test
"Sybrand Bakker" <postbus_at_sybrandb.demon.nl> wrote in message
news:81uiet4nq2bgt5mbk7r2iqpfck4sfejc9b_at_4ax.com...
> On Fri, 27 Apr 2001 13:29:35 +0200, hegyvari_at_ardents.hu wrote:
>
> >Greets,
> >
> >Config: 8.0.5 running on RedHat 6.1.
> >
> >I had a strange problem this morning: some of the queries took forever to
> >finish. Upon close examination I found that those queries all had one
> >thing in common: all of them needed the temp tablespace for sorting.
> >Started Storage Manager, clicked on tablespace and I had to wait at least
> >20 seconds (instead of one) to get the list of tablespaces. Shooed away
> >all the users and tried to remove the TEMP tablespace. I could see the
> >process start on the server, but after 50 minutes of waiting it was still
> >running, so I shut down the instance. Started it again and assigned the
> >database users to a different temporary tablespace, let the users come in
> >and everything went fine afterwards. I suspect there must be a bug in
> >SMON which messes up TEMP, and also my rollback segments seem to be very
> >large. Anyone experiencing anything similar?
> >
> >Best Regards,
> >
> >Hegyvari Krisztian
>
> An unlikely story. Most likely though your temp tablespace has a very
> small initial and next default storage. This results in temporary
> segments with many extents.
> If you are going to drop that tablespace you will have very high
> activitity on fet$ and uet$ in the dictionary, as the change in status
> of the extents need to be administrated.
> BTW: 8.0.5 has been desupported more than a year ago, so in the very
> unlikely case it is a new bug, Oracle is not going to address it.
>
> Regards,
>
>
> Sybrand Bakker, Oracle DBA
Received on Mon Apr 30 2001 - 12:45:09 CDT
