Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Re: ORA-600: Hardware ok
Thanks for your answer,
I think hardware is ok (with HACMP we were able to change to the 2nd processor
which has his own memory, but the problem occured again) and I know that it is
difficult for oracle to help us. May be the combination of our software
(8.1.6.2, Intermedia and IMS Access Manager) on AIX is not often used. May be
there is someone having the same combination and no problems???
Regards
Jean-Claude Matarazzo schrieb:
> Hello,
>
> What's the Oracle Support had said to you ?
>
> * the problem of Oracle : the meaning of the codes ORA-00600 [] []
> problem of disk ? problem of memory ? Oracle bug ?
>
> * Upgrade your version 8.1.6.2 to ?
>
> Perhaps, you can have hard problems :
> * the disks don't work very well
> * the RAM memory have problems
> ...
> So Oracle could give you ORA-00600.
>
> Regards.
>
> "maxd64" <Wilfried_Binder_at_t-online.de> a écrit dans le message news:
> 3A5F943D.BCADDB39_at_t-online.de...
> > We see corrupt data blocks and instability in our database 8.1.6.2 with
> > the intermedia option (context) running on an IBM RS6000 S80 with AIX
> > 4.3.3 (32bit) and HACMP.
> >
> > The stored data is not static, it is some kind of a mailing system with
> > incoming messages from another IBM-system that pushes data via oracle's
> > IMS ACCESS MANAGER (database link). There are about 10.000 incoming docs
> > a day, the same number is deleted, so that that the number of docs is
> > always about 250.000.
> > The context index for the incoming messages is immediately actualised.
> > The user's application is forms 4.5 based (we know we have to change!).
> > Retrieval is done with the intermedia option (context).
> >
> > We have heavy problems, once a week corrupted data (the servers has to
> > run 24 hours/7 days), messages are ORA-600 (subcodes 2123 or 25252 or
> > 12700 or KCBZPB_2 or ...). There are about 6 high used tables in the
> > application, together with the also high used context index tables they
> > are the candidates for corruption.
> > We cannot reproduce the problem.
> > The database has been rebuilt (complete export and import), 10 days
> > later, the error came again. Hardware has been checked, the fibre
> > channel bug from IBM is fixed.
> >
> > Is there anybody with better hints than oracle's field support?
> >
> > Thanks
> >
Received on Sat Jan 13 2001 - 16:28:23 CST