Wireshark-dev: Re: [Wireshark-dev] fuzz failures not generating bugs
From: Gerald Combs <gerald@xxxxxxxxxxxxx>
Date: Fri, 30 Nov 2012 11:05:00 -0800
On 11/30/12 9:22 AM, Evan Huus wrote:
> On Fri, Nov 30, 2012 at 12:17 PM, Jeff Morriss
> <jeff.morriss.ws@xxxxxxxxx <mailto:jeff.morriss.ws@xxxxxxxxx>> wrote:
> 
>     Hi Gerald,
> 
>     The fuzz-bot seems to be generating fuzz failures but they're not
>     showing up as bugs.  For example the latest fuzz failure
> 
>     http://buildbot.wireshark.org/__trunk/builders/Clang-Code-__Analysis/builds/1620/steps/__fuzz-menagerie/logs/stdio
>     <http://buildbot.wireshark.org/trunk/builders/Clang-Code-Analysis/builds/1620/steps/fuzz-menagerie/logs/stdio>
> 
>     was copied to the automated captures area:
> 
>     https://www.wireshark.org/__download/automated/captures/__fuzz-2012-11-30-20336.pcap
>     <https://www.wireshark.org/download/automated/captures/fuzz-2012-11-30-20336.pcap>
> 
>     but no bug was generated.
> 
> 
> I'd been wondering about that. It's probably related to the recent
> bugzilla upgrade?

It looks like I should have read the release notes more closely. Fuzz
failure reporting uses the bugzilla-submit script, which requires
converting to a new status workflow in Bugzilla 4.0 and 4.2:

http://www.bugzilla.org/releases/4.2/release-notes.html#v40_feat_workflow

Converting would mean going from this:

  http://www.bugzilla.org/docs/3.6/en/html/lifecycle.html

to this:

  http://www.bugzilla.org/docs/4.2/en/html/lifecycle.html

Unless anyone really prefers the old, more complex workflow I'll
schedule the conversion for tomorrow morning PST.