[prev in list] [next in list] [prev in thread] [next in thread] 

List:       bugtraq
Subject:    Re: Bug in WinNT 4.0 SP4
From:       David LeBlanc <dleblanc () MINDSPRING ! COM>
Date:       1999-04-27 20:13:54
[Download RAW message or body]

At 04:36 PM 4/26/99 -0400, Paul Gracy wrote:
>I must disagree.  Any action that a program takes that can crash a server is
>a bug.  Period.

I did not say it wasn't a bug.  A bug, by definition, is something that
causes an application (or even the whole OS) to crash or otherwise
malfunction.  So you are not disagreeing with anything I _said_.  If you
can make something go splat, then it is a bug.  No arguments there.

>The fact that properly using the SDK and following all the 'rules of
>microsoft' would prevent the crash is not an excuse.

No excuses were being made.  Please do not manufacture excuses when they
are not present.

The only point was that Alvaro seemed to think that it was a problem that
moving a folder could result in a total path which is > MAX_PATH.  So far
as I know, this isn't a problem, since if you are correctly handling the
open, you can deal with extremely long paths.  I thought that others might
have the same sort of issue, and also thought that few people would know
that bit of arcane trivia, so I was trying to point out how you might deal
with this correctly.  In general, using API calls correctly, and knowing
various bits of trivia from the documentation is a Good Thing, and perhaps
might save others from having their app go down.

I was NOT saying that crashing is not a bug.  That would be ridiculous.
Neither the little backup app that comes with NT, or the Seagate product
(which as far as I know, both sprung from Arcada, which Seagate bought) are
favorites of mine.  And before anyone asks, I really don't have something I
can recommend.


David LeBlanc
dleblanc@mindspring.com

[prev in list] [next in list] [prev in thread] [next in thread]