7:49 p.m.
My continuing adventures in fedora 14 alpha have finally
gotten it installed, and I can even boot, but it insists
on coming up in runlevel 3 (apparently). It won't start
KDM and I get no graphical login.
I have edited /etc/inittab to change :3: to :5: in the
one line in there, but that didn't change anything.
(Apparently it picked runlevel 3 as the default because
X wouldn't work in the installer and I had to use VNC).
I can't find any logs that make it appear as though it
even attempted to start X, but if I run "startx" manually
after a console login, X starts up fine and everything
appears to work, so there shouldn't be any fundamental
obstacle blocking X startup.
I'm completely mystified, and I can't find any place
that any upstart script would be starting it to even
attempt to add debug code.
8:02 p.m.
On Wed, 2010-08-25 at 20:49 -0400, Tom Horsley wrote:
My continuing adventures in fedora 14 alpha have finally
gotten it installed, and I can even boot, but it insists
on coming up in runlevel 3 (apparently). It won't start
KDM and I get no graphical login.
I have edited /etc/inittab to change :3: to :5: in the
one line in there, but that didn't change anything.
Fedora 14 is using systemd as the init system, and it doesn't
honor /etc/inittab:
https://bugzilla.redhat.com/show_bug.cgi?id=626855
--
Matt
8:18 p.m.
On Wed, 25 Aug 2010 21:02:54 -0400
Matt McCutchen wrote:
Fedora 14 is using systemd as the init system, and it doesn't
honor /etc/inittab:
https://bugzilla.redhat.com/show_bug.cgi?id=626855
http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
Another example of the Vogon effect :-).
8:34 p.m.
2010/8/26 Tom Horsley <horsley1953(a)gmail.com>:
On Wed, 25 Aug 2010 21:02:54 -0400
Matt McCutchen wrote:
> Fedora 14 is using systemd as the init system, and it doesn't
> honor /etc/inittab:
>
> https://bugzilla.redhat.com/show_bug.cgi?id=626855
http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
Another example of the Vogon effect :-).
a really good one and i can see parallels but minus the survey ;).
kind regards,
rudolf kastl
--
test mailing list
test(a)lists.fedoraproject.org
To unsubscribe:
https://admin.fedoraproject.org/mailman/listinfo/test
10:02 p.m.
On Wed, 2010-08-25 at 21:18 -0400, Tom Horsley wrote:
On Wed, 25 Aug 2010 21:02:54 -0400
Matt McCutchen wrote:
> Fedora 14 is using systemd as the init system, and it doesn't
> honor /etc/inittab:
>
> https://bugzilla.redhat.com/show_bug.cgi?id=626855
http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
Another example of the Vogon effect :-).
No, not really. It's not the Vogon effect. This is a very public, very
open project; the use of systemd has been discussed extensively on this
list, and on -test list, and is listed in the Alpha release notes - it's
the very first item under 'what's new'
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#Better_System...
- and the Alpha release announcement -
https://fedoraproject.org/wiki/F14_Alpha_release_announcement - again it's the first
item. If you're going to run a Fedora alpha without reading the release announcement,
the release notes, or either of the two mailing lists most relevant to Fedora
pre-releases, I don't think you get to complain too much.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
10:13 p.m.
On Wed, Aug 25, 2010 at 08:02:45PM -0700, Adam Williamson wrote:
On Wed, 2010-08-25 at 21:18 -0400, Tom Horsley wrote:
> On Wed, 25 Aug 2010 21:02:54 -0400
> Matt McCutchen wrote:
>
> > Fedora 14 is using systemd as the init system, and it doesn't
> > honor /etc/inittab:
> >
> > https://bugzilla.redhat.com/show_bug.cgi?id=626855
>
> http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
>
> Another example of the Vogon effect :-).
No, not really. It's not the Vogon effect. This is a very public, very
open project; the use of systemd has been discussed extensively on this
list, and on -test list, and is listed in the Alpha release notes - it's
the very first item under 'what's new'
Hrrm--Adam, I think you know how much I respect you and your efforts for
we, the users, and so also hope you'll forgive the sarcasm but....
Hrrm, somehow, I'm missing the part that says--if you want to change
your runlevel the way you've done it since it's inception, simply do....
So, I fear that I would have to say this is the Volgan effect. You
mention a change, without really saying how it will change the everyday
stuff. Actually, it implies that there won't be a difference as it says
that it maintains compatibility.
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Anya: I have witnessed a millennium of treachery and oppression
from the males of the species, and I have nothing but contempt
for the whole libidinous lot of them.
Xander: Then why are you talking to me?
Anya: I don't have a date for the prom.
Xander: Well, gosh, I wonder why not? It couldn't possibly have
anything to do with your sales pitch.
Anya: Men are evil. Will you go with me?
11:13 p.m.
