Re: Running with _system_trig_enabled set to false

"Paul Brewer" <paul_at_paul.brewers.org.uk> wrote in message news:<3e36fcee_1_at_mk-nntp-1.news.uk.worldonline.com>...
> "Mark D Powell" <Mark.Powell_at_eds.com> wrote in message
> news:2687bb95.0301280730.2f50b15e_at_posting.google.com...
> >
> > Joe, if you have metalink support look up the following documents:
> > (bug no) 2187432, Doc Id 143639.1, 222467.1, and 2670738.1.
> >
> > The bug appears to match your problem since it appears to relate to
> > JAVA code.
> >
> > As far as turning off database event triggers goes: This is a feature
> > of Oracle that you do not have to use and if your shop does not now or
> > intend to use the feature then turning it off should be fine.
> >
>
> I would be very wary of running in this mode.
>
> We are on 8.7.1.4 and have a THIRD_PARTY three-tier application. All the
> developers know the Oracle Username/Password which the application server
> uses to connect to the database - and it would be useless to change it.
>
> All the business rules are in the app. There are no constraints in the
> database (other than primary keys) and I have *no* influence here.
>
> The best I've been able to do so far is to restrict that Oracle username to
> logging in from named terminals (i.e. the application servers), using an
> after_event logon trigger owned by SYSTEM.
>
> Any suggestions would be welcomed.
>
> Regards,
> Paul

Paul, I think you need database event triggers, but most shops do not use them so turning them off until a bug patch is available is not going to affect anything. In your case if you have hit this problem then you probably need to file an iTAR to get Oracle's help. It appears from one of the documents I listed that the problem is JAVA related and removal of the code might be the solution, that or re-validation of the invalid referenced code/object from a trigger. There were several other documents, but I listed the ones with recent dates that looked like they would apply.

IMHO -- Mark D Powell -- Received on Wed Jan 29 2003 - 08:47:39 CST