4:24 p.m.
Hi there,
According to https://fedoraproject.org/wiki/BugZappers/BugStatusWorkFlow
CLOSED NEXTRELEASE/RAWHIDE resolution is for maintainers only.
But for bug https://bugzilla.redhat.com/show_bug.cgi?id=498361 is
already clear now.
Can I shortcut the path and do not disturb the maintainer by setting
this bug to ASSIGNED?
--
Best regards,
Sergey Rudchenko
7:17 p.m.
On Thu, 2009-09-24 at 00:24 +0300, Sergey Rudchenko wrote:
Hi there,
According to https://fedoraproject.org/wiki/BugZappers/BugStatusWorkFlow
CLOSED NEXTRELEASE/RAWHIDE resolution is for maintainers only.
But for bug https://bugzilla.redhat.com/show_bug.cgi?id=498361 is
already clear now.
Can I shortcut the path and do not disturb the maintainer by setting
this bug to ASSIGNED?
It's rare that you can ever do CLOSED NEXTRELEASE on your own
initiative, because you don't know what the maintainer's intention is.
You can do CLOSED RAWHIDE in appropriate cases, but this one doesn't
meet that definition, because the bug is filed on 11, not Rawhide. You
can't close a bug filed on a stable release as CLOSED RAWHIDE. It's up
to the maintainer, now, to choose whether he intends to fix this for 11
as well as Rawhide. If he does, he can make the fix and then do CLOSED
ERRATA. If he only wants to fix in Rawhide, he can do CLOSED WONTFIX
with a comment that the bug is fixed in Rawhide but won't be fixed for
11.
(That's the theory, anyhow. Since Matthias is somewhat both the reporter
and maintainer, he could actually just change the bug from 11 to Rawhide
then close it as CLOSED RAWHIDE. But it should be his decision, not
yours, anyway.)
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
8:21 p.m.
On Wed, 2009-09-23 at 17:17 -0700, Adam Williamson wrote:
On Thu, 2009-09-24 at 00:24 +0300, Sergey Rudchenko wrote:
> Hi there,
>
> According to https://fedoraproject.org/wiki/BugZappers/BugStatusWorkFlow
> CLOSED NEXTRELEASE/RAWHIDE resolution is for maintainers only.
>
> But for bug https://bugzilla.redhat.com/show_bug.cgi?id=498361 is
> already clear now.
>
> Can I shortcut the path and do not disturb the maintainer by setting
> this bug to ASSIGNED?
It's rare that you can ever do CLOSED NEXTRELEASE on your own
initiative, because you don't know what the maintainer's intention is.
You can do CLOSED RAWHIDE in appropriate cases, but this one doesn't
meet that definition, because the bug is filed on 11, not Rawhide. You
can't close a bug filed on a stable release as CLOSED RAWHIDE.
I believe this bug _was_ filed against rawhide before F11 was released,
and then move to F11 by some bug triage sweep. Looks like shooting
yourself in your own foot, if you don't have the power to close the bug
anymore, after you've moved it...
It's up
to the maintainer, now, to choose whether he intends to fix this for 11
as well as Rawhide. If he does, he can make the fix and then do CLOSED
ERRATA. If he only wants to fix in Rawhide, he can do CLOSED WONTFIX
with a comment that the bug is fixed in Rawhide but won't be fixed for
11.
I've put the bug out of its misery now. Closed rawhide, moving on...
8:54 p.m.
On Wed, 2009-09-23 at 21:21 -0400, Matthias Clasen wrote:
I believe this bug _was_ filed against rawhide before F11 was
released,
and then move to F11 by some bug triage sweep. Looks like shooting
yourself in your own foot, if you don't have the power to close the bug
anymore, after you've moved it...
yeah, that's a system that's somewhat controversial. there's no perfect
answer to the question 'should bugs filed on rawhide be switched to a
stable version once it's released'. At present we _do_ do this. It does
make sense sometimes; it potentially makes sense here. After all, if you
think about it, the initial bug request could be read as 'please port
this to PolicyKit 1.0 for Fedora 11', since it was filed during Fedora
11 development period.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
9:16 p.m.
On Wed, 2009-09-23 at 18:54 -0700, Adam Williamson wrote:
On Wed, 2009-09-23 at 21:21 -0400, Matthias Clasen wrote:
> I believe this bug _was_ filed against rawhide before F11 was released,
> and then move to F11 by some bug triage sweep. Looks like shooting
> yourself in your own foot, if you don't have the power to close the bug
> anymore, after you've moved it...
yeah, that's a system that's somewhat controversial. there's no perfect
answer to the question 'should bugs filed on rawhide be switched to a
stable version once it's released'. At present we _do_ do this. It does
make sense sometimes; it potentially makes sense here. After all, if you
think about it, the initial bug request could be read as 'please port
this to PolicyKit 1.0 for Fedora 11', since it was filed during Fedora
11 development period.