On Wed, 2010-08-25 at 23:13 -0400, Scott Robbins wrote:
On Wed, Aug 25, 2010 at 08:02:45PM -0700, Adam Williamson wrote:
> On Wed, 2010-08-25 at 21:18 -0400, Tom Horsley wrote:
> > On Wed, 25 Aug 2010 21:02:54 -0400
> > Matt McCutchen wrote:
> >
> > > Fedora 14 is using systemd as the init system, and it doesn't
> > > honor /etc/inittab:
> > >
> > > https://bugzilla.redhat.com/show_bug.cgi?id=626855
> >
> > http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
> >
> > Another example of the Vogon effect :-).
>
> No, not really. It's not the Vogon effect. This is a very public, very
> open project; the use of systemd has been discussed extensively on this
> list, and on -test list, and is listed in the Alpha release notes - it's
> the very first item under 'what's new'
Hrrm--Adam, I think you know how much I respect you and your efforts for
we, the users, and so also hope you'll forgive the sarcasm but....
Hrrm, somehow, I'm missing the part that says--if you want to change
your runlevel the way you've done it since it's inception, simply do....
So, I fear that I would have to say this is the Volgan effect. You
mention a change, without really saying how it will change the everyday
stuff. Actually, it implies that there won't be a difference as it says
that it maintains compatibility.
Vogon, not Volgan. :) And yes, that issue isn't documented, mainly
because it isn't intended to be permanent (systemd will likely grow
compatibility with the runlevel specified in inittab somehow). In the
meantime, it is mentioned in the systemd FAQ, at
http://www.freedesktop.org/wiki/Software/systemd/FrequentlyAskedQuestions . I was pointing
out, though, that the fact that F14a changes the default init system *is* extensively
documented, and as your original post's reference to upstart clearly indicated, you
hadn't noticed that. If you had it would probably have been more obvious that systemd
was the reason it wasn't working, and you'd probably have found the various
explanations of the issue.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
11:39 p.m.
On Wed, Aug 25, 2010 at 21:13:20 -0700,
Adam Williamson <awilliam(a)redhat.com> wrote:
Vogon, not Volgan. :) And yes, that issue isn't documented, mainly
That's not fair. The plans for the destruction of Earth were on display.
If people couldn't be bothered to read them, that's their problem.
8:36 a.m.
On Wed, Aug 25, 2010 at 11:39 PM, Bruno Wolff III <bruno(a)wolff.to> wrote:
On Wed, Aug 25, 2010 at 21:13:20 -0700,
Adam Williamson <awilliam(a)redhat.com> wrote:
>
> Vogon, not Volgan. :) And yes, that issue isn't documented, mainly
That's not fair. The plans for the destruction of Earth were on display.
If people couldn't be bothered to read them, that's their problem.
<SNIP>
I disagree, there really is only so much we as an internet based
community can do to help the users and publicly providing the
documentation in multiple outlets is about as good as it gets. I don't
have everyone's phone number who downloaded the Alpha image so I'm not
able to call them individually and warn them of the changelog.
-AdamM
--
http://maxamillion.googlepages.com
---------------------------------------------------------
() ascii ribbon campaign - against html e-mail
/\ www.asciiribbon.org - against proprietary attachments
8:43 a.m.
On Thu, 2010-08-26 at 08:36 -0500, Adam Miller wrote:
I disagree, there really is only so much we as an internet based
community can do to help the users and publicly providing the
documentation in multiple outlets is about as good as it gets. I don't
have everyone's phone number who downloaded the Alpha image so I'm not
able to call them individually and warn them of the changelog.
I think the point is that the one thing we can do to help users is to
"throw it all away and start over" LESS often.
-sv
8:52 a.m.
On Thu, Aug 26, 2010 at 09:43:14AM -0400, seth vidal wrote:
On Thu, 2010-08-26 at 08:36 -0500, Adam Miller wrote:
> I disagree, there really is only so much we as an internet based
> community can do to help the users and publicly providing the
> documentation in multiple outlets is about as good as it gets. I don't
> have everyone's phone number who downloaded the Alpha image so I'm not
> able to call them individually and warn them of the changelog.
I think the point is that the one thing we can do to help users is to
"throw it all away and start over" LESS often.
I think the point is don't have it all over the Internet. Have it in
ONE place where people can go to check.
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Mayor Wilkins: She's pretty, Angel. A little skinny. Still
don't understand why it couldn't work out with you and my Faith. I
guess you kinda just have strange tastes in women.
Angel: Yeah, well, what can I say? I like 'em sane.
11:46 a.m.
On Thu, 2010-08-26 at 09:52 -0400, Scott Robbins wrote:
On Thu, Aug 26, 2010 at 09:43:14AM -0400, seth vidal wrote:
> On Thu, 2010-08-26 at 08:36 -0500, Adam Miller wrote:
>
> > I disagree, there really is only so much we as an internet based
> > community can do to help the users and publicly providing the
> > documentation in multiple outlets is about as good as it gets. I don't
> > have everyone's phone number who downloaded the Alpha image so I'm not
> > able to call them individually and warn them of the changelog.
>
>
> I think the point is that the one thing we can do to help users is to
> "throw it all away and start over" LESS often.
I think the point is don't have it all over the Internet. Have it in
ONE place where people can go to check.
