Re: Locally managed tablespaces and migrated rows

From: Nuno Souto <nsouto_at_nsw.bigpond.net.au.nospam>
Date: Wed, 09 May 2001 11:24:36 GMT
Message-ID: <3af924d9.8580919@news-server>

On Wed, 9 May 2001 11:44:53 +0100, "Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> wrote:

>
>I hope that was a rhetorical statement - I happen

Yup. I hate those 3rd parties and their "knowledge of how things should be coded". But I have to take it, this would be a very small world without them. Necessary evil, IMHO.

>so that we don't get hit with smon kicking the
>dictionary to death every five minutes looking for
>space to coalesce. Have you seen what happens
>to I/O on the SYSTEM tablespace when your
>database has 25,000 tablespaces !!!

25K? Gosh Jonathan, I know you like to test things to destruction but aren't you getting carried away? Or is that 25K in TS$ but a lot less in reality? BTW, I often wonder where the "drop the oldest partition" thing is gonna get us when it's been done a few thousand times. On second thought, get carried away! We need someone to do this so we don't get bitten by "easter eggs". :-D

>
>(Unless you set events 10061 and 10269, or
>hack sql.bsq of course).
>

Yup. I noticed that too with mine. No more weird I/O going on in SYSTEM. Well, not much anyway. Didn't know 10061 and 10269 had that effect, will look them up. Thanx a lot.

Cheers
Nuno Souto
nsouto_at_bigpond.net.au.nospam
http://www.users.bigpond.net.au/the_Den/index.html Received on Wed May 09 2001 - 06:24:36 CDT