Re: [gentoo-user] Reading a DVD as an ISO image

[Daniel Pielmeier]

2009/11/17 Sebastian Beßler :
>> What about mounting the iso and create it again from the mountpoint.
>
> That he has deleted the iso?

Well, he then can create the iso again from the mounted DVD, without
readcd or dd, but with the same program (mkisofs?) he used before.

-- 
Daniel Pielmeier

equery depends vs. emerge -pv --depclean (was: Re: [gentoo-user] MySql versus sqlite. Don't want sqlite anymore.)

[Daniel Pielmeier]

Dale schrieb am 17.11.2009 22:53:
>
> r...@smoker / # equery depends sqlite
> [ Searching for packages depending on sqlite... ]
> app-pda/libopensync-0.22 (>=dev-db/sqlite-3)

If you have this package installed and want to keep it you are forced to
keep sqlite as it is a hard dependency. Meaning this package needs
sqlite and there is no alternative.

> app-portage/eix-0.17.0 (sqlite? >=dev-db/sqlite-3)

Here sqlite is optional via use flag. So you can get rid of the
dependency it by disabling the use flag.

> Those appear to be in the ebuild and MySql is not a option for
> replacement.  Is there a way?  I have already emerged these with the
> -sqlite and +mysql USE flag.  I notice something about sqlite3 in one of
> the ebuilds.  What is that?

The main reason why I respond to this post is that there are still a lot
of people which are using "equery depends", although it is known to
print false positive by not taking use flags into account. You have to
scan through the output and check if the use flag is activated which
pulls in the package you want to check. Like for eix in the above
example. Eix only depends on sqlite if the sqlite use flag is activated.
"Equery depends" lists it in any case (activated or not) as it doesn't
take the use flags into account.

A more convenient way to find out the correct reverse dependencies with
portage is "emerge -pv --depclean "[1]. Running only "emerge -pv
--depclean" checks the complete world and system set, but if you pass a
package to it only the reverse dependencies of the package in question
are examined.

[1] http://www.gentoo.org/proj/en/portage/doc/faq.xml

-- 
Daniel Pielmeier



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] mplayer plugin

[Daniel Pielmeier]

Mick schrieb am 22.11.2009 14:48:
> http://www.amd.com/us-
> en/assets/content_type/DownloadableAssets/T64X2_animation.wmv
> 

Works fine with gecko-mediaplayer and download-helper.

-- 
Daniel Pielmeier



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] mplayer plugin

[Daniel Pielmeier]

Mick schrieb am 22.11.2009 20:02:
> On Sunday 22 November 2009 17:27:26 Daniel Pielmeier wrote:
>> Mick schrieb am 22.11.2009 14:48:
>>> http://www.amd.com/us-
>>> en/assets/content_type/DownloadableAssets/T64X2_animation.wmv
>> Works fine with gecko-mediaplayer and download-helper.
> 
> Yes, thank you.  I've installed it now and it looks neat.
> 
> Any luck with Apple Trailers?  It doesn't ask me where to save the download.

I am not able to play the Apple Trailers by clicking at the link to the
movie (maybe a problem with the java script popup). When I open the link
in a new window I can watch/download with gecko-mediaplayer like normal.
Download-helper seems to have problems though.

-- 
Daniel Pielmeier



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Bittorent black box

[Daniel Troeder]

On Tue, 2009-11-24 at 00:02 +0100, laurent wrote:
> Hi,
> 
> I would like to use my server as bittorent client and maybe tracker later.
> I would like to have a web interface but I could also develop it myself 
> later.
> 
> I saw, qbittorent and hrktorrent that looks good. I don't know any 
> features of the last one.
> What would you recommend to use?
> 
> thanks
> Laurent
I'd recommend net-p2p/deluge (http://deluge-torrent.org/).
It has all the features I ever wanted and its written in Python using
libtorrent as engine.

It runs as a non-ui daemon to which you can connect clients via network.
Existing clients are: console, web and gtk+.

I start the daemon and a web client on my server, and connect ether via
web or with the gtk-ui :)

It has no build-in tracker, but I recommend using a public one anyway
(for example http://openbittorrent.com/).

