11:39 p.m.
So, tor-0.2.1.25-1200.fc12 is in updates-testing for Fedora 12.
It has a subpackage, tor-lsb, which provides the standard old-style init
scripts. There's another "tor-upstart" package, which I presume does the
new
thing. Maybe soon we'll get "tor-systemd". And maybe eventually Fedora will
settle on a standard here.
For the purposes of this complaint, I don't care. I do care that whenever
you install the package, it spits out this gem:
oouch... redhat-lsb is still broken. See the report
https://bugzilla.redhat.com/show_bug.cgi?id=522053
for details.
Which points to a bug that's CLOSED|DUPLICATE of another bug that's
CLOSED|RAWHIDE.
So, clearly, there's some disagreement about what's fixed and what's broken.
But printing out a passive-agressive warning to end-users is not the
solution. The error message is confusing and very, very unhelpful. Worse,
it's not _meant_ to be helpful to the poor end user -- it's meant to try to
goad the other packager into action. Such things need to be taken up with
FESCO, not fought about in user-visible debug output.
--
Matthew Miller <mattdm(a)mattdm.org>
Senior Systems Architect -- Instructional & Research Computing Services
Harvard School of Engineering & Applied Sciences
12:06 a.m.
On Sun, May 30, 2010 at 00:39:14 -0400,
Matthew Miller <mattdm(a)mattdm.org> wrote:
So, clearly, there's some disagreement about what's fixed and what's broken.
But printing out a passive-agressive warning to end-users is not the
solution. The error message is confusing and very, very unhelpful. Worse,
it's not _meant_ to be helpful to the poor end user -- it's meant to try to
goad the other packager into action. Such things need to be taken up with
FESCO, not fought about in user-visible debug output.
See: https://fedorahosted.org/fesco/ticket/347
12:14 a.m.
New subject: tor-lsb -- hey, look, package script, don't complain to _me_. I'm just installing you.
2010/5/30 Bruno Wolff III <bruno(a)wolff.to>:
On Sun, May 30, 2010 at 00:39:14 -0400,
Matthew Miller <mattdm(a)mattdm.org> wrote:
>
> So, clearly, there's some disagreement about what's fixed and what's
broken.
> But printing out a passive-agressive warning to end-users is not the
> solution. The error message is confusing and very, very unhelpful. Worse,
> it's not _meant_ to be helpful to the poor end user -- it's meant to try to
> goad the other packager into action. Such things need to be taken up with
> FESCO, not fought about in user-visible debug output.
See: https://fedorahosted.org/fesco/ticket/347
--
devel mailing list
devel(a)lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
I don't why the tor maintainer don't want to keep consistence with
fedora package guideline and tor upstream.
12:20 a.m.
On Sun, May 30, 2010 at 12:06:12AM -0500, Bruno Wolff III wrote:
Yeah, I remember this coming up before with the issue of zillions of
dependencies.
The problem here is the output. I know (as discussed in that ticket,
actually) that the Fedora guidelines don't forbid output in the post
scripts. I think it _should_ be forbidden except in the case of errors, but
that's not the issue here. The problem is _what_ the message says, its tone,
and to whom it is addressed. All unhelpful and bad for Fedora.
--
Matthew Miller mattdm(a)mattdm.org <http://mattdm.org/>
12:38 a.m.
New subject: tor-lsb -- hey, look, package script, don't complain to _me_. I'm just installing you.
2010/5/30 Matthew Miller <mattdm(a)mattdm.org>:
On Sun, May 30, 2010 at 12:06:12AM -0500, Bruno Wolff III wrote:
> See: https://fedorahosted.org/fesco/ticket/347
Yeah, I remember this coming up before with the issue of zillions of
dependencies.
The problem here is the output. I know (as discussed in that ticket,
actually) that the Fedora guidelines don't forbid output in the post
scripts. I think it _should_ be forbidden except in the case of errors, but
that's not the issue here. The problem is _what_ the message says, its tone,
and to whom it is addressed. All unhelpful and bad for Fedora.
--
Matthew Miller mattdm(a)mattdm.org <http://mattdm.org/>
--
It's actually the same problem and both caused by the misusing of
redhat-lsb. The tor package looks very different from other daemons in
fedora, e.g. vsftpd squid etc, a small package with so many
subpackages and a metapackage seems quite strange.
