Re: decreasing initial extent

"Richard Foote" <richard.foote_at_bigpond.com> wrote in message news:69f6c1c8.0301152003.55b98cff_at_posting.google.com...
> "Howard J. Rogers" <howardjr2000_at_yahoo.com.au> wrote in message
news:<H0mV9.24997$jM5.65909_at_newsfeeds.bigpond.com>...
> >
> > Come off it Richard! (And I mean that in a loving way). You queried the
> > wrong view. Here's a rather more telling test:
> >
>
> Hi Howard,
>
> You spoil sport, you went and ruined it all ;)
>
> I'm sorry, I was being a bit mischievous (again).
>
> With LMT, the term "Initial Extent" is just a touch ambiguous as the
> Initial storage parameter (along with the others) is used to determine
> the initial *size* of the segment, which is listed as Initial_Extent
> in the _table views. IMHO, once DMT are gone forever, it would be nice
> if this column be renamed "Initial Size" (or some such).

I'll buy that argument. Just.

> With LMT and the *uniform* clause the "true" initial extent (the first
> extent belonging to a segment), resizing is a tad difficult.

Is that a bowiesque way of sayin g 'impossible'?

> However, the "true" initial extent can be resized in a LMT (I'll be a
> bit more serious this time although yes, it's a bit of a dirty trick
> ;)

Not really. One can expect that in autoallocate, and (to me) it seems perfectly respectable, not a dirty trick at all. It's rather smarter than I would have assumed Oracle would have made it, which is a nice bonus. But the point again though is that it didn't do what you expected it to (resize to 50K), but *something* happened.

[snip]

> Pretty clever I thought !!

Thank you!

Regards
HJR Received on Wed Jan 15 2003 - 23:21:52 CST