Bye,
Daniel

-- 
PGP key @ http://pgpkeys.pca.dfn.de/pks/lookup?search=0xBB9D4887&op=get
# gpg --recv-keys --keyserver hkp://subkeys.pgp.net 0xBB9D4887



signature.asc
Description: This is a digitally signed message part

Re: [gentoo-user] question about wireless setting

[Daniel Pielmeier]

2009/12/1 Willie Wong :
> On Tue, Dec 01, 2009 at 03:46:51PM +0900, Penguin Lover ?? squawked:
>> I could see wireless driver when typed lspci.
>
> No you can't. You see the pci devices. You don't see drivers. You need
> a drive either emerged or compiled into your kernel before the device
> will function.
>

Afaik lspci -v shows the driver in use.

-- 
Daniel Pielmeier

Re: [gentoo-user] What magic does portage use?

[Daniel Troeder]

On Fri, 2009-12-11 at 07:16 -0600, Dale wrote:
> Helmut Jarausch wrote:
> > On 11 Dec, Alan McKinnon wrote:
> >   
> >> On Friday 11 December 2009 11:11:41 Helmut Jarausch wrote:
> >> 
> >>> Hi,
> >>>
> >>> I'm curious how portage solves its most difficult part (in my eyes).
> >>>
> >>> When installing a dynamic library (by "hand") I have often got an
> >>> error messages if the corresponding library is currently in use.
> >>>
> >>> How does portage succeed anyway.
> >>> (I have the suspicion that it does not succeed always, since sometimes
> >>> only rebooting solves some very strange problems)
> >>>
> >>> How to replace fundamental X11-libaries on a system running X11
> >>> or even more suprising, how can I replace a running glibc ?
> >>>
> >>> Many thanks for enlightening me,
> >>> Helmut.
> >>>
> >>>   
> >> Portage does nothing special, as dealing with this is a Unix thing.
> >>
> >> On Unix, the inode is the file, not the directory entry. If you want to 
> >> replace an open file, the system simply does it and updates the dentry to 
> >> point to a new inode. Any spp using the old file will continue to use it 
> >> as it 
> >> still has a handle to the inode. The inode is only fully deleted when the 
> >> last 
> >> app using it closes it
> >>
> >> If you update a library to a new version with an API break, the lib should 
> >> get 
> >> a new soname so the file is a different name, hence no collision (symlinks 
> >> to 
> >> libraries excepted).
> >>
> >> This is how it should work, any code that tries to do it a different way 
> >> is by 
> >> definition broken, that's why portage needs take no special measures.
> >>
> >> All of this is in complete contrast to other broken systems, such as 
> >> Windows 
> >> for example. On Windows, the filename IS the file, so upgrades are 
> >> horrible. 
> >> Installers must put the file somewhere else and have the final steps and 
> >> registry updates done at next reboot before anything has a chance to open 
> >> libs. This is why fairly deep updates on Windows often require multiple 
> >> reboot 
> >> - multiple apps installed multiple libs to be fiddled with multiple times 
> >> 
> >>
> >> 
> >
> > Many thanks Alan,
> >
> > so I conclude that rebooting IS necessary to get the new libraries used,
> > isn't it?
> > On the other hand running applications should continue to run, which is
> > not always the case, e.g. recently using cvs as non-root user just
> > hanged. Rebooting the system solved it (since I update my system nearly
> > each day).
> >
> > Thanks again Alan,
> > Helmut.
> >   
> 
> I'll add two cents here.  Let's say I upgrade Seamonkey which is my web 
> browser / email program.  I sync and notice there is a update to 
> Seamonkey available and I let emerge update it.  When the install is 
> complete, I don't have to reboot or even log out of KDE.  All I have to 
> do is close Seamonkey and start it again.  It will then load the new 
> updated version and run it.
> 
> The same could be said for a service like cups.  If you update cups, all 
> you have to do is restart the service.  It will stop the old service 
> then load up the new service that was just installed.  Just a simple 
> "/etc/init.d/cupsd restart" will work just fine. 
> 
> If you upgrade something kde, say kdelibs or some other kde base 
> package, then all you need to do is log out of KDE and log back in 
> again.  Sort of the same with updating xorg, logout, go to a console and 
> restart xdm or whatever you use to start X.  I usually use the ctrl alt 
> backspace key but restarting the service is better, or so some have said 
> anyway. 
> 
> Rebooting will also do all of this but it is not needed.  From a 
> technical stand point, the only time you must reboot is to load a new 
> kernel. 
> 
> Hope that helps a little.
> 
> Dale
> 
> :-)  :-) 

I absolutely concur with Alan and Dale, I just want to warn a bit about
complex X11-environments like gnome or kde. If you logout and login
again, it is NOT secure, that all gnome/kde/qt apps have been closed.
There are services (gconf, kded, pulse, etc) that take a time to quit,
or sometimes just don't :(

Then, when an app was linked against a symlink, and that left-over-app
too, the dynamic loader may not load a newly installed library, but
reuses the one in memory (from the left-over-app). [1]

It's still valid, that no reboot is needed, but you sure can be
unlucky :)

Bye,
Daniel


[1] don't take this info for granted, I'm no expert in this - just what
I understood from reading...