No, not really. While this bug was filed during the F11 cycle, the plan
for PolicyKit 1 has always been to land it early in the F12 cycle. The
idea was to have most of the porting done by the time F12 opens. Just
looking at when a bug was filed does not always tell you the full story
of what version it is targeted at...
9:29 p.m.
On Wed, 2009-09-23 at 22:16 -0400, Matthias Clasen wrote:
No, not really. While this bug was filed during the F11 cycle, the
plan
for PolicyKit 1 has always been to land it early in the F12 cycle. The
idea was to have most of the porting done by the time F12 opens. Just
looking at when a bug was filed does not always tell you the full story
of what version it is targeted at...
yep, that's why I said it's controversial. neither policy will always be
correct (if we _didn't_ transfer rawhide bugs to stable releases,
there'd be cases which looked equally odd, only in the other way).
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
1:57 a.m.
You can't close a bug filed on a
stable release as CLOSED RAWHIDE. It's up to the maintainer,
now, to choose whether he intends to fix this for 11 as well
as Rawhide. If he does, he can make the fix and then do
CLOSED ERRATA. If he only wants to fix in Rawhide, he can do
CLOSED WONTFIX with a comment that the bug is fixed in
Rawhide but won't be fixed for 11.
ahem, I have to disagree a bit (or correct my understanding of
the workflow :-))
Bugzilla help at
https://bugzilla.redhat.com/page.cgi?id=fields.html#status says:
NEXTRELEASE
A fix for the problem described will be included in the next
major release of the product, and not in the release against the
release which the bug was filed. When a bug is closed to this
resolution, include information on the release in which the fix
will be included in the summary.
so, if a bug is filed on a stable release, and the developer
chooses to fix it just in rawhide, and for some reason the CLOSED
RAWHIDE state wouldn't be used, then it still falls under CLOSED
NEXTRELEASE (as the fix in rawhide is going to be cloned for the
next release on its branching)
on the other hand, CLOSED WONTFIX cannot be used:
WONTFIX
The problem described is a bug which will never be fixed. This
may happen when there is significant divergence from upstream. An
explaination of why this resolution has been chosen should be
supplied.
- if the bug is fixed in rawhide, you can't say that it will
never be fixed
AFAIK, the fact that it will never be fixed for some old version
does not matter, as upgrade is proper way of getting the bugs
fixed
K.
--
Karel Volný
QE BaseOs/Daemons Team
Red Hat Czech, Brno
tel. +420 532294274
(RH: +420 532294111 ext. 8262074)
xmpp kavol(a)jabber.cz
:: "Never attribute to malice what can
:: easily be explained by stupidity."
2:09 a.m.
On Thu, 2009-09-24 at 08:57 +0200, Karel Volný wrote:
> You can't close a bug filed on a
> stable release as CLOSED RAWHIDE. It's up to the maintainer,
> now, to choose whether he intends to fix this for 11 as well
> as Rawhide. If he does, he can make the fix and then do
> CLOSED ERRATA. If he only wants to fix in Rawhide, he can do
> CLOSED WONTFIX with a comment that the bug is fixed in
> Rawhide but won't be fixed for 11.
ahem, I have to disagree a bit (or correct my understanding of
the workflow :-))
Bugzilla help at
https://bugzilla.redhat.com/page.cgi?id=fields.html#status says:
This is not the canonical reference for Fedora, as it says at the top of
the page:
"Note: Bugs reported against Fedora products have a slightly different
life cycle which is described in more detail here."
'here' links to
https://fedoraproject.org/wiki/BugZappers/BugStatusWorkFlow . That is
the canonical reference to Fedora. The page on redhat.com is the
canonical reference for RHEL.
on the other hand, CLOSED WONTFIX cannot be used:
WONTFIX
The problem described is a bug which will never be fixed. This
may happen when there is significant divergence from upstream. An
explaination of why this resolution has been chosen should be
supplied.
- if the bug is fixed in rawhide, you can't say that it will
never be fixed
Yes, you can, if the bug is filed on a specific Fedora release which is
not Rawhide. A bug which is filed against Fedora 11 is not fixed unless
an update is released for Fedora 11. If it is fixed in Rawhide but no
update is ever shipped for F11, and the reporter and maintainer do not
agree to re-assign the bug to Rawhide, the correct resolution is
WONTFIX, as the maintainer has chosen not to fix the bug in the release
against which it was reported.
This is somewhat a consequence of the fact that Bugzilla does not work
well for tracking bugs against multiple releases of a distribution,
which leads to these somewhat odd-looking cases. But it's all we've got
to work with.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
2:19 a.m.
ok, thanks for the clarification
btw,
"The resolutions CANTFIX, WONTFIX, and WORKSFORME are for use by
maintainers only, and *are self-explanatory*."
- oh my ...
K.
