"Niall Litchfield" <n-litchfield_at_audit-commission.gov.uk> wrote in message news:<3f39033e$0$15034$ed9e5944_at_reading.news.pipex.net>...
> "quarkman" <quarkman_at_myrealbox.com> wrote in message
> news:oprtr886hxzkogxn_at_haydn...
> > With autoextensioin turned off, you also get poor performance. Very poor.
> > As in, the thing needing the space which it can no longer get
> automatically
> > gives up the ghost and announces to you that it's 'unable to extend by xx
> > in tablespace TEMP'. The transaction or query then blows up (and gets
> > rolled back if need be. Though in 9i, you can arrange for it to suspend
> > itself until you sort out the disk space issue, rather than immediately
> > blowing up). Hence the need to pro-actively manage the thing, and make
> sure
> > sufficient space is provided up-front.
>
> Hi Howard, good to have you back.
>
> Actually I think that this 'poor performance' is a good thing. There is a
> class of users who won't complain when their process takes 25 minutes -
> because it is using 45 million lios - after all its always been that long
> hasn't it.
>
> They do however tend to complain when they get "A funny Oracle error message
> something about extending tablepalaces or something just came up and when I
> clicked OK my report died is the internet down or something?". Theres no
> V$view quite so helpful at identifying problem processes as the end-user
> whinging at you down the phone.
>
> Obviously it would be nice to always allocate enough space to temp up front,
> but there will always be that one 8 way cartesian join to screw things up
> for you.
>
>
> --
> Niall Litchfield
> Oracle DBA
> Audit Commission UK
>
> P.S. I haven't understood how the set of users that can wait 10 minutes for
> a drop down list to, well drop down, can coexist with the set that complain
> that it takes 5 seconds to save this form and so the whole damn system is
> unusable crap.
Don't forget the group that waits a full 4 weeks to tell you something
stopped working.
Received on Thu Aug 14 2003 - 08:53:49 CDT