The release notes are the one canonical thing you should always read. I
just threw in the other links to make it clear that systemd is being
listed as a New Thing in F14 just about everywhere we're *talking* about
F14.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
noon
On Thu, 2010-08-26 at 09:46 -0700, Adam Williamson wrote:
The release notes are the one canonical thing you should always read.
I have edited the release notes to state that systemd is a "mostly
compatible" replacement instead of a "drop-in" replacement, and I have
copied the FAQ and Tips and Tricks links from the release announcement.
Hopefully this will help.
--
Matt
12:26 p.m.
On Thu, 2010-08-26 at 13:00 -0400, Matt McCutchen wrote:
On Thu, 2010-08-26 at 09:46 -0700, Adam Williamson wrote:
> The release notes are the one canonical thing you should always read.
I have edited the release notes to state that systemd is a "mostly
compatible" replacement instead of a "drop-in" replacement, and I have
copied the FAQ and Tips and Tricks links from the release announcement.
Hopefully this will help.
That's good, yep. In general, though, I think you shouldn't directly
edit the release notes without consulting with the docs team first, as
they maintain such important pages. Thanks for the update.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
9:18 a.m.
2010/8/26 seth vidal <skvidal(a)fedoraproject.org>
I think the point is that the one thing we can do to help users is to
"throw it all away and start over" LESS often.
-sv
good point.
--
Among the maxims on Lord Naoshige's wall, there was this one: "Matters of
great concern should be treated lightly." Master Ittei commented, "Matters
of small concern should be treated seriously."
(Ghost Dog : The Way of The Samurai)
12:17 a.m.
On Wed, Aug 25, 2010 at 08:02:45PM -0700, Adam Williamson wrote:
On Wed, 2010-08-25 at 21:18 -0400, Tom Horsley wrote:
>
> http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
>
> Another example of the Vogon effect :-).
No, not really. It's not the Vogon effect. This is a very public,
very open project; the use of systemd has been discussed
extensively on this list, and on -test list, and is listed in the
Alpha release notes - it's the very first item under 'what's new'
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#Better_System...
- and the Alpha release announcement -
https://fedoraproject.org/wiki/F14_Alpha_release_announcement -
again it's the first item.
OK, let's try this "first item" in its entirety.
Fedora 14 introduces systemd, a smarter, more efficient way of
starting up and managing background daemons and services. It is
a drop-in replacement for sysvinit and upstart, maintaining
compatibility with SysV and LSB init scripts.
Ahem! So far AFAICT "smarter, more efficient" are quite far from
being demonstrated, "drop-in" is debatable and "maintaining
compatibility" in the curent state it certainly is not. So where
this leaves us? Searching bugzilla for systemd and reading through
search results can be an enlightening experience.
One way around these issues would be to make sure that /sbin/upstart
is installed and booting with 'init=/sbin/upstart' in kernel
parameters. Of course then you are not testing systemd. OTOH
adding '5' there may just work (maybe).
Michal
12:41 a.m.
On Wed, Aug 25, 2010 at 11:17:19PM -0600, Michal Jaegermann wrote:
On Wed, Aug 25, 2010 at 08:02:45PM -0700, Adam Williamson wrote:
> On Wed, 2010-08-25 at 21:18 -0400, Tom Horsley wrote:
> >
> > http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
> >
> > Another example of the Vogon effect :-).
>
> No, not really. It's not the Vogon effect. This is a very public,
> very open project; the use of systemd has been discussed
> extensively on this list, and on -test list, and is listed in the
> Alpha release notes - it's the very first item under 'what's new'
>
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#Better_System...
> - and the Alpha release announcement -
> https://fedoraproject.org/wiki/F14_Alpha_release_announcement -
> again it's the first item.
I reckon that perhaps there should be, somehow, a place--one
place--where people could always go to find the things that violate POLA
(for those who missed earlier installments, Principle of Least
Astonishment). Granted, there would always be something overlooked, and
what might astonish some wouldn't astonish others.
As it is, people are, as Tom's Vogon (sorry about the Volgan to all you
Vogans who may have been offended) page indicates, are supposed to
follow X and Y and Z and perhaps Q, and if it's in any of those 4-10
places (generally speaking, I realize I only specified 4), it's their
fault.
Perhaps an announcements mailing list, read only, which could have had
something like, We have this great new thing, systemd--at present, it
means you can't change runlevels in inittab so to do that, edit $FILE.
Looking at the page originally mentioned,
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#Better_System...
such a list (or page) DOESN'T NEED the things the kernel does (the
several points below), only needs mention of spice if it is going to
completely change the way KVM now works, no need to mention the JPEG
stuff, or even mention of the D language unless it breaks C and C++. In
other words, a page that just mentions, Ok this thing that you've always
done will no longer work. Do this instead.
When someone says, hey, what's this spice stuff, how come I didn't hear
about it, THEN, one can point them to the relevant pages or links, but
it's not something that will make the way they do what they've been
doing no longer work.
(Or point them to a Spice Girls video, of course, but I digress.)
It seems that in Open Source, the onus is always on the user--the way
Apple put the blame on its users to hold their iPhones in a certain way,
and if it dropped calls, well, it's your fault for the way you hold it,
even though that way has worked for you since the beginning of the iPhone.
