bugzilla triage madness :-/

Michael Schwendt mschwendt at gmail.com
Tue Apr 8 23:03:08 UTC 2008


On Tue, 08 Apr 2008 14:24:21 -0700, John Poelstra wrote:

> Michael Schwendt wrote:
> > On Mon, 07 Apr 2008 21:23:28 -0700, John Poelstra wrote:
> > 
> >> Pekka Savola said the following on 04/04/2008 11:07 AM Pacific Time:
> >>  > On Fri, 4 Apr 2008, Michael Schwendt wrote:
> >>  >> What do I care? When I filed the bug, the product was fresh and
> >>  >> maintained,
> >>  >> wasn't it? But more than a year later I'm no longer willing to spend 
> >> time
> >>  >> on the same issues without a single sign of life from the package owner.
> >>
> >> I think this is completely reasonable.  I'd do the same thing and would 
> >> thus close the bug.
> > 
> > The message sais the tickets will be closed automatically after 30 days.
> > That is 30 days for the maintainer to take a last look and see whether the
> > issue is fixed or not.
> > 
> 
> No, it is not as drastic as you describe.

It is, it is. The maintainer, who has ignored a ticket for a long
time, won't take a look at a closed ticket anymore either. The
maintainer does not answer the NEEDINFO call either, because you
believe it's solely the reporter's duty to take action. And according
to your theory no maintainer has the time to handle incoming
tickets. And with that we're back at my initial post. I've seen the
request to "retest against a newer release" several times before (just
not with EasyFix bugs), and I don't like that game if it's applied to
all tickets brute-force.

> Yes, EOL bugs will automatically be closed after 30 days.  They should 
> have been closed a long time ago anyway :)

Yes, but by a human being, who decides between FIXED, WONTFIX or
CANTFIX. Not by a bot that can't even determine whether a maintainer is
AWOL or deliberately ignorant.

I don't mind if you handle some packages in a special way,
e.g. kernel, anaconda, xorg, yum, perhaps GCC, and a few others with
hundreds to thousands of bugs per Fedora release. But I don't like it
at all that packages with a single open ticket in bugzilla (!) are
also handled by a bot without any word from the package maintainer in
over a year. If the Fedora package collection is in such a desolate
state, I rather not report any new bugs.

> For other bugs that remain 
> in NEEDINFO for thirty days, they will be automatically closed too--the 
> normal process for bugs in NEEDINFO.  If you think this is the wrong 
> approach please lobby FESCo to change the process they have approved.

Why? If FESCo believes the process is best-effort, so be it. It's
possible to reopen closed tickets. It's possible to open new tickets
if something bugs me. I can open tickets automatically whenever I
want to. But I rather not waste time in bugzilla and on retesting issues,
which are considered lowest priority apparently. Perhaps in the future
it will be possible again to commit fixes in cvs.

> At any time before these 30 days are up you can change the bugs to 
> version 'rawhide' and change the state to ASSIGNED, thereby avoiding 
> auto-closure and giving you many months to review bugs as far back as 
> you like.
> 
> And if you wanted to be a little more creative you could query all of 
> the bugs assigned to you and change them all at once through the 
> bugzilla GUI.  Contact us on #fedora-qa we'd be glad to walk you through 
> the process :)

No, thanks. Filing the bugs has been a one-time offer.




More information about the fedora-devel-list mailing list