Bug lifecycle page revised / expanded

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi, everyone - just a quick note that, last week, I revised / expanded
the bug lifecycle page:

https://fedoraproject.org/wiki/BugZappers/BugStatusWorkFlow

mostly I just re-ordered it in a logical fashion and extended it to
cover all normal process, states and resolutions.

I'd hope this can now serve as the proper reference for bug workflow for
Fedora: the 'official' reference is currently
https://bugzilla.redhat.com/page.cgi?id=fields.html , but that talks
almost entirely about the RHEL process, not the Fedora one. I'm going to
ask the Bugzilla maintainer to add a little link pointing to the Wiki
page from the Bugzilla one.

I guess the significant points are mostly the 'valid' resolutions, and
resolving Rawhide bugs.

Resolutions - the biggest point is that CURRENTRELEASE is not intended
for use by Fedora.  CURRENTRELEASE, for the RHEL process, does not mean
what it looks like it means. What it means is:

"The problem described has been fixed and only ever appeared in
unsupported or unreleased products."

Which really doesn't apply to Fedora at all. So, don't use it :) We
could declare it to mean something different for Fedora, but that would
just be needlessly confusing. The correct resolution to use for a bug in
a stable release that's fixed by an update should be ERRATA, as
mentioned on the page now.

On the use of ERRATA - Jesse Keating informed me that this was decided
against in the past on the grounds that it doesn't mean quite the same
as in RHEL (Fedora doesn't issue erratas exactly), but it's a lot closer
to correct than CURRENTRELEASE or NEXTRELEASE, per their RHEL
definitions. So the choice is use ERRATA, which means almost the same
thing, or invent yet *another* Fedora-only resolution. Which wouldn't be
fun. Currently Bodhi, when used, closes bugs fixed by updates as
NEXTRELEASE; I'll file an issue to switch it to ERRATA.

On resolving Rawhide bugs - this was discussed in a QA meeting, and we
all felt that the MODIFIED / reporter confirms fix cycle should be
optional for Rawhide bugs. So, as the page now says, outside of a freeze
period, you (maintainers) can go directly from ASSIGNED to CLOSED
RAWHIDE if you're confident you've definitely fixed the issue, or you
can take a report out of MODIFIED to CLOSED RAWHIDE if it's been sitting
there for a while and no-one seems inclined to confirm the fix. This is
entirely up to the maintainer's discretion, so it's up to you whether to
never close the bug until it's been confirmed, use MODIFIED but close
after a while if there's no feedback, or just go straight to CLOSED
RAWHIDE and skip MODIFIED. Whichever works best for you. 

Everything else should just be an explanation of current practice and
doesn't really need any comment...let me know if anything seems wrong or
odd.

The graphic on the page, for now, reflects only the cycle for bugs in
stable releases, it doesn't quite cover the Rawhide case. It's being
revised.
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net

-- 
fedora-devel-list mailing list
fedora-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/fedora-devel-list

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Fedora Announce]     [Fedora Kernel]     [Fedora Testing]     [Fedora Formulas]     [Fedora PHP Devel]     [Kernel Development]     [Fedora Legacy]     [Fedora Maintainers]     [Fedora Desktop]     [PAM]     [Red Hat Development]     [Gimp]     [Yosemite News]
  Powered by Linux