It's not even in this case, a matter of assigning blame--it's just a
suggestion to find a place or mailing list to be sure to mention what
will break--and only what will break, so that people who wish to use
Fedora don't feel they have to subscribe to 4 lists and constantly check
4 pages, nor wade through a bunch of things that won't affect them.
Even the release notes will have a what's changed, but the vast majority
of it won't necessarily affect how people HAVE to do things (as opposed
to how they can now do things.)
FreeBSD has a couple of lists (though these aren't announce only, but
they have a different, and far smaller, user base) where when major
changes are coming, they're announced with a subject line beginning
HEADSUP. They also will have an UPDATING section that is updated with
each upgrade, as well as a second UPDATING in their ports (3rd party
programs) section, so that if you miss it, it is fair to put the onus on
you. I'm not saying that's practical for Fedora, but simply to mention
the sort of things that should be covered.
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Willow: Why couldn't he be possessed by a puppy, or some
ducks?
3:10 a.m.
On 08/26/2010 07:17 AM, Michal Jaegermann wrote:
On Wed, Aug 25, 2010 at 08:02:45PM -0700, Adam Williamson wrote:
> On Wed, 2010-08-25 at 21:18 -0400, Tom Horsley wrote:
>>
>> http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
>>
>> Another example of the Vogon effect :-).
>
> No, not really. It's not the Vogon effect. This is a very public,
> very open project; the use of systemd has been discussed
> extensively on this list, and on -test list, and is listed in the
> Alpha release notes - it's the very first item under 'what's new'
>
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#Better_System...
> - and the Alpha release announcement -
> https://fedoraproject.org/wiki/F14_Alpha_release_announcement -
> again it's the first item.
OK, let's try this "first item" in its entirety.
Fedora 14 introduces systemd, a smarter, more efficient way of
starting up and managing background daemons and services. It is
a drop-in replacement for sysvinit and upstart, maintaining
compatibility with SysV and LSB init scripts.
Ahem! So far AFAICT "smarter, more efficient" are quite far from
being demonstrated, "drop-in" is debatable and "maintaining
compatibility" in the curent state it certainly is not. So where
this leaves us? Searching bugzilla for systemd and reading through
search results can be an enlightening experience.
One way around these issues would be to make sure that /sbin/upstart
is installed and booting with 'init=/sbin/upstart' in kernel
parameters. Of course then you are not testing systemd. OTOH
adding '5' there may just work (maybe).
Michal
Hi Michal,
I made the experience that I can add in F14 the runlevel in the GRUB
boot line, so I can boot into rl 5 or 3 without problem. Without any
change in /etc/inittab.
Kind regards
--
Joachim Backes <joachim.backes(a)rhrk.uni-kl.de>
http://www.rhrk.uni-kl.de/~backes
9:06 a.m.
On Thu, Aug 26, 2010 at 10:10:28AM +0200, Joachim Backes wrote:
On 08/26/2010 07:17 AM, Michal Jaegermann wrote:
> OTOH adding '5' there may just work (maybe).
I made the experience that I can add in F14 the runlevel in the GRUB
boot line, so I can boot into rl 5 or 3 without problem. Without any
change in /etc/inittab.
That 3 in a boot command works I know; with 5 I never tried.
Switching runlevels once you booted, to other than 0 or 6, currently
ranges from difficult to impossible - depending on what you are
trying to achieve - but hopefuly that will get fixed.
https://bugzilla.redhat.com/show_bug.cgi?id=627014
Michal
8:47 a.m.
On Thu, Aug 26, 2010 at 12:17 AM, Michal Jaegermann <michal(a)harddata.com> wrote:
On Wed, Aug 25, 2010 at 08:02:45PM -0700, Adam Williamson wrote:
> On Wed, 2010-08-25 at 21:18 -0400, Tom Horsley wrote:
> >
> > http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
> >
> > Another example of the Vogon effect :-).
>
> No, not really. It's not the Vogon effect. This is a very public,
> very open project; the use of systemd has been discussed
> extensively on this list, and on -test list, and is listed in the
> Alpha release notes - it's the very first item under 'what's new'
>
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#Better_System...
> - and the Alpha release announcement -
> https://fedoraproject.org/wiki/F14_Alpha_release_announcement -
> again it's the first item.
OK, let's try this "first item" in its entirety.
Fedora 14 introduces systemd, a smarter, more efficient way of
starting up and managing background daemons and services. It is
a drop-in replacement for sysvinit and upstart, maintaining
compatibility with SysV and LSB init scripts.
Ahem! So far AFAICT "smarter, more efficient" are quite far from
being demonstrated, "drop-in" is debatable and "maintaining
compatibility" in the curent state it certainly is not. So where
this leaves us? Searching bugzilla for systemd and reading through
search results can be an enlightening experience.
<SNIP>
Oh, I'm sorry ... did you miss the Alpha tag on the ISO you
downloaded? Did you not see Anaconda pop up a warning to let you know
that this is not a stable or completed release?
