Re: [gentoo-user] Switching to unstable

2010-04-14 Thread Alan McKinnon
On Tuesday 13 April 2010 16:49:44 Helmut Jarausch wrote:
> On 13 Apr, Peter Humphrey wrote:
> > On Monday 12 April 2010 16:55:38 Paul Hartman wrote:
> >> I've been using portage unmasked for a very long time and don't
> >> remember having any portage-related problems. I'm sure there must be
> >> some (or else why is it still RC?) but for me the new features are
> >> worth the potential risk of using less-tested code.
> > 
> > There was a problem with its preserved-rebuild feature for a while;
> > several people reported here that they were running it and being told
> > they still needed to run it again.
> > 
> > As far as I know, that's the only thing preventing release of v2, and I
> > think it's been fixed anyway.
> 
> No, I don't think so.
> Just recently, I had to unmerge then emerge wxpython since
> emerge @preserved-rebuild
> couldn't solve it itself.
> 
> One more "buglet".
> When doing  emerge -j 
> - which is a very useful feature on a multicore machine -
> sometime a package just stops to build (which is reported
> as failing package).
> Just emerge it again.
> So, I'd say some new features are not ready, yet, but still
> very useful as they are.
> Helmut.

Those sound like ebuild and build system bugs, not something with portage 
itself.

Especially the -j bug - sounds like a race condition caused by a pathetic 
Makefile

-- 
alan dot mckinnon at gmail dot com



Re: [gentoo-user] Switching to unstable

2010-04-13 Thread Peter Humphrey
On Tuesday 13 April 2010 15:49:44 Helmut Jarausch wrote:
> On 13 Apr, Peter Humphrey wrote:
> > As far as I know, that's the only thing preventing release of v2,
> > and I think it's been fixed anyway.
> 
> No, I don't think so.
> Just recently, I had to unmerge then emerge wxpython since
> emerge @preserved-rebuild
> couldn't solve it itself.

Ah. I supposed that it had been fixed as I haven't seen it recently.
 
> One more "buglet".
> When doing  emerge -j 
> - which is a very useful feature on a multicore machine -
> sometime a package just stops to build (which is reported
> as failing package).
> Just emerge it again.

I've noticed that too. I don't know where the problem lies, but as you 
say, it's easily evaded.

> So, I'd say some new features are not ready, yet, but still
> very useful as they are.

Agreed.

-- 
Rgds
Peter.



Re: [gentoo-user] Switching to unstable

2010-04-13 Thread Helmut Jarausch
On 13 Apr, Peter Humphrey wrote:
> On Monday 12 April 2010 16:55:38 Paul Hartman wrote:
> 
>> I've been using portage unmasked for a very long time and don't
>> remember having any portage-related problems. I'm sure there must be
>> some (or else why is it still RC?) but for me the new features are
>> worth the potential risk of using less-tested code.
> 
> There was a problem with its preserved-rebuild feature for a while; 
> several people reported here that they were running it and being told 
> they still needed to run it again.
> 
> As far as I know, that's the only thing preventing release of v2, and I 
> think it's been fixed anyway.
> 

No, I don't think so.
Just recently, I had to unmerge then emerge wxpython since
emerge @preserved-rebuild
couldn't solve it itself.

One more "buglet".
When doing  emerge -j 
- which is a very useful feature on a multicore machine -
sometime a package just stops to build (which is reported
as failing package).
Just emerge it again.
So, I'd say some new features are not ready, yet, but still
very useful as they are.
Helmut.

-- 
Helmut Jarausch

Lehrstuhl fuer Numerische Mathematik
RWTH - Aachen University
D 52056 Aachen, Germany



Re: [gentoo-user] Switching to unstable

2010-04-13 Thread Peter Humphrey
On Monday 12 April 2010 16:55:38 Paul Hartman wrote:

> I've been using portage unmasked for a very long time and don't
> remember having any portage-related problems. I'm sure there must be
> some (or else why is it still RC?) but for me the new features are
> worth the potential risk of using less-tested code.

There was a problem with its preserved-rebuild feature for a while; 
several people reported here that they were running it and being told 
they still needed to run it again.

As far as I know, that's the only thing preventing release of v2, and I 
think it's been fixed anyway.

