on plain dbfs? No ;DBC, no sofisticated buffering is going on! Just that the data were not flushed in the good old dos sense. And btw, from other stations, I did get the correct information back for sql select ...
>>>
>>>I persist. Scary _and_ weird.
>>>
>>>Marc
>>
>>Marc, it may be related to a problem I just posted. (It's in 'Troubleshooting', thread '91327') about data not being available even when a table is closed and re-opened. If I learn anything, I'll pass it on.
>>
>>Barbara (no help, just sympathy)
>
>Did I mention that UNLOCK solved the problem? The incriminated record was locked before the update.
>
>... and Barbara, in these difficult times, sympathy _is_ help. :)
>
>Marc

Speaking of UNLOCK, I've found that when using RECALL, one must UNLOCK all records across all Private Datasessions in order for it to work. It doesn't bomb out either or give any error messages. In the debugger, the code even appears to work but when you go back and look at it, the record(s) are still deleted. I've had to traverse each datasession doing an unlock all before attempting the recall.

Steve