I understand everyone's frustrations, but in reality there are always
going to be growing pains when new tech is introduced, especially such
drastic changes as ripping and replacing the init system. We as a
community, especially as a QA community, should understand that by now
and trust in the developers who are working so hard to make these
innovations. Those of us on this mailing list spend time testing
Fedora and we should provide *constructive* feedback, triage bugs, and
all around try our hardest to help make sure Fedora is the best it can
be in any way we can. There *are* going to be bugs, that's just part
of the territory of writing software.
-AdamM
--
http://maxamillion.googlepages.com
---------------------------------------------------------
() ascii ribbon campaign - against html e-mail
/\ www.asciiribbon.org - against proprietary attachments
9:11 a.m.
On Thu, 26 Aug 2010 08:47:17 -0500
Adam Miller wrote:
Those of us on this mailing list spend time testing
Fedora and we should provide *constructive* feedback
I thought the suggestion earlier in this thread to provide
a single place where "heads up" documentation about things
users are going to be forced to change was constructive. I'd
like to see it as the very first chunk of text in the
release notes. And complaints from alpha testers about
undocumented surprises are a good way to find the list of
things that belong in that "heads up" section.
9:12 a.m.
On Thu, Aug 26, 2010 at 08:47:17AM -0500, Adam Miller wrote:
>
> OK, let's try this "first item" in its entirety.
Oh, I'm sorry ... did you miss the Alpha tag on the ISO you
downloaded?
Somebody who is missing something is you; namely that an information
to deal with a fall out is lacking. A piece of a marketting fluff
instead is a rather poor substitute.
Michal
9:17 a.m.
On Thu, 2010-08-26 at 08:12 -0600, Michal Jaegermann wrote:
On Thu, Aug 26, 2010 at 08:47:17AM -0500, Adam Miller wrote:
> >
> > OK, let's try this "first item" in its entirety.
>
> Oh, I'm sorry ... did you miss the Alpha tag on the ISO you
> downloaded?
Somebody who is missing something is you; namely that an information
to deal with a fall out is lacking. A piece of a marketting fluff
instead is a rather poor substitute.
I believe its a wiki. Please go there and help improve the release
notes.
9:27 a.m.
On Thu, Aug 26, 2010 at 10:17:23AM -0400, Matthias Clasen wrote:
On Thu, 2010-08-26 at 08:12 -0600, Michal Jaegermann wrote:
> On Thu, Aug 26, 2010 at 08:47:17AM -0500, Adam Miller wrote:
> > >
> > > OK, let's try this "first item" in its entirety.
> >
> > Oh, I'm sorry ... did you miss the Alpha tag on the ISO you
> > downloaded?
>
> Somebody who is missing something is you; namely that an information
> to deal with a fall out is lacking. A piece of a marketting fluff
> instead is a rather poor substitute.
I believe its a wiki. Please go there and help improve the release
notes.
The notes already have a link to the common bugs page, which now has a
section for the systemd v8 issue:
https://fedoraproject.org/wiki/Common_F14_bugs#systemd_package_update_bre...
The QA team is going to check out what I drafted there.
From a documentation point of view it doesn't make sense to be
updating two pages with the same detailed information about specific
issues, since they invariably will get out of sync.
--
Paul W. Frields http://paul.frields.org/
gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233 5906 ACDB C937 BD11 3717
http://redhat.com/ - - - - http://pfrields.fedorapeople.org/
Where open source multiplies: http://opensource.com
9:33 a.m.
On Thu, 2010-08-26 at 10:27 -0400, Paul W. Frields wrote:
On Thu, Aug 26, 2010 at 10:17:23AM -0400, Matthias Clasen wrote:
> On Thu, 2010-08-26 at 08:12 -0600, Michal Jaegermann wrote:
> > Somebody who is missing something is you; namely that an information
> > to deal with a fall out is lacking. A piece of a marketting fluff
> > instead is a rather poor substitute.
>
> I believe its a wiki. Please go there and help improve the release
> notes.
The notes already have a link to the common bugs page, which now has a
section for the systemd v8 issue:
https://fedoraproject.org/wiki/Common_F14_bugs#systemd_package_update_bre...
This thread is about systemd not honoring /etc/inittab, not the issue
with systemd version 8.
--
Matt
12:25 p.m.
On Thu, 2010-08-26 at 10:27 -0400, Paul W. Frields wrote:
> I believe its a wiki. Please go there and help improve the
release
> notes.
The notes already have a link to the common bugs page, which now has a
section for the systemd v8 issue:
https://fedoraproject.org/wiki/Common_F14_bugs#systemd_package_update_bre...
The QA team is going to check out what I drafted there.
>From a documentation point of view it doesn't make sense to be
updating two pages with the same detailed information about specific
issues, since they invariably will get out of sync.
There's a clear demarcation between release notes and common bugs.
Release notes documents *intentional* changes; common bugs documents
*unintentional* changes (AKA bugs). :)
The fact that we have a brand new init system with some changes to how
it should be configured is not a bug, it's a new feature: it goes in
release notes. The fact that /etc/inittab both doesn't work to set
default runlevel any more and there's no comment in it to indicate this
is a bug, so it goes on common bugs; we will certainly either make it
work or add a comment to the file alerting you to the fact that it
doesn't work any more, and you should use another method to change the
default boot target (systemd's equivalent of runlevels). So that can be
mentioned in common bugs. I'll add it today, if no-one else does.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
2:16 p.m.