Chen Lei
1:10 a.m.
On Sun, May 30, 2010 at 01:38:15PM +0800, Chen Lei wrote:
It's actually the same problem and both caused by the misusing
of
redhat-lsb. The tor package looks very different from other daemons in
fedora, e.g. vsftpd squid etc, a small package with so many
subpackages and a metapackage seems quite strange.
I'd really like to separate out that issue -- use or misuse of redhat-lsb,
packaging for generality vs. packaging for fedora -- from this one.
And this one is: packages should not print out messages complaining about
the state of other packages in Fedora. That's not the right process for
solving those issues. If redhat-lsb is broken, there's a procedure for
dealing with that, and it isn't "give confusing warnings to the end users!"
--
Matthew Miller <mattdm(a)mattdm.org>
Senior Systems Architect -- Instructional & Research Computing Services
Harvard School of Engineering & Applied Sciences
1:49 p.m.
On Sat 29 May 2010 11:10:35 pm Matthew Miller wrote:
And this one is: packages should not print out messages complaining
about
the state of other packages in Fedora. That's not the right process for
solving those issues. If redhat-lsb is broken, there's a procedure for
dealing with that, and it isn't "give confusing warnings to the end
users!"
+1
Airing out our dirty laundry for our users to see is not something that we
should allow or promote. I'm all for reporting errors, but b*tching to
users? No. I'm going to file a bug on this if someone else has not.
It's pretty non-excellent of this package's maintainer, but judging by his
previous actions I am not surprised. He has shown in the past that he has no
qualms with breaking the "be excellent to eachother" rule.
Ryan
--
Ryan Rix
== http://hackersramblings.wordpress.com | http://rix.si/ ==
== http://rix.si/page/contact/ if you need a word ==
3:55 p.m.
On Mon, 31 May 2010, Ryan Rix wrote:
Airing out our dirty laundry for our users to see is not something
that we
should allow or promote. I'm all for reporting errors, but b*tching to
users? No. I'm going to file a bug on this if someone else has not.
It's been filed many times, duplicated many times, closed many times.
https://bugzilla.redhat.com/show_bug.cgi?id=532373
I am sure there are more instances of reporting this bug.
Fixing init scripts and %post is now left in the state "maintainer is too
busy to fix". In the past I already offered co-maintainership or taking
over the package due to my close relationship with upstream. It's a lame
excuse for leaving it in the current state, since that's the preference
of the maintainer, which violates fedora packagaging policies.
Paul
5:31 p.m.
On Mon 31 May 2010 1:55:26 pm Paul Wouters wrote:
since that's the preference
of the maintainer, which violates fedora packagaging policies
Then a provenpackager should fix it regardless of whether the maintainer "is
too busy to fix it." and even then, they shouldn't be maintaining packages
they are too busy to fix! That's just as bad as blatently refusing to fix
this issue.
--
Ryan Rix
== http://hackersramblings.wordpress.com | http://rix.si/ ==
== http://rix.si/page/contact/ if you need a word ==
8:43 p.m.
New subject: tor-lsb -- hey, look, package script, don't complain to _me_. I'm just installing you.
2010/6/1 Ryan Rix <ry(a)n.rix.si>:
On Mon 31 May 2010 1:55:26 pm Paul Wouters wrote:
> since that's the preference
> of the maintainer, which violates fedora packagaging policies
Then a provenpackager should fix it regardless of whether the maintainer "is
too busy to fix it." and even then, they shouldn't be maintaining packages
they are too busy to fix! That's just as bad as blatently refusing to fix
this issue.
--
Ryan Rix
== http://hackersramblings.wordpress.com | http://rix.si/ ==
== http://rix.si/page/contact/ if you need a word ==
The maintainer refuse some others to co-maintain tor package or help
him to solve this issue. It's a bit complicated to fix this, fedora
policy seems don't permit provenpackagers to commit a package if the
maintainer are very unwilling to do so. It should be decided by fesco
in which condition that a provenpackager can commit a package
regardless the unwillingness of the package owner.
Chen Lei
7:55 p.m.
Chen Lei wrote:
The maintainer refuse some others to co-maintain tor package or help
him to solve this issue. It's a bit complicated to fix this, fedora
policy seems don't permit provenpackagers to commit a package if the
maintainer are very unwilling to do so. It should be decided by fesco
in which condition that a provenpackager can commit a package
regardless the unwillingness of the package owner.