-- 
Rgds
Peter.



Re: [gentoo-user] Switching to unstable

2010-04-13 Thread Dale

Tanstaafl wrote:

On 2010-04-12 12:23 PM, Dale wrote:
   

+1  I been using the latest portage for a long time too.  I don't recall
any problems with it and the new features sure do help.

If you keyword portage, you need to do the same for its friends.  Mainly
gentoolkit and eix.  They seem to go together better.  If you run one
without the other, it can do some weird things.
 

Ok, I'm seriously considering it... thanks. Are those really the only 3?

Thanks again...

   


Those are the most commonly used portage type packages that I use 
anyway.  I know recently I had portage running on the latest then 
equery  got a tummy ache and sort of puked at me.  So, I had to get the 
latest for equery too.  Then I needed the latest eix as well.  It's been 
pretty much happy since then tho.


It works fine tho.  They seem to test portage pretty well before it hits 
the tree.


Dale

:-)  :-)



Re: [gentoo-user] Switching to unstable

2010-04-12 Thread Tanstaafl
On 2010-04-12 12:23 PM, Dale wrote:
> +1  I been using the latest portage for a long time too.  I don't recall
> any problems with it and the new features sure do help.
> 
> If you keyword portage, you need to do the same for its friends.  Mainly
> gentoolkit and eix.  They seem to go together better.  If you run one
> without the other, it can do some weird things.

Ok, I'm seriously considering it... thanks. Are those really the only 3?

Thanks again...

-- 

Charles



Re: [gentoo-user] Switching to unstable

2010-04-12 Thread Mark Knecht
On Mon, Apr 12, 2010 at 12:06 AM, Marc Joliet  wrote:
> Am Sun, 11 Apr 2010 17:13:26 -0700
> schrieb Mark Knecht :
>
>> On Sun, Apr 11, 2010 at 5:00 PM, Neil Bothwick  wrote:
>> > On Sun, 11 Apr 2010 16:11:53 -0700, Mark Knecht wrote:
>> >
>> >> > Doing system first makes good sense. Then you can update your config
>> >> > files, follow the openrc update etc and then reboot. The world part of
>> >> > the update will take quite a while, especially if you use KDE or
>> >> > GNOME.
>> >>
>> >>  Less than most PC. It's an i7-980x with 24GB of RAM with RAID1.
>> >
>
> Gah, I'm getting PC envy ;-) !
>
>> A couple of packages in this OpenRC upgrade aren't building. I hope
>> they are less important. So far groff and help2man have failed so I
>> did --resume --skip-first and moved on for now.
>
> Just a note: you can also specify "--keep-going" and portage will do that for
> you and even recalculate the dependencies before continuing. After it's done,
> it gives you a message listing the packages that failed.
>

Thanks for pointing that out. When doing a long set it means I don't
have to keep coming in here to make sure it's doing the most it can.
Nice addition.

Cheers,
Mark



Re: [gentoo-user] Switching to unstable

2010-04-12 Thread Dale

Paul Hartman wrote:

On Mon, Apr 12, 2010 at 10:18 AM, Tanstaafl  wrote:
   

On 2010-04-12 11:05 AM, Paul Hartman wrote:
 

I use the --keep-going always, it was a great addition and especially
helpful when there is a bad package that won't compile for a week or
two, it makes it easier to just ignore it.
   

Hopefully no one will mind a slight OT question, but still related...

Is the testing version of portage 2.2 stable enough for production
machines? There are a number of new features I'd like to take advantage
of, but have always hesitated to and any non-stable system packages.
 

I've been using portage unmasked for a very long time and don't
remember having any portage-related problems. I'm sure there must be
some (or else why is it still RC?) but for me the new features are
worth the potential risk of using less-tested code.

   


+1  I been using the latest portage for a long time too.  I don't recall 
any problems with it and the new features sure do help.


If you keyword portage, you need to do the same for its friends.  Mainly 
gentoolkit and eix.  They seem to go together better.  If you run one 
without the other, it can do some weird things.


