Re: changing initial extent size with hash partitioned index

samuels_at_blue.seas.upenn.edu wrote:

> there's no need to be rude.

And what, pray, do you consider rude?

I asked why you were mucking around with extent sizes when the whole point of locally managed tablespace is that that sort of waste of time is old hat and no longer necessary. Only an overly-sensitive person indeed could construe the suggestion that they have missed the point of LMT as being rude.

Or perhaps it was the word "bloody"? In which case, do you live in Utah or something??

> version is 8.1.7.4.0, platform is aix5.1

Good. Always post this information. It frequently makes a difference to the answers you might receive. Or was it rude of me to ask you for it?

> my question has more to do with the fact that when rebuilding a non-partitioned or range partitioned index,
> an initial extent size is allowed, and with a hash partitioned index it is not (apparently).
>
> if you can't post a well-mannered constructive follow-up, then you shouldn't at all.

And my reply to you was: Why are you trying to change an initial extent in the first place? A question you haven't answered.

That a normal index *can* have a storage clause supplied when rebuilding doesn't mean you *should* supply a storage clause. Still less that you should actually rebuild the thing.

In 8i and above, *no* segment should have a storage clause supplied, ever. That's the wonder of LMTs. It was also good practice even in 8.0 with the MINIMUM EXTENT clause, but results were less predictable.

Neither would I take the syntactical abilities of a normal index as any kind of indication of the syntactical capabilities of a partitioned index.

So your question has more to do with the fact that you dislike encountering a "limitation" when in fact no limitation actually exists, and that you are encountering it whilst engaged in an activity that in 8i and above is largely redundant (ie. worrying about extent sizes).

Now I don't know what you class as rude or polite, well-mannered or ill-mannered. But the above is accurate. So you can take it or leave it as you will.

HJR Received on Wed May 19 2004 - 20:00:50 CDT