FYI, FESCo decided on this particular issue that a provenpackager can fix
tor to comply with our initscripts guidelines for released Fedoras. (As far
as I know, the maintainer already fixed the Rawhide package.)
Kevin Kofler
9:46 p.m.
New subject: tor-lsb -- hey, look, package script, don't complain to _me_. I'm just installing you.
2010/6/2 Kevin Kofler <kevin.kofler(a)chello.at>:
Chen Lei wrote:
> The maintainer refuse some others to co-maintain tor package or help
> him to solve this issue. It's a bit complicated to fix this, fedora
> policy seems don't permit provenpackagers to commit a package if the
> maintainer are very unwilling to do so. It should be decided by fesco
> in which condition that a provenpackager can commit a package
> regardless the unwillingness of the package owner.
FYI, FESCo decided on this particular issue that a provenpackager can fix
tor to comply with our initscripts guidelines for released Fedoras. (As far
as I know, the maintainer already fixed the Rawhide package.)
Kevin Kofler
No yet, as I known:), he only add a sysv initscripr to cvs, the
package in rawhide still use -lsb and -upstart. Also the upstart
subpackage works silly, it may need further optimization or obsolete
from tor package.
See
http://koji.fedoraproject.org/koji/buildinfo?buildID=176044
http://koji.fedoraproject.org/koji/fileinfo?rpmID=1999845&filename=/e...
12:48 p.m.
On Wed, Jun 02, 2010 at 02:55:53AM +0200, Kevin Kofler wrote:
FYI, FESCo decided on this particular issue that a provenpackager can
fix
tor to comply with our initscripts guidelines for released Fedoras. (As far
as I know, the maintainer already fixed the Rawhide package.)
It's true; it is fixed in Rawhide. Okay then.
--
Matthew Miller <mattdm(a)mattdm.org>
Senior Systems Architect -- Instructional & Research Computing Services
Harvard School of Engineering & Applied Sciences
10:07 a.m.
On Mon, May 31, 2010 at 16:55:26 -0400,
Paul Wouters <paul(a)xelerance.com> wrote:
Fixing init scripts and %post is now left in the state "maintainer is too
busy to fix". In the past I already offered co-maintainership or taking
over the package due to my close relationship with upstream. It's a lame
excuse for leaving it in the current state, since that's the preference
of the maintainer, which violates fedora packagaging policies.
Does FESCO know you'd be willing to become the maintainer?
10:48 a.m.
On Tue, 1 Jun 2010, Bruno Wolff III wrote:
> Fixing init scripts and %post is now left in the state
"maintainer is too
> busy to fix". In the past I already offered co-maintainership or taking
> over the package due to my close relationship with upstream. It's a lame
> excuse for leaving it in the current state, since that's the preference
> of the maintainer, which violates fedora packagaging policies.
Does FESCO know you'd be willing to become the maintainer?
I've definately talked to quite a few of them (online and in person) over
the years this has been going on. I even had a tor package made and
submitted it, but Enrico and my package crossed paths and his was a day
earlier, so his "personal" version instead of a "fedora" version got
accepted:
https://bugzilla.redhat.com/show_bug.cgi?id=175799
https://bugzilla.redhat.com/show_bug.cgi?id=175433
https://bugzilla.redhat.com/show_bug.cgi?id=532373
I don't care who maintains it, as long as we can get the package up to
spec so upstream does not feel they need to require to tell their
users "don't use the fedora package, use our rpm". That, and the
repeated tor discussions on package guidelines violations clearly
shows a maintainer issue.
I'm getting seriously tired of this tor package discussion every six
months. Seriously, just rip out the childish %post crap, and remove
all the non-fedora initscript sub package nonsense. This is not the
Enrico Project.
Paul
10:51 a.m.
On Tue, Jun 01, 2010 at 11:48:02 -0400,
Paul Wouters <paul(a)xelerance.com> wrote:
On Tue, 1 Jun 2010, Bruno Wolff III wrote:
>>Fixing init scripts and %post is now left in the state "maintainer is too
>>busy to fix". In the past I already offered co-maintainership or taking
>>over the package due to my close relationship with upstream. It's a lame
>>excuse for leaving it in the current state, since that's the preference
>>of the maintainer, which violates fedora packagaging policies.