Dale

:-)  :-)



Re: [gentoo-user] Switching to unstable

2010-04-12 Thread Paul Hartman
On Mon, Apr 12, 2010 at 10:18 AM, Tanstaafl  wrote:
> On 2010-04-12 11:05 AM, Paul Hartman wrote:
>> I use the --keep-going always, it was a great addition and especially
>> helpful when there is a bad package that won't compile for a week or
>> two, it makes it easier to just ignore it.
>
> Hopefully no one will mind a slight OT question, but still related...
>
> Is the testing version of portage 2.2 stable enough for production
> machines? There are a number of new features I'd like to take advantage
> of, but have always hesitated to and any non-stable system packages.

I've been using portage unmasked for a very long time and don't
remember having any portage-related problems. I'm sure there must be
some (or else why is it still RC?) but for me the new features are
worth the potential risk of using less-tested code.



Re: [gentoo-user] Switching to unstable

2010-04-12 Thread Tanstaafl
On 2010-04-12 11:05 AM, Paul Hartman wrote:
> I use the --keep-going always, it was a great addition and especially
> helpful when there is a bad package that won't compile for a week or
> two, it makes it easier to just ignore it.

Hopefully no one will mind a slight OT question, but still related...

Is the testing version of portage 2.2 stable enough for production
machines? There are a number of new features I'd like to take advantage
of, but have always hesitated to and any non-stable system packages.

Portage 2.2 is just taking forever to go stable... currently its on the
67th release candidate? That must be some kind of record (although
earlier versions of dovecot came close)... :P

-- 

Charles



Re: [gentoo-user] Switching to unstable

2010-04-12 Thread Crístian Viana
>
> when there is a bad package that won't compile for a week or two


I've already seen packages doing that, but they shouldn't happen, right? :-)


Re: [gentoo-user] Switching to unstable

2010-04-12 Thread Paul Hartman
On Mon, Apr 12, 2010 at 2:06 AM, Marc Joliet  wrote:
> Am Sun, 11 Apr 2010 17:13:26 -0700
> schrieb Mark Knecht :
>> A couple of packages in this OpenRC upgrade aren't building. I hope
>> they are less important. So far groff and help2man have failed so I
>> did --resume --skip-first and moved on for now.
>
> Just a note: you can also specify "--keep-going" and portage will do that for
> you and even recalculate the dependencies before continuing. After it's done,
> it gives you a message listing the packages that failed.

I use the --keep-going always, it was a great addition and especially
helpful when there is a bad package that won't compile for a week or
two, it makes it easier to just ignore it.

Also on big upgrade emerges like that sometimes packages will fail
during "the big emerge" but after you finish, etc-update, env-update
whatever, they will build okay, for whatever reason. Maybe related to
gcc/binutils/other libs stuff not matching for a short while...



Re: [gentoo-user] Switching to unstable

2010-04-12 Thread Marc Joliet
Am Sun, 11 Apr 2010 17:13:26 -0700
schrieb Mark Knecht :

> On Sun, Apr 11, 2010 at 5:00 PM, Neil Bothwick  wrote:
> > On Sun, 11 Apr 2010 16:11:53 -0700, Mark Knecht wrote:
> >
> >> > Doing system first makes good sense. Then you can update your config
> >> > files, follow the openrc update etc and then reboot. The world part of
> >> > the update will take quite a while, especially if you use KDE or
> >> > GNOME.
> >>
> >>  Less than most PC. It's an i7-980x with 24GB of RAM with RAID1.
> >

Gah, I'm getting PC envy ;-) !

> A couple of packages in this OpenRC upgrade aren't building. I hope
> they are less important. So far groff and help2man have failed so I
> did --resume --skip-first and moved on for now.

Just a note: you can also specify "--keep-going" and portage will do that for
you and even recalculate the dependencies before continuing. After it's done,
it gives you a message listing the packages that failed.