Dne Čt 24. září 2009 09:09:15 Adam Williamson napsal(a):
On Thu, 2009-09-24 at 08:57 +0200, Karel Volný wrote:
> > You can't close a bug filed on a
> > stable release as CLOSED RAWHIDE. It's up to the
> > maintainer, now, to choose whether he intends to fix this
> > for 11 as well as Rawhide. If he does, he can make the fix
> > and then do CLOSED ERRATA. If he only wants to fix in
> > Rawhide, he can do CLOSED WONTFIX with a comment that the
> > bug is fixed in Rawhide but won't be fixed for 11.
>
> ahem, I have to disagree a bit (or correct my understanding
> of the workflow :-))
>
> Bugzilla help at
> https://bugzilla.redhat.com/page.cgi?id=fields.html#status
> says:
This is not the canonical reference for Fedora, as it says at
the top of the page:
"Note: Bugs reported against Fedora products have a slightly
different life cycle which is described in more detail here."
'here' links to
https://fedoraproject.org/wiki/BugZappers/BugStatusWorkFlow .
That is the canonical reference to Fedora. The page on
redhat.com is the canonical reference for RHEL.
> on the other hand, CLOSED WONTFIX cannot be used:
>
> WONTFIX
> The problem described is a bug which will never be fixed.
> This may happen when there is significant divergence from
> upstream. An explaination of why this resolution has been
> chosen should be supplied.
>
> - if the bug is fixed in rawhide, you can't say that it will
> never be fixed
Yes, you can, if the bug is filed on a specific Fedora release
which is not Rawhide. A bug which is filed against Fedora 11
is not fixed unless an update is released for Fedora 11. If
it is fixed in Rawhide but no update is ever shipped for F11,
and the reporter and maintainer do not agree to re-assign the
bug to Rawhide, the correct resolution is WONTFIX, as the
maintainer has chosen not to fix the bug in the release
against which it was reported.
This is somewhat a consequence of the fact that Bugzilla does
not work well for tracking bugs against multiple releases of
a distribution, which leads to these somewhat odd-looking
cases. But it's all we've got to work with.
--
Karel Volný
QE BaseOs/Daemons Team
Red Hat Czech, Brno
tel. +420 532294274
(RH: +420 532294111 ext. 8262074)
xmpp kavol(a)jabber.cz
:: "Never attribute to malice what can
:: easily be explained by stupidity."
9:57 a.m.
On Thu, 2009-09-24 at 09:19 +0200, Karel Volný wrote:
ok, thanks for the clarification
btw,
"The resolutions CANTFIX, WONTFIX, and WORKSFORME are for use by
maintainers only, and *are self-explanatory*."
- oh my ...
well, I should probably explain the release versioning wrinkle there,
yeah. But what they _mean_ is pretty self-explanatory. 'Can't fix',
'Won't fix', 'Works for me'. What's not to get? :)
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
2:45 a.m.
Dne Čt 24. září 2009 16:57:04 Adam Williamson napsal(a):
On Thu, 2009-09-24 at 09:19 +0200, Karel Volný wrote:
> ok, thanks for the clarification
>
> btw,
>
> "The resolutions CANTFIX, WONTFIX, and WORKSFORME are for
> use by maintainers only, and *are self-explanatory*."
>
> - oh my ...
well, I should probably explain the release versioning wrinkle
there, yeah. But what they _mean_ is pretty self-explanatory.
'Can't fix', 'Won't fix', 'Works for me'. What's not
to get?
:)
the part we already discussed: wontfix = I'll fix it, but not
backport the fix
(sorry if I broke the policy in the past because of my ignorance)
K.
--
Karel Volný
QE BaseOs/Daemons Team
Red Hat Czech, Brno
tel. +420 532294274
(RH: +420 532294111 ext. 8262074)
xmpp kavol(a)jabber.cz
:: "Never attribute to malice what can
:: easily be explained by stupidity."
9:48 a.m.
On Fri, 2009-09-25 at 09:45 +0200, Karel Volný wrote:
Dne Čt 24. září 2009 16:57:04 Adam Williamson napsal(a):
> On Thu, 2009-09-24 at 09:19 +0200, Karel Volný wrote:
> > ok, thanks for the clarification
> >
> > btw,
> >
> > "The resolutions CANTFIX, WONTFIX, and WORKSFORME are for
> > use by maintainers only, and *are self-explanatory*."
> >
> > - oh my ...
>
> well, I should probably explain the release versioning wrinkle
> there, yeah. But what they _mean_ is pretty self-explanatory.
> 'Can't fix', 'Won't fix', 'Works for me'.
What's not to get?
> :)
the part we already discussed: wontfix = I'll fix it, but not
backport the fix
As I said, that's 'the release versioning wrinkle'. I'll try and
remember to update the page.
(sorry if I broke the policy in the past because of my ignorance)
It's not a big deal. The world doesn't end if someone gets it wrong :)
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
5339
days inactive
5341
days old
11 comments
4 participants
participants (4)
-
Adam Williamson -
Karel Volný -
Matthias Clasen -
Sergey Rudchenko