Re: Where are these table stats coming from?

Ok, I'll start running some tests.
(I was originally wondering if it was doing a full compute on the table for each index rebuild. This would account for why it's taking so much time to rebuild the indexes. I'm now less inclined to think that is what's happening. I think it's just updating a couple of fields in the user_tables view.)

thanks so much for your response!

On 6/8/05, Wolfgang Breitling <breitliw_at_centrexcc.com> wrote:
> Barbara,
>=20
> AVG_ROW_LEN =3D 100 is the default. BLOCKS is gotten out of the segment
> header for the table and this is accurate, and I suppose gathering index
> statistics requires the number of rows in the table - it is after all
> one of the index statistics - so it seems that "compute statistics" on
> index builds and rebuilds updates the table stats while it's at it. In
> that it behaves consistent with gathering histograms using dbms_stats.
> It too update the num_rows table statistics (and perhaps blocks, I
> haven't checked that).
> What's left to verify is if it will override existing table statistics.
>=20
> Barbara Baker wrote:
> > Wolfgang: Interesting! I should have looked
> > Yes, both are filled in. But the value for avg row length is WAY off.
> > Should be about 900.
> >
> > NUM_ROWS : 16233354
> > BLOCKS : 3250702
> > EMPTY_BLOCKS : 0
> > AVG_SPACE : 0
> > CHAIN_CNT : 0
> > AVG_ROW_LEN : 100
> --
> Regards
>=20
> Wolfgang Breitling
> Centrex Consulting Corporation
> www.centrexcc.com
>

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