Re: Strange issue after patching to 10.2.0.4 CPU 24

From: Jared Still <jkstill_at_gmail.com>
Date: Tue, 8 Dec 2009 10:35:29 -0800
Message-ID: <bf46380912081035s32f971c5j406b4d73cedaf610_at_mail.gmail.com>

On Mon, Dec 7, 2009 at 12:58 PM, Taylor, Chris David < ChrisDavid.Taylor_at_ingrambarge.com> wrote:

> We went from 10.2.0.3 CPU 12 (I think it was) to 10.2.0.4 CPU 24 and
> discovered a bug I didn’t know existed.
>
> If you’re using ANSI SQL in 10.2.0.3 prior to 10.2.0.4 CPU 2, ORA-00918
> Ambiguous Column may not get raised.
>
> After patching, you’ll have code that starts failing if you failed to
> properly alias duplicate columns.
>

Ah, for lack of a better name, a "reverse effect bug"

This is a "bug" that rears it's head when only when Oracle fixes the bug, and code that would have previously been broken, but actually ran due to the bug, then stops working.

If someone has a better name, please chime in.

There's probably one already coined somewhere, I just haven't looked for it.

I recall a similar PL/SQL bug from several years ago when some in-house code started failing after a minor patch was applied, a patch that fixed a bug in the PL/SQL whereby an error message was not being raised.

Not unlike the situation you refer to.

In both cases the proper coding technique had long been documented, and apparently long ignored. :)

Jared Still
Certifiable Oracle DBA and Part Time Perl Evangelist Oracle Blog: http://jkstill.blogspot.com Home Page: http://jaredstill.com

--
http://www.freelists.org/webpage/oracle-l

Received on Tue Dec 08 2009 - 12:35:29 CST