On Thu, Aug 26, 2010 at 10:27:51AM -0400, Paul W. Frields wrote:
section for the systemd v8 issue:
https://fedoraproject.org/wiki/Common_F14_bugs#systemd_package_update_bre...
The "less fragile update" should be pushed to the testing repo RSN.
--
Matthew Miller <mattdm(a)mattdm.org>
Senior Systems Architect -- Instructional & Research Computing Services
Harvard School of Engineering & Applied Sciences
9:41 a.m.
On Thu, Aug 26, 2010 at 10:17:23AM -0400, Matthias Clasen wrote:
On Thu, 2010-08-26 at 08:12 -0600, Michal Jaegermann wrote:
> On Thu, Aug 26, 2010 at 08:47:17AM -0500, Adam Miller wrote:
> > >
> > > OK, let's try this "first item" in its entirety.
> >
> > Oh, I'm sorry ... did you miss the Alpha tag on the ISO you
> > downloaded?
>
> Somebody who is missing something is you; namely that an information
> to deal with a fall out is lacking. A piece of a marketting fluff
> instead is a rather poor substitute.
I believe its a wiki. Please go there and help improve the release
notes.
Do you happen to mean that:
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#Better_System...
I just quoted the whole entry. If somewhere else there is something
"real" then I would start by putting a reference to this into
release notes.
Or maybe you were thinking of this:
https://fedoraproject.org/wiki/F14_Alpha_release_announcement
NetworkManager and PulseAudio as examples in the context are
positively scary. Do I need to explain why?
Michal
10:01 a.m.
On Thu, 2010-08-26 at 08:41 -0600, Michal Jaegermann wrote:
Do you happen to mean that:
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#Better_System...
I just quoted the whole entry. If somewhere else there is something
"real" then I would start by putting a reference to this into
release notes.
I was not thinking about any specific page, I was just trying to move
this discussion away from complaining and toward contributing.
Or maybe you were thinking of this:
https://fedoraproject.org/wiki/F14_Alpha_release_announcement
NetworkManager and PulseAudio as examples in the context are
positively scary. Do I need to explain why?
I do not think the examples are scary. But no thanks, I don't want you
to explain either.
11:07 a.m.
On Thu, Aug 26, 2010 at 11:01:02AM -0400, Matthias Clasen wrote:
On Thu, 2010-08-26 at 08:41 -0600, Michal Jaegermann wrote:
> Or maybe you were thinking of this:
> https://fedoraproject.org/wiki/F14_Alpha_release_announcement
> NetworkManager and PulseAudio as examples in the context are
> positively scary. Do I need to explain why?
I do not think the examples are scary.
Well, both NetworkManager and PulseAudio are not trouble free
after long time and many iterations although nowadays _usually_
they tend to work if a situation is not too demanding. We are
talking here about such basic thing as booting. Maybe you are
substantially braver then me.
But no thanks, I don't want you to explain either.
OK, I will stay away from specific examples.
Michal
3 p.m.
On Thu, Aug 26, 2010 at 8:11 PM, Michal Jaegermann wrote:
Do you happen to mean that:
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#Better_System...
I just quoted the whole entry. If somewhere else there is something
"real" then I would start by putting a reference to this into
release notes.
Everyone is free to add "real" info to the same page. I have done so
already FYI. Took less time than the back and forth going on here.
Rahul
3:34 p.m.
On Sun, Aug 29, 2010 at 01:30:35AM +0530, Rahul Sundaram wrote:
On Thu, Aug 26, 2010 at 8:11 PM, Michal Jaegermann wrote:
Do you happen to mean that:
http://fedoraproject.org/wiki/Fedora_14_Alpha_release_notes#
Better_System_and_Session_Management
I just quoted the whole entry. If somewhere else there is something
"real" then I would start by putting a reference to this into
release notes.
Everyone is free to add "real" info to the same page.
It helps immensly if you are in a position to distinguish between
"real" info, a pure voodoo magic which once gave an impression that
it worked, accidental and/or intentional workarounds for bugs which
will be gone in some future version - maybe, misfeatures and etc.,
etc...
I try to document encountered problems in bugzilla.
Michal
5:39 p.m.
On 08/29/2010 02:04 AM, Michal Jaegermann wrote:
It helps immensly if you are in a position to distinguish between
"real" info, a pure voodoo magic which once gave an impression that
it worked, accidental and/or intentional workarounds for bugs which
will be gone in some future version - maybe, misfeatures and etc.,
etc...
I try to document encountered problems in bugzilla.
In this case, the problems were already documented in bugzilla and the
solutions to the problems were also available there. It just requires
someone to take the next step and put up the info in a more accessible
way for other testers. Again, the time required to do that is pretty
small (maybe as small as ten minutes) and I do hope more testers take
that initiative.
Rahul
4:45 a.m.