> Thanks,
> Mark

HTH
-- 
Marc Joliet
--
Lt. Frank Drebin: "It's true what they say: cops and women don't mix. Like
eating a spoonful of Drāno; sure, it'll clean you out, but it'll leave you
hollow inside."


signature.asc
Description: PGP signature


Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Mark Knecht
On Sun, Apr 11, 2010 at 5:33 PM, Mark Knecht  wrote:

> So help2man won't build due to some missing perl module.
>
> I'm assuming this isn't bad enough to stop a reboot from being
> successful but @system is @system so no reboot until I hear something
> back. (Or I get bored waiting...) ;-)
>
> Cheers,
> Mark
>
>
> checking for x86_64-pc-linux-gnu-gcc option to accept ISO C89... none
> needed
> checking for library containing dlsym... -ldl
> checking for library containing bindtextdomain... none required
> configure: error: perl module Locale::gettext required
>
> !!! Please attach the following file when seeking support:
> !!! /var/tmp/portage/sys-apps/help2man-1.37.1/work/help2man-1.37.1/config.log
>  * ERROR: sys-apps/help2man-1.37.1 failed:
>  *   econf failed
>  *
>  * Call stack:
>  *     ebuild.sh, line   48:  Called src_compile
>  *   environment, line 2332:  Called econf '--enable-nls'
>  *     ebuild.sh, line  538:  Called die
>  * The specific snippet of code:
>  *                      die "econf failed"
>  *
>  * If you need support, post the output of 'emerge --info
> =sys-apps/help2man-1.37.1',
>  * the complete build log and the output of 'emerge -pqv
> =sys-apps/help2man-1.37.1'.
>  * The complete build log is located at
> '/var/tmp/portage/sys-apps/help2man-1.37.1/temp/build.log'.
>  * The ebuild environment file is located at
> '/var/tmp/portage/sys-apps/help2man-1.37.1/temp/environment'.
>  * S: '/var/tmp/portage/sys-apps/help2man-1.37.1/work/help2man-1.37.1'
>
 Failed to emerge sys-apps/help2man-1.37.1, Log file:
>
  '/var/tmp/portage/sys-apps/help2man-1.37.1/temp/build.log'
>

emerge -e @system cleared that problem up and the reboot appears to
have been clean.

I don't know exactly what to make of the output from df. Clearly this
is a new way of looking at things:

cruncher ~ # df
Filesystem   1K-blocks  Used Available Use% Mounted on
rootfs51612984   8799528  40191652  18% /
/dev/root 51612984   8799528  40191652  18% /
rc-svcdir 102468   956   7% /lib64/rc/init.d
udev 10240   256  9984   3% /dev
shm6152104 0   6152104   0% /dev/shm
cruncher ~ #

rootfs? /dev/root? Did I miss something in my editing? I mount
/dev/md3 in fstab. It becomes something else in this environment I
guess...

Anyway, it rebooted cleanly and for the most part the instructions
were pretty good. XFCE4 is rebuilding, I'll check that X works and
then on to Gnome and KDE.

Cheers,
Mark



Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Mark Knecht
On Sun, Apr 11, 2010 at 5:22 PM, Mark Knecht  wrote:
> On Sun, Apr 11, 2010 at 5:13 PM, Mark Knecht  wrote:
> 
>>
>> A couple of packages in this OpenRC upgrade aren't building. I hope
>> they are less important. So far groff and help2man have failed so I
>> did --resume --skip-first and moved on for now.
>>
> 
>
> So it's done and I'm editing. In /etc/init.d I see a blinking link:
>
> runscript.sh -> ../../sbin/runscript.sh
>
> Remove it or is there a problem?
>
> - Mark
>

So help2man won't build due to some missing perl module.

I'm assuming this isn't bad enough to stop a reboot from being
successful but @system is @system so no reboot until I hear something
back. (Or I get bored waiting...) ;-)

Cheers,
Mark


checking for x86_64-pc-linux-gnu-gcc option to accept ISO C89... none
needed
checking for library containing dlsym... -ldl
checking for library containing bindtextdomain... none required
configure: error: perl module Locale::gettext required

!!! Please attach the following file when seeking support:
!!! /var/tmp/portage/sys-apps/help2man-1.37.1/work/help2man-1.37.1/config.log
 * ERROR: sys-apps/help2man-1.37.1 failed:
 *   econf failed
 *
 * Call stack:
 * ebuild.sh, line   48:  Called src_compile
 *   environment, line 2332:  Called econf '--enable-nls'
 * ebuild.sh, line  538:  Called die
 * The specific snippet of code:
 *  die "econf failed"
 *
 * If you need support, post the output of 'emerge --info
=sys-apps/help2man-1.37.1',
 * the complete build log and the output of 'emerge -pqv
=sys-apps/help2man-1.37.1'.
 * The complete build log is located at
'/var/tmp/portage/sys-apps/help2man-1.37.1/temp/build.log'.
 * The ebuild environment file is located at
'/var/tmp/portage/sys-apps/help2man-1.37.1/temp/environment'.
 * S: '/var/tmp/portage/sys-apps/help2man-1.37.1/work/help2man-1.37.1'

