Re: Is this serious?

Your shared pool is way too small for even the most trivial database. Bump it up to at least 32m. Even more may be required.

The upgrade to 8.1.7 is simply a patch and takes about 30 minutes to apply. The upgrade process is to shut down the instance, back up the database (in case anything goes wrong), apply the patch, startup the instance, and run the upgrade script $ORACLE_HOME/rdbms/admin/u0801050.sql.

--
Chuck

"Tom" <tomNOSPAM_at_eazyriders.com> wrote in message
news:7H9d9.9593$ZH6.74399860_at_news-text.cableinet.net...


> Hi,

>

> The shared pool size is

>

> shared_pool_size = 6291456

>

> this database is not massive in size but it does have quite a number of

 open


> connections. The box had 4gig of memory and currently supports 2

 instances.


> What would be an optimum size for the shared pool do you think?

>

> We do run 8.1.7 but this box is one of the older ones and i'm not sure it

> would be worth the upgrade from 8.1.5. On the upgrade note what do people

> recomend on the best upgrade route? Install the newer version and then exp

> out of the old and imp into the new?

>

> A full exp of this database is only about 5gigs so it is not unmanageable.

> Incedently i exp'd the whole instance the other day and actually caused

 one


> of these errors to occur. Does that give any more clues?

>

> thanks again

>

> Tom

>

> "Chuck" <chuckh_at_softhome.net> wrote in message

> news:al3523$1mn5r0$1_at_ID-85580.news.dfncis.de...

> > Are you pinning frequently used objects in the shared pool? If not you

> > should be. That reduces fragmentation which is usually the cause of

> > ORA-4031.

> >

> > How big is your shared pool?

> >

> > There are also known memory leaks in the shared pool in earlier versions

 of


> > 8i. 8.1.6 was the worst. Avoid it like the plague. As someone else

 stated


> > you need to get up to the latest release of 8.1.7 for your platform.

> >

> >

> > "Tom" <tomNOSPAM_at_eazyriders.com> wrote in message

> > news:n67d9.9261$Me5.70638758_at_news-text.cableinet.net...

> > > Hi,

> > >

> > > We are running 8.1.5 on Solaris7.

> > > About once every month or so i get errors from the instance that are

 the


> > > following....

> > >

> > > Exception in thread "main" uk.co.goodtech.listener.ListenerException:

> > > uk.co.goodtech.listener.ListenerException: Problem generating index:

> > > java.sql.SQLException:

> > > ORA-04031: unable to allocate 4096 bytes of shared memory ("shared

> > > pool","DBMS_LOB","PL/SQL MPCODE","BAMIMA: Bam Buffer")

> > > ORA-06508: PL/SQL: could not find program unit being called

> > > ORA-06512: at line 1 at

> > >

> >

>

 uk.co.goodtech.listener.ListenerManager.notifyListeners(ListenerManager.java


> > > :136)

> > > at

 uk.co.goodtech.cms.RebuildIndexes.indexSection(RebuildIndexes.java:81)


> > > at uk.co.goodtech.cms.RebuildIndexes.main(RebuildIndexes.java:60)

> > >

> > > now the other instances seem fine when this happens and it seems to

 revover


> > > itself before i can get to fix it. Does this problem look serious as

 when
 it


> > > occurred once i had to bounce the instance. Reading up on the error it

 seems


> > > it could be memory related but i'll be interested to hear if others

 have


> > > seem this. A search on google shows the problem but not really a

 resolution.


> > >

> > > thanks

> > >

> > > Tom

> > >

> > >

> > >

> >

> >

>

>