On Sunday 29 of August 2010 00:39:52 Rahul Sundaram wrote:
Again, the time required to do that is pretty small (maybe as
small as ten minutes)
... if you already have all the background knowledge, and the
needed abilities
don't forget about that, please
(sorry, I'm a bit tired of hearing "oh, you just need 5 minutes
to do this, why don't you do it?" ... someone can spend 5 minutes
on something, someone else can spend 5 hours on the same thing
and in the end it is screwed - reminds me of trying to draw a
simple SVG which ended in studying the specs and writing the
source in text editor ...)
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."
7:13 a.m.
On 08/30/2010 03:15 PM, Karel Volný wrote:
On Sunday 29 of August 2010 00:39:52 Rahul Sundaram wrote:
> Again, the time required to do that is pretty small (maybe as
> small as ten minutes)
... if you already have all the background knowledge, and the
needed abilities
don't forget about that, please
I have done the search, found the info and written up the documentation
and didn't take much up much time either. Fedora 11 Ext4 FAQ for
example. I will likely do it for systemd as well if noone else steps
up but I don't have any unique abilities you don't. It is just some
time and interest to do it.
Rahul
7:37 a.m.
hi,
On Thursday 26 of August 2010 05:02:45 Adam Williamson wrote:
No, not really. It's not the Vogon effect. This is a very
public, very open project; the use of systemd has been
discussed extensively on this list, and on -test list, and is
listed in the Alpha release notes - it's the very first item
under 'what's new'
ok, great ... I remember seeing the info that there will be some
"systemd" thing in F14
but I really do not remember seeing the info that inittab won't
be usable any longer
see the difference?
- I guess Tom has the right to complain
and btw, I don't see it mentioned here nor in the referenced
bugreport - why /etc/inittab gets installed at all if it is not
used?
(I've run into that issue too, but as I was interested in another
problem, I simply ignored that ... if inittab wouldn't be in its
place, I'd go ask "what's new?", but it was there and hasn't
worked, so I thought "oh, there is a bug")
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."
9:33 a.m.
On Fri, Aug 27, 2010 at 02:37:08PM +0200, Karel Volný wrote:
and btw, I don't see it mentioned here nor in the referenced
bugreport - why /etc/inittab gets installed at all if it is not
used?
It's used by upstart, which is currently installed in parallel as a
fallback. Confusing, I know.
--
Matthew Miller <mattdm(a)mattdm.org>
Senior Systems Architect -- Instructional & Research Computing Services
Harvard School of Engineering & Applied Sciences
10:48 a.m.
On Fri, 2010-08-27 at 14:37 +0200, Karel Volný wrote:
ok, great ... I remember seeing the info that there will be some
"systemd" thing in F14
but I really do not remember seeing the info that inittab won't
be usable any longer
see the difference?
Yes, and we already addressed that earlier. It's an *Alpha*. Stuff gets
broken.
- I guess Tom has the right to complain
and btw, I don't see it mentioned here nor in the referenced
bugreport - why /etc/inittab gets installed at all if it is not
used?
Because it's an Alpha, and not everything is fixed yet.
(I've run into that issue too, but as I was interested in another
problem, I simply ignored that ... if inittab wouldn't be in its
place, I'd go ask "what's new?", but it was there and hasn't
worked, so I thought "oh, there is a bug")
K.
--
test mailing list
test(a)lists.fedoraproject.org
To unsubscribe:
https://admin.fedoraproject.org/mailman/listinfo/test
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net
4:08 p.m.
Tom Horsley wrote:
On Wed, 25 Aug 2010 21:02:54 -0400
Matt McCutchen wrote:
> Fedora 14 is using systemd as the init system, and it doesn't
> honor /etc/inittab:
>
> https://bugzilla.redhat.com/show_bug.cgi?id=626855
http://home.comcast.net/~tomhorsley/wisdom/braindump/oss-happens.html
Another example of the Vogon effect :-).
I sent that link to about a dozen developers and admins I've worked with over
the years. Totally priceless, and unfortunately spot on.
--
Bill Davidsen <davidsen(a)tmr.com>
"We have more to fear from the bungling of the incompetent than from
the machinations of the wicked." - from Slashdot
8:30 p.m.
On Wed, Aug 25, 2010 at 09:02:54PM -0400, Matt McCutchen wrote:
On Wed, 2010-08-25 at 20:49 -0400, Tom Horsley wrote:
> My continuing adventures in fedora 14 alpha have finally
> gotten it installed, and I can even boot, but it insists
> on coming up in runlevel 3 (apparently). It won't start
> KDM and I get no graphical login.
>
> I have edited /etc/inittab to change :3: to :5: in the
> one line in there, but that didn't change anything.
I remember this happening with upstart--briefly, and I mentioned that a
comment should go in /etc/inittab, and I think Bill Nottingham fixed it
in a few hours (without me even filing a bug.)
As long as there's a comment in /etc/inittab on how to fix it, should be
alright.
It's one of the things that if it's changed, will surprise a lot of
people--putting the onus on the user to have read x, y, or z to know
about it is not fair. The natural place to look is in /etc/inittab. If
changing it there doesn't fix it, they have to start to google, which is
disrespectful of the user's time, especially when a line or two in
inittab will not take much time to put in.
