Not sure what preciely you mean. The 'set unused' option simply updates the
data dictionary to make it seem as if the column(s) affected no longer
exist. However, you are right that no change is actually made to the row
itself. That should answer the rest of your questions: future updates to
that row are unaffected (so long as you're not trying to update the columns
set as unused, of course), and behave as they have always done. Batch
updates are therefore treated exactly the same as they always have been.
Regards
HJR
--
Resources for Oracle: http://www.hjrdba.com
===============================
"IAmTheRepoMan" <IAmTheRepoMan_at_hotmail.com> wrote in message
news:Il0J7.141$Il6.137320_at_news20...
> I have some very large tables in which I'd like to drop some columns.
Based
> on the overhead of a drop column revisiting every block to coalese the
free
> space, I'd like to mark them unusable for the time being and just reorg
them
> in a few weeks. I'm pretty sure that new inserts to the table will fill
> blocks correctly. But ,what exactly happens when I update a row that
> contains a column that has been marked unusable. Does the row get
adjusted
> during the update or does the entire block get adjusted? What about a
large
> batch update to these rows?
>
> TIA
> Craig
>
>
Received on Thu Nov 15 2001 - 22:21:43 CST
