Re: PLUS80W.EXE

"The Highlander (aka Gordon)" wrote:

> "DA Morgan" <damorgan_at_exxesolutions.com> wrote
> > Larry wrote:
> > > The SQLPLUS application on the database can be used, but then it would
> > > require moving all scripts to the Unix environment as it cannot access
> > > our NT Network, so this is not a viable option
> > >
> > > By your responses, I can tell that you are well versed in
> > > understanding Oracle theory.
> >
> > Dumbfounded actually refers to my impression of your entire series of
> postings.
>
> You are the one who queried the use of the word.
>
> > Leaving your DBA out, at your request, you are absolutely 100% incorrect
> about cancel terminating a query.
> > And it has never been different in any version of Oracle going back at
> least 14 years.
>
> Really? In my experience which is not quite as academic as yours and not
> quite so extensive as Larry's I have noted that when using the Windows
> SQLplus client on some versions of Oracle, the client does in fact allow the
> user to cancel the query long before any result set has been generated. In
> testing with my DBA we found that yes, indeed when the cancel is selected in
> the Windows client it results in Oracle, at the server end, killing the
> query, not the session. If you kill the entire sqlplus windows client the
> whole session dies, not just the query.
>
> I noted your quoted link, theory is one thing, imperical trial and error on
> a live system is quite another. All the Oracle documentation in the world is
> no substitute for getting out there and trying things.

It only cancels writing to SQL*Plus. If Oracle is on a separate machine (almost always the case) the SQL continues to execute.

Want some fun ... try this in any version of SQL*Plus.

CREATE PROCEDURE infinite_loop IS

BEGIN
   LOOP
      NULL;
   END LOOP;
END infinite_loop;
/

after youve started it close SQL*Plus.

Check the server. It will run until the DBA or SysAdmin stops it by blunt trauma.

Daniel Morgan Received on Fri Apr 04 2003 - 07:18:39 CEST