>>> Failed to emerge sys-apps/help2man-1.37.1, Log file:

>>>  '/var/tmp/portage/sys-apps/help2man-1.37.1/temp/build.log'



Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Mark Knecht
On Sun, Apr 11, 2010 at 5:13 PM, Mark Knecht  wrote:

>
> A couple of packages in this OpenRC upgrade aren't building. I hope
> they are less important. So far groff and help2man have failed so I
> did --resume --skip-first and moved on for now.
>


So it's done and I'm editing. In /etc/init.d I see a blinking link:

runscript.sh -> ../../sbin/runscript.sh

Remove it or is there a problem?

- Mark



Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Mark Knecht
On Sun, Apr 11, 2010 at 5:00 PM, Neil Bothwick  wrote:
> On Sun, 11 Apr 2010 16:11:53 -0700, Mark Knecht wrote:
>
>> > Doing system first makes good sense. Then you can update your config
>> > files, follow the openrc update etc and then reboot. The world part of
>> > the update will take quite a while, especially if you use KDE or
>> > GNOME.
>>
>>  Less than most PC. It's an i7-980x with 24GB of RAM with RAID1.
>
> Oh, is that all, I thought you had something quite fast?
>
> :P
>

Fast for under 100 lbs of iron. I'm sure the big boys must have things
much faster. Full KDE in 109 minutes seemed pretty good to me, and
it's fun to watch 12 processors max out at the same time. 3 drive
RAID1 for Gentoo and backups, 2 drive RAID0 for where the VMs will
run. 1 extra drive in the box in case a drive goes bad one of these
days.

A couple of packages in this OpenRC upgrade aren't building. I hope
they are less important. So far groff and help2man have failed so I
did --resume --skip-first and moved on for now.

If I get to the end, do the hand edits and manage to reboot then I'll
do an emerge -e @system just to make sure things are really OK before
dealing with @world.

Thanks,
Mark



Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Neil Bothwick
On Sun, 11 Apr 2010 16:11:53 -0700, Mark Knecht wrote:

> > Doing system first makes good sense. Then you can update your config
> > files, follow the openrc update etc and then reboot. The world part of
> > the update will take quite a while, especially if you use KDE or
> > GNOME. 
> 
>  Less than most PC. It's an i7-980x with 24GB of RAM with RAID1.

Oh, is that all, I thought you had something quite fast?

:P


-- 
Neil Bothwick

God is real, unless specifically declared integer.


signature.asc
Description: PGP signature


Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Mark Knecht
On Sun, Apr 11, 2010 at 3:55 PM, Neil Bothwick  wrote:
> On Sun, 11 Apr 2010 14:59:07 -0700, Mark Knecht wrote:
>
>> 1) I don't see any mention of hald & dbus in the upgrade guide. I
>> currently have them turned on. Are they still necessary? I know hald
>> is going away one of these days. Is it too early for me to dump it.
>> Possibly dump hald before the upgrade, make sure everything is
>> consistent, and then do the ~amd64 work?
>
> Don't try to second guess what the system needs, just upgrade and do a
> --depclean. #~Portage will let you know if anything is no longer needed.
> You'll probably find depclean wants to remove a few things, but hal won't
> be one of them.

OK.

>
>> 2) Any problem if I do emerge -DuN @system and then make the changes
>> in the upgrade guide followed by @world? I've turned of xdm. I don't
>> have a UPS on this system yet so @system would shrink the window of a
>> power outage causing me problems. (It's pouring rain here today. It
>> will likely happen no matter what) ;-)
>
> Doing system first makes good sense. Then you can update your config
> files, follow the openrc update etc and then reboot. The world part of
> the update will take quite a while, especially if you use KDE or GNOME.
>

 Less than most PC. It's an i7-980x with 24GB of RAM with RAID1.
My first really nice machine in years.

Thanks!

