I have tablespace that keeps getting space eaten up for no clear
reason.

I have some tables set up that keep 5 of months of data. Every night
a stored procedure deletes (delete from sector_tab where tstamp <=
(sysdate-150); commit older data.

The number of rows of data remain constant. I've checked that this is
true using count(*).

Also in this tablespace I use views and materialized views
(snapshots).

A procedure runs every hour that refreshes the snapshot and loads new
data into the tables.

I also use partitions on the tables. However, we increase these as
necessary and delete the older ones.

Lastly, I have logging turned on. Could these logs be eating up
space?

Our dba has backed up the tables by exporting the data, dropped the
tables and recreated. This did not help. Both of our dba's are at
loss to explain what's causing this and their solution is to increase
the disk space.

Any ideas about what else could be causing this? Any suggestions
would be appreciated.

-David

Did your DBAs try to ALTER TABLE ... MOVE the tables to rebuild them? When
they did the export / import, did they use COMPRESS? When you load the
tables, do you use direct-load inserts or any other
"above-the-high-watermark" method? Are there indexes that could be
accounting for the change in available space? Try re-posting with a little
more specific information. These space-eating problems can be tricky to
figure out without a database connection right in front of us.
--
Cheers,
Chris

___________________________________

Chris Leonard, The Database Guy
http://www.databaseguy.com

Brainbench MVP for Oracle Admin
http://www.brainbench.com

MCSE, MCDBA, OCP, CIW
___________________________________

"wireless" <wireless200 (AT) yahoo (DOT) com> wrote