>
>Does FESCO know you'd be willing to become the maintainer?
I've definately talked to quite a few of them (online and in person) over
the years this has been going on. I even had a tor package made and
submitted it, but Enrico and my package crossed paths and his was a day
earlier, so his "personal" version instead of a "fedora" version got
accepted:
The reason I asked is that they might be more willing to yank the package
from the current maintainer if there is someone willing to step in and
fix things rather than having to orphan it.
11 a.m.
On Tue, 1 Jun 2010, Bruno Wolff III wrote:
>> Does FESCO know you'd be willing to become the
maintainer?
>
> I've definately talked to quite a few of them (online and in person) over
> the years this has been going on. I even had a tor package made and
> submitted it, but Enrico and my package crossed paths and his was a day
> earlier, so his "personal" version instead of a "fedora" version
got
> accepted:
The reason I asked is that they might be more willing to yank the package
from the current maintainer if there is someone willing to step in and
fix things rather than having to orphan it.
I am willing to maintain or co-maintain it, and pull it into compliance
with fedora package guidelines.
Paul
11:07 a.m.
New subject: tor-lsb -- hey, look, package script, don't complain to _me_. I'm just installing you.
2010/6/2 Paul Wouters <paul(a)xelerance.com>:
On Tue, 1 Jun 2010, Bruno Wolff III wrote:
>>> Does FESCO know you'd be willing to become the maintainer?
>>
>> I've definately talked to quite a few of them (online and in person) over
>> the years this has been going on. I even had a tor package made and
>> submitted it, but Enrico and my package crossed paths and his was a day
>> earlier, so his "personal" version instead of a "fedora"
version got
>> accepted:
>
> The reason I asked is that they might be more willing to yank the package
> from the current maintainer if there is someone willing to step in and
> fix things rather than having to orphan it.
I am willing to maintain or co-maintain it, and pull it into compliance
with fedora package guidelines.
Paul
+1 for you.
As the maintainer of vidalia and polipo, I really like to see tor
fedora to be more compliance with Fedora package guideline and the tor
package from upstream.
Chen Lei
11:04 a.m.
New subject: tor-lsb -- hey, look, package script, don't complain to _me_. I'm just installing you.
2010/6/1 Bruno Wolff III <bruno(a)wolff.to>:
> I've definately talked to quite a few of them (online and in
person) over
> the years this has been going on. I even had a tor package made and
> submitted it, but Enrico and my package crossed paths and his was a day
> earlier, so his "personal" version instead of a "fedora" version
got
> accepted:
The reason I asked is that they might be more willing to yank the package
from the current maintainer if there is someone willing to step in and
fix things rather than having to orphan it.
--
devel mailing list
devel(a)lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
I'd like to see that fesco can assign some co-maintainers for tor and
maybe some more packages from Enrico.
Regardless of violating fedora package guideline, his package style is
quite strance, e,g,
He add noarch documention to tor main package, then leave tor binary
into -core subpackage, he also add an useless upstart conf as an
alternatives to initsrcipt, the package layout is very different with
tor upstream and other packages in fedora.
Chen Lei
1:11 p.m.
On Tue 1 June 2010 8:48:02 am Paul Wouters wrote:
I'm getting seriously tired of this tor package discussion every
six
months. Seriously, just rip out the childish %post crap, and remove
all the non-fedora initscript sub package nonsense. This is not the
Enrico Project.
Halfway there, if you're up for testing:
https://bugzilla.redhat.com/show_bug.cgi?id=598213#c3
Ryan
--
Ryan Rix
== http://hackersramblings.wordpress.com | http://rix.si/ ==
== http://rix.si/page/contact/ if you need a word ==
12:25 p.m.
On Sun, 30 May 2010, Matthew Miller wrote:
For the purposes of this complaint, I don't care. I do care that
whenever
you install the package, it spits out this gem:
oouch... redhat-lsb is still broken. See the report
https://bugzilla.redhat.com/show_bug.cgi?id=522053
for details.
This comes up every six months. Everyone but one single individual agrees
with you. It just needs a provenpackager to tear it out.
Paul
5076
days inactive
5080
days old
20 comments
6 participants
participants (6)
-
Bruno Wolff III -
Chen Lei -
Kevin Kofler -
Matthew Miller -
Paul Wouters -
Ryan Rix