Cheers,
Mark
>
> --
> Neil Bothwick
>
> Everywhere is walking distance if you have the time.
>



Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Neil Bothwick
On Sun, 11 Apr 2010 14:59:07 -0700, Mark Knecht wrote:

> 1) I don't see any mention of hald & dbus in the upgrade guide. I
> currently have them turned on. Are they still necessary? I know hald
> is going away one of these days. Is it too early for me to dump it.
> Possibly dump hald before the upgrade, make sure everything is
> consistent, and then do the ~amd64 work?

Don't try to second guess what the system needs, just upgrade and do a
--depclean. #~Portage will let you know if anything is no longer needed.
You'll probably find depclean wants to remove a few things, but hal won't
be one of them.

> 2) Any problem if I do emerge -DuN @system and then make the changes
> in the upgrade guide followed by @world? I've turned of xdm. I don't
> have a UPS on this system yet so @system would shrink the window of a
> power outage causing me problems. (It's pouring rain here today. It
> will likely happen no matter what) ;-)

Doing system first makes good sense. Then you can update your config
files, follow the openrc update etc and then reboot. The world part of
the update will take quite a while, especially if you use KDE or GNOME.


-- 
Neil Bothwick

Everywhere is walking distance if you have the time.


signature.asc
Description: PGP signature


Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Neil Bothwick
On Sun, 11 Apr 2010 14:24:18 -0700, Mark Knecht wrote:

> I have a new machine that just came up yesterday. I was thinking of
> running ~arch on it and seeing how things work out. Seems like it's a
> good time to do it if I'm ever going to as I haven't started using it
> and it's going to get busy. If your answers are dependent on the work
> done then this machine is (hopefully) going to run a bunch of vmware
> instances at the same time. Any problem doing that under ~amd64? I've
> never run more than 1 in the past. This time hopefully 5 at a time.
>
I have run several VMs at once in workstation.

> Question - after adding ~amd64 to make.conf do I then ever need
> anything in package.keywords again because the whole system is ~amd64?

Only if you want to use packages that do not have amd64 or ~amd64
keywords.

> It seems that there would be some use flag changes required:

That's not strictly down to the use of ~amd64, it's a change that will
affect everyone in due course, you just get to find out these things that
bit sooner now.


-- 
Neil Bothwick

On the other hand, you have different fingers.


signature.asc
Description: PGP signature


Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Mark Knecht
On Sun, Apr 11, 2010 at 2:24 PM, Mark Knecht  wrote:
> On Sun, Apr 11, 2010 at 12:43 PM, Neil Bothwick  wrote:
>> On Sun, 11 Apr 2010 20:36:37 +0200, Damian wrote:
>>
>>> The reason for doing so is that what is considered as unstable as been
>>> regarded as stable releases for the developers, and the truth is that
>>> the problems I got for using outdated software were more that the ones I
>>> had for using unstable versions.
>>
>> That's because what you describe as unstable has nothing to do with
>> stability of the software. ~arch is testing ebuilds, they are unstable in
>> that they change more often, but as you say, the upstream software is
>> considered fit for use.
>>
>>> Thus, I'm thinking about switching all of my system to the unstable
>>> branch. But first I want to be sure that this is reasonable given the
>>> problems I described before.
>>>
>>> Can you provide me some useful advice according to your experience?
>>
>> Switching to testing is as easy as changing ACCEPT_KEYWORDS in make.conf
>> and doing emerge -uaDN @world. Switching back is less easy, but if this
>> is what you want to do, then go for it. I have run testing for years,
>> with far less problems than some people running mixed arch systems.
>>
>>
>> --
>> Neil Bothwick
>
> I have a new machine that just came up yesterday. I was thinking of
> running ~arch on it and seeing how things work out. Seems like it's a
> good time to do it if I'm ever going to as I haven't started using it
> and it's going to get busy. If your answers are dependent on the work
> done then this machine is (hopefully) going to run a bunch of vmware
> instances at the same time. Any problem doing that under ~amd64? I've
> never run more than 1 in the past. This time hopefully 5 at a time.
>
> Question - after adding ~amd64 to make.conf do I then ever need
> anything in package.keywords again because the whole system is ~amd64?
>
> It seems that there would be some use flag changes required:
>
> cruncher ~ # emerge -pvDuN @world
>
> These are the packages that would be merged, in order:
>
> Calculating dependencies... done!
>
> emerge: there are no ebuilds built with USE flags to satisfy
> ">=sys-fs/udev-145[extras]".
> !!! One of the following packages is required to complete your request:
> - sys-fs/udev-151-r1 (Change USE: +extras)
> (dependency required by "gnome-base/gvfs-1.4.3" [ebuild])
> (dependency required by "gnome-base/libgnome-2.28.0" [ebuild])
> (dependency required by "gnome-base/libbonoboui-2.24.3" [ebuild])
> (dependency required by "gnome-base/libgnomeui-2.24.3" [ebuild])
> (dependency required by "net-libs/xulrunner-1.9.2.3-r1" [ebuild])
> (dependency required by "www-client/mozilla-firefox-3.6.3" [ebuild])
> (dependency required by "@world" [argument])
>
> cruncher ~ #
>
> Will there be a lot of this before I execute the emerge? I know udev
> is pretty special.
>
> - Mark
>

