Re: SIGSEGV a bug or a feature?

There are possibly some gaps here.

As root or NT Administrator ( for example) one can always jump into a process' memory pool and start playing about

inside the ring you can do this. Although a USER is mentioned - OK , point.

second, no program can be declared completely safe

since Oracle just has too many inputs and outputs

The more money you pay Oracle support the more likelihood they will fix it however if they do not want to there is nothing you can do about it "Andreas Plesner Jacobsen" <apj_at_daarligstil.dk> wrote in message news:slrnc27sls.qs5.apj_at_slartibartfast.nerd.dk...
> On 2004-02-04, Joel Garry <joel-garry_at_home.com> wrote:
> >
> > Yeah, you are right. There shouldn't be anything a user could do to
> > sigsegv an _oracle_ process.
> >
> > Can you give them a simple and easily replicable example using only
> > Oracle and platform native software? It's been too long since I've
> > DBI'd, but I seem to recall there are translators to C. You might ask
> > the perlmongers, too.
>
> The exact cause of the problem has been established, I'm able to
> reproduce it using DBI and 8.1.7.4 oci libs, and I probably COULD do a C
> version if I had the time.
>
> But we (I and the Oracle Support personnel) know exactly why the oracle
> process crashes, so I don't see how this would change anything.
>
> I was sort of hoping that I would be able to get some oracle corp
> lurkers' attention with this thread, because it is my distinct feeling
> that I will never (ever, ever, ever) persuade the support-personnel that
> this is a bug (my TAR is in soft close at the moment). To me, a SIGSEGV
> caused by network-based input just screams "BUG!" regardless of the
> client application providing the data.
>
> --
> Andreas Plesner Jacobsen | When eating an elephant take one bite at a
time.
> | -- Gen. C. Abrams
Received on Sun Feb 08 2004 - 11:14:10 CST