RE: Question About Space Usage with LONG RAW Column

Hi Mark

In my opinion two possibilities:

1. Many rows have been deleted. Since the delete statements have no = influence on high water mark (which is shown in user_tables.blocks), = empty blocks are not counted.
2. If you have migrated or chained rows (or both at the same time...) by = selecting the rowid you only have the blocks containing the head pieces.

Chris

>-----Original Message-----
>From: oracle-l-bounce_at_freelists.org =

[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Mark
>Strickland
>Sent: 20 October 2004 22:51
>To: oracle-l_at_freelists.org
>Subject: Question About Space Usage with LONG RAW Column
>
>I did a count of the number of blocks of a Kana table (kc_rawtext) with
>this query: =3D20
>
> select count (distinct substr(rowid,7,9)) from kc_rawtext. =3D20
>
>The result is 2,198,800 blocks. However, user_tables.blocks shows
>4,113,400 blocks. I'm trying to figure out the discrepancy. The table
>has a LONG RAW column and it appears that the column is not the last
>column in the table (column_id for the LONG RAW column is #3 out of =
five
>columns). The table grows quickly. Can someone help me understand
>where the space is going? Thx!
>
>Mark Strickland
>--
>http://www.freelists.org/webpage/oracle-l

--
http://www.freelists.org/webpage/oracle-l