I agree with Tom's comment in the bug report. Put the comment in
inittab, as that's where people will look if it suddenly starts booting
into an unexpected runlevel. (not that he was the only one to suggest
it of course.)
From what I see of the bug report, Bill is going to fix it shortly.
--
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6
Willow: Xander, wanna stay and help me?
Xander: Are you kidding?
Willow: Yes, it was a joke I made up.
2:02 a.m.
On 08/26/2010 07:00 AM, Scott Robbins wrote:
I remember this happening with upstart--briefly, and I mentioned that
a
comment should go in /etc/inittab, and I think Bill Nottingham fixed it
in a few hours (without me even filing a bug.)
As long as there's a comment in /etc/inittab on how to fix it, should be
alright.
This has been suggested by me earlier in bugzilla. However note that
the decision on whether to go with systemd for Fedora 14 general release
has not been taken yet and some of the details (syntax etc) are not yet
set in stone. So considering it is just a alpha release, it would be
useful to give it more time rather than rant. systemd is extensively
documented and I don't think one can complain about that much.
Rahul
7:03 a.m.
On 08/26/2010 12:49 AM, Tom Horsley wrote:
My continuing adventures in fedora 14 alpha have finally
gotten it installed, and I can even boot, but it insists
on coming up in runlevel 3 (apparently). It won't start
KDM and I get no graphical login.
What happens if you start kdm in run level 3
I have edited /etc/inittab to change :3: to :5: in the
one line in there, but that didn't change anything.
(Apparently it picked runlevel 3 as the default because
X wouldn't work in the installer and I had to use VNC).
That wont work inittab is slowly being deprecated add 1, 3 or 5 to
kernel command line during start or in grub.conf directly instead.
So it seems you have encountered two bugs one in Anaconda and another in
KDM
I can't find any logs that make it appear as though it
even attempted to start X, but if I run "startx" manually
after a console login, X starts up fine and everything
appears to work, so there shouldn't be any fundamental
obstacle blocking X startup.
The problem seems to be KDM not starting
JBG
9:14 a.m.
> I have edited /etc/inittab to change :3: to :5: in the
> one line in there, but that didn't change anything.
> (Apparently it picked runlevel 3 as the default because
> X wouldn't work in the installer and I had to use VNC).
>
That wont work inittab is slowly being deprecated add 1, 3 or 5 to
kernel command line during start or in grub.conf directly instead.
So it seems you have encountered two bugs one in Anaconda and another in
KDM
The only possible anaconda bug here is that X "doesn't work", but we
don't have enough data to go on to know whether that's anaconda's fault
or X's fault or what.
The VNC -> runlevel 3 implication is working as designed.
- Chris
9:18 a.m.
On Thu, 26 Aug 2010 10:14:59 -0400
Chris Lumens wrote:
The VNC -> runlevel 3 implication is working as designed.
Though, I'd really like anaconda to have a screen that
let's me choose the runlevel during install, but it isn't
a big deal (until changing the runlevel becomes an
adventure :-).
12:06 p.m.
On Thu, 2010-08-26 at 10:18 -0400, Tom Horsley wrote:
On Thu, 26 Aug 2010 10:14:59 -0400
Chris Lumens wrote:
> The VNC -> runlevel 3 implication is working as designed.
Though, I'd really like anaconda to have a screen that
let's me choose the runlevel during install, but it isn't
a big deal (until changing the runlevel becomes an
adventure :-).
The theory here is that if graphics didn't work at install time they're
probably not going to work afterwards either. So if you needed to
resort to VNC or text mode...
- ajax
12:59 p.m.
On Thu, 26 Aug 2010 13:06:55 -0400
Adam Jackson wrote:
The theory here is that if graphics didn't work at install time
they're
probably not going to work afterwards either. So if you needed to
resort to VNC or text mode...
Yea, but my most frequent use case is to do a graphical install
on a machine that is really going to be a server which I would
like to default to runlevel 3 instead of 5.
2:18 p.m.
On Thu, Aug 26, 2010 at 01:59:01PM -0400, Tom Horsley wrote:
Yea, but my most frequent use case is to do a graphical install
on a machine that is really going to be a server which I would
like to default to runlevel 3 instead of 5.
Note that this is a common use case, because the text-gui installer does not
implement all of the functionality of the X-gui one. Notably for servers: no
disk layout tool.
--
Matthew Miller <mattdm(a)mattdm.org>
Senior Systems Architect -- Instructional & Research Computing Services
Harvard School of Engineering & Applied Sciences
4980
days inactive
4992
days old
46 comments
20 participants
participants (20)
-
Adam Jackson -
Adam Miller -
Adam Williamson -
Bill Davidsen -
Bruno Wolff III -
Chris Lumens -
cornel panceac -
Joachim Backes -
Jóhann B. Guðmundsson -
Karel Volný -
Matt McCutchen -
Matthew Miller -
Matthias Clasen -
Michal Jaegermann -
Paul W. Frields -
Rahul Sundaram -
Rudolf Kastl -
Scott Robbins -
Seth Vidal -
Tom Horsley