Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Sat, 1 Oct 2022 09:01:47 -0400
From:   "Theodore Ts'o" <tytso@mit.edu>
To:     Bagas Sanjaya <bagasdotme@gmail.com>
Cc:     Laurent Pinchart <laurent.pinchart@ideasonboard.com>,
        "Artem S. Tashkinov" <aros@gmx.com>,
        "Bird, Tim" <Tim.Bird@sony.com>,
        Thorsten Leemhuis <linux@leemhuis.info>,
        Slade Watkins <srw@sladewatkins.net>,
        Konstantin Ryabitsev <konstantin@linuxfoundation.org>,
        "workflows@vger.kernel.org" <workflows@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        "regressions@lists.linux.dev" <regressions@lists.linux.dev>,
        "ksummit@lists.linux.dev" <ksummit@lists.linux.dev>
Subject: Re: Planned changes for bugzilla.kernel.org to reduce the "Bugzilla
 blues"
Message-ID: <Yzg6O5BQUAYpj+bH@mit.edu>
References: <95c3384b-53d0-fd6c-6ec5-a7e03fdeddfc@gmx.com>
 <F300ED64-5E8E-4060-89DC-C98BC5FF08E6@sladewatkins.net>
 <YzXK6Px+BrNuuMZH@pendragon.ideasonboard.com>
 <a86adc6d-05db-ec2e-c5de-d280aad9fb8a@leemhuis.info>
 <Yzbtuz6L1jlDCf9/@pendragon.ideasonboard.com>
 <BYAPR13MB250377AAFCC43AC34E244795FD569@BYAPR13MB2503.namprd13.prod.outlook.com>
 <3cfaef48-744f-000f-1be5-6f96d64dea24@gmx.com>
 <YzcdnnjJA4M09dNH@pendragon.ideasonboard.com>
 <YzeVVV+nPaxsqS0V@mit.edu>
 <Yzfy5wXuU0uxdIqr@debian.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Yzfy5wXuU0uxdIqr@debian.me>
Precedence: bulk

On Sat, Oct 01, 2022 at 02:57:27PM +0700, Bagas Sanjaya wrote:
> > Funny thing.  I've largely given up on getting any kind of useful bug
> > report from Launchpad, so I've largely ignored it.  In contast, the
> > bug reports I get for e2fsprogs from Debian are generally far more
> > actionable, with bug reports that have all of the data so I can
> > actually root cause the problem, and help the user.
> 
> So no matter how the bug tracker interface is, the etiquette is:
> Whenever something buggy happens, try to gather all information related
> to that event (reproduction steps and reproducer, logs, crash dumps,
> etc), then file the polished report. From your experience, it seems like
> Debian people knows it.

Another critical part of the bug tracker etiquette is when in doubt,
always file a separate bug report.  More than once, both with
Launchpad or Kernel Bugzilla, users will do a web search for "my file
system lost data" or "EXT4-fs error" and assume it's the same problem
as what they are seeing.

In some cases, for a bug report that is years old and already closed.
That's actually less damaging, because it's obviously noise, and it
can be ignored.  The more annoying one is when the bug is actively
being worked, and people dog-pile onto that bug, and the bugs might be
caused by hardware issues (more often than not, a "bug report" is
really due to someone with a failing hard drive, or a USB stick which
is sticking out of the laptop, and gets jostled).  Even it's a real
software bug, if there are two bugs whose bug reports are getting
jumbled together into a single bug tracker web page, it can get
horribly confusing for the poor maintainer being asked to work the
issue, and the two users who start aguing amongst themselves about
their pet theory.

(Another bug reporter etiquette: clearly differenciate between *facts*
that you are reporting, and your pet theories about what might be
going wrong.  If you're so smart that you think you know the problem,
express your theory in the form of a patch.  Otherwise, putting
theories into a bug report which is not backed up by facts is worse
than useless.)

Of course, all of this can happy with bug reports filed by e-mail, or
via the Debian BTS.  However, it seems that people who are smart
enough to figure out how to send e-mail to a vger.kernel.org mailing
list, or how to use Debian's command-line interface "reportbug"
script, generally have enough experience that they can file a decent
bug report.  Whereas people who can only fill in a web page.... tend
not to have that (minimal) filter applied.

					- Ted