OK - I walked through the list until it told me things were OK. These
are the use flag additions I had to add:

sys-fs/udev extras
gnome-base/gvfs gdu
sys-apps/parted device-mapper
sys-auth/consolekit policykit
net-libs/opal sip
net-libs/ptlib wav
x11-base/xorg-server kdrive

A couple more questions:

1) I don't see any mention of hald & dbus in the upgrade guide. I
currently have them turned on. Are they still necessary? I know hald
is going away one of these days. Is it too early for me to dump it.
Possibly dump hald before the upgrade, make sure everything is
consistent, and then do the ~amd64 work?

2) Any problem if I do emerge -DuN @system and then make the changes
in the upgrade guide followed by @world? I've turned of xdm. I don't
have a UPS on this system yet so @system would shrink the window of a
power outage causing me problems. (It's pouring rain here today. It
will likely happen no matter what) ;-)

Thanks,
Mark



Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Mark Knecht
On Sun, Apr 11, 2010 at 12:43 PM, Neil Bothwick  wrote:
> On Sun, 11 Apr 2010 20:36:37 +0200, Damian wrote:
>
>> The reason for doing so is that what is considered as unstable as been
>> regarded as stable releases for the developers, and the truth is that
>> the problems I got for using outdated software were more that the ones I
>> had for using unstable versions.
>
> That's because what you describe as unstable has nothing to do with
> stability of the software. ~arch is testing ebuilds, they are unstable in
> that they change more often, but as you say, the upstream software is
> considered fit for use.
>
>> Thus, I'm thinking about switching all of my system to the unstable
>> branch. But first I want to be sure that this is reasonable given the
>> problems I described before.
>>
>> Can you provide me some useful advice according to your experience?
>
> Switching to testing is as easy as changing ACCEPT_KEYWORDS in make.conf
> and doing emerge -uaDN @world. Switching back is less easy, but if this
> is what you want to do, then go for it. I have run testing for years,
> with far less problems than some people running mixed arch systems.
>
>
> --
> Neil Bothwick

I have a new machine that just came up yesterday. I was thinking of
running ~arch on it and seeing how things work out. Seems like it's a
good time to do it if I'm ever going to as I haven't started using it
and it's going to get busy. If your answers are dependent on the work
done then this machine is (hopefully) going to run a bunch of vmware
instances at the same time. Any problem doing that under ~amd64? I've
never run more than 1 in the past. This time hopefully 5 at a time.

Question - after adding ~amd64 to make.conf do I then ever need
anything in package.keywords again because the whole system is ~amd64?

It seems that there would be some use flag changes required:

cruncher ~ # emerge -pvDuN @world

These are the packages that would be merged, in order:

Calculating dependencies... done!

emerge: there are no ebuilds built with USE flags to satisfy
">=sys-fs/udev-145[extras]".
!!! One of the following packages is required to complete your request:
- sys-fs/udev-151-r1 (Change USE: +extras)
(dependency required by "gnome-base/gvfs-1.4.3" [ebuild])
(dependency required by "gnome-base/libgnome-2.28.0" [ebuild])
(dependency required by "gnome-base/libbonoboui-2.24.3" [ebuild])
(dependency required by "gnome-base/libgnomeui-2.24.3" [ebuild])
(dependency required by "net-libs/xulrunner-1.9.2.3-r1" [ebuild])
(dependency required by "www-client/mozilla-firefox-3.6.3" [ebuild])
(dependency required by "@world" [argument])

cruncher ~ #

Will there be a lot of this before I execute the emerge? I know udev
is pretty special.

