Re: fuzz failures not generating bugs

[Guy Harris <guy () alum mit edu>]

On Nov 30, 2012, at 1:08 PM, Evan Huus <eapache () gmail com> wrote:

> I would think so. It's bothered me for a while that we didn't have a way to distinguish between "brand new, nobody 
> has looked at it yet" bugs and "solution identified, but nobody wants to work on it" bugs.

CONFIRMED is "somebody's looked at it and determined that it's really a bug (as opposed to, for example, that's what 
it's supposed to do)"; IN_PROGRESS is "solution identified and somebody's working on it".  There's no separate 
"solution identified, but nobody wants to work on it", or even "yes, it's a bug, but we haven't figured out the cause 
yet" state; CONFIRMED covers those two.

> Separating our current NEW bugs into either UNCONFIRMED or CONFIRMED states seems like the right way to do that.
>
> While on the topic, I'd also love an "INCOMPLETE" state like Launchpad (for bugs that are waiting on the submitter 
> for more information -- we seem to have a fair number of those), but I suppose one thing at a time :)

Yes - a bug database I've worked with a state like that.

It also had resolutions along the lines of

        NOTABUG - "that's what the software's supposed to do";

        NOTOURBUG - "it's a bug in some other software that we use (and that we can't or shouldn't work around)"

which we currently lump under INVALID.  (It also has resolutions equivalent to the existing resolutions WONTFIX and 
WORKSFORME.)
___________________________________________________________________________
Sent via:    Wireshark-dev mailing list <wireshark-dev () wireshark org>
Archives:    http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
             mailto:wireshark-dev-request () wireshark org?subject=unsubscribe