- Mark



Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Alex Schuster
Damian writes:

> Thus, I'm thinking about switching all of my system to the unstable
> branch. But first I want to be sure that this is reasonable given the
> problems I described before.
> 
> Can you provide me some useful advice according to your experience?

I have asked a similar question here, I suggest you read the thread [*]. I 
did not regret the switch and would also suggest running ~arch. Beware the 
update to openrc, and do what the elog messages tell you to do before 
rebooting.

Wonko

[*] http://www.mail-archive.com/gentoo-user@lists.gentoo.org/msg94206.html



Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Alan McKinnon
On Sunday 11 April 2010 20:36:37 Damian wrote:
> Hello,
> 
> I've been using the gentoo stable branch since I began with this distro
> (around 4 years ago), but lately I've been unmasking almost all packages
> I use in my daily work (emacs, firefox, gnome*, xmonad, etc).
> 
> The reason for doing so is that what is considered as unstable as been
> regarded as stable releases for the developers, and the truth is that
> the problems I got for using outdated software were more that the ones I
> had for using unstable versions.
> 
> Thus, I'm thinking about switching all of my system to the unstable
> branch. But first I want to be sure that this is reasonable given the
> problems I described before.
> 
> Can you provide me some useful advice according to your experience?
> 
> Thanks in advance,
> Damian.

I run ~arch everywhere, and I seldom have trouble. The hiccups that do occur 
are usually blockers between the latest version of something and the previous 
version, very easy to fix if you have read the portage man pages. But these 
days portage knows how to deal with that, so it doesn't expect you to hunt, 
search, unmerge, merge. It just tells you what it will do and waits for you to 
tell it to get on with it. Nice :-)

Reading this list for many years leads me to conclude:

- running ~arch is OK if you can drive portage. Few breakages, on the whole 
they are easily fixed
- running arch. Tends to run a bit behind latest stuff out there, but stable.
- mixing the two. No, this is not something you want to do. You are doing it 
now, so just update make.conf to use ~arch and be done with it.


-- 
alan dot mckinnon at gmail dot com



Re: [gentoo-user] Switching to unstable

2010-04-11 Thread Neil Bothwick
On Sun, 11 Apr 2010 20:36:37 +0200, Damian wrote:

> The reason for doing so is that what is considered as unstable as been
> regarded as stable releases for the developers, and the truth is that
> the problems I got for using outdated software were more that the ones I
> had for using unstable versions.

That's because what you describe as unstable has nothing to do with
stability of the software. ~arch is testing ebuilds, they are unstable in
that they change more often, but as you say, the upstream software is
considered fit for use.

> Thus, I'm thinking about switching all of my system to the unstable
> branch. But first I want to be sure that this is reasonable given the
> problems I described before.
> 
> Can you provide me some useful advice according to your experience?

Switching to testing is as easy as changing ACCEPT_KEYWORDS in make.conf
and doing emerge -uaDN @world. Switching back is less easy, but if this
is what you want to do, then go for it. I have run testing for years,
with far less problems than some people running mixed arch systems.


-- 
Neil Bothwick

CPU: (n.) acronym for Central Purging Unit. A device which discards or
distorts data sent to it, sometimes returning more data and sometimes
merely over-heating.


signature.asc
Description: PGP signature


[gentoo-user] Switching to unstable

2010-04-11 Thread Damian
Hello,

I've been using the gentoo stable branch since I began with this distro
(around 4 years ago), but lately I've been unmasking almost all packages
I use in my daily work (emacs, firefox, gnome*, xmonad, etc). 

The reason for doing so is that what is considered as unstable as been
regarded as stable releases for the developers, and the truth is that
the problems I got for using outdated software were more that the ones I
had for using unstable versions.

Thus, I'm thinking about switching all of my system to the unstable
branch. But first I want to be sure that this is reasonable given the
problems I described before.

Can you provide me some useful advice according to your experience?

Thanks in advance,
Damian.