-- 
PGP key @ http://pgpkeys.pca.dfn.de/pks/lookup?search=0xBB9D4887&op=get
# gpg --recv-keys --keyserver hkp://subkeys.pgp.net 0xBB9D4887



signature.asc
Description: This is a digitally signed message part

Re: [gentoo-user] What magic does portage use?

[Daniel Troeder]

On 12/11/2009 08:00 PM, Dale wrote:
> Alan McKinnon wrote:
>> On Friday 11 December 2009 17:07:17 Dale wrote:
>>  
>>> Alan McKinnon wrote:
>>>
>>>> On Friday 11 December 2009 15:16:01 Dale wrote:
>>>>  
>>>>> Rebooting will also do all of this but it is not needed.  From a
>>>>> technical stand point, the only time you must reboot is to load a new
>>>>> kernel.
>>>>> 
>>>> And these days, not even then :-)
>>>>
>>>> [it requires some voodoo but is certainly possible]
>>>>
>>>> [[and I don't mean build and install a new kernel, I really do mean loa
>>>> ti into memory and run it, dispensing with the old one]]
>>>>   
>>> I have read about that but never read something from someone who has
>>> actually done it.  I have always been curious as to how that would work,
>>> in reality not just theory.
>>> 
>>
>> kexec and CONFIG_RELOCATABLE
>>
>>  
>>> I have also wondered why a person would go to all that trouble.
>>> Wouldn't all the services have to be restarted anyway?
>>> 
>>
>> Nope. userspace ABI is stable so services just carry on as normal once
>> he new kernel comes up. You don't need to restart SeaMonkey if you
>> restart a local apache on your machine - same thing
>>
>>   
> 
> That would be cool of you had a system that just couldn't be rebooted. 
> Is there such a thing tho?  What would be the reason a machine just
> could not be rebooted?  I guess one would be if the puter was on planet
> Mars maybe?  Is that how NASA does it?  lol  Could you imagine getting a
> blue screen of death on a computer that is on Mars?  O_O
> 
> Dale
> 
> :-)  :-)
A real world scenario would be a bank server doing transactions. Those
big irons do never ever get shut down.
(But they also don't ever get really updated ;)

Did you know, that they still use cobol-code from decades ago. The code
has to interact with newer systems, but the existing code is not allowed
to be altered, they just run it inside hugh java application servers on
their main frames :D

Bye,
Daniel

-- 
use PGP key @ http://pgpkeys.pca.dfn.de/pks/lookup?search=0xBB9D4887&op=get
# gpg --recv-keys --keyserver hkp://subkeys.pgp.net 0xBB9D4887



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Blocks with xz-utils lzma-utils

[Daniel Pielmeier]

Dale schrieb am 17.12.2009 16:55:
> 
> [blocks b ] kde-base/kde-meta:4.3[-kdeprefix]
> ("kde-base/kde-meta:4.3[-kdeprefix]" is blocking kde-base/kde-meta-)
> [blocks B ] app-arch/lzma-utils ("app-arch/lzma-utils" is blocking
> app-arch/xz-utils-)
> [blocks B ] app-arch/xz-utils ("app-arch/xz-utils" is blocking
> app-arch/lzma-utils-4.32.7)
> 
> Total: 278 packages (17 upgrades, 19 new, 242 in new slots, 248
> uninstalls), Size of downloads: 240,422 kB
> Conflict: 511 blocks (2 unsatisfied)
> Portage tree and overlays:
> [0] /usr/portage
> [1] /usr/local/portage/layman/kde
> 
> * Error: The above package list contains packages which cannot be
> * installed at the same time on the same system.
> 
>  ('ebuild', '/', 'app-arch/xz-utils-', 'merge') pulled in by
>app-arch/xz-utils required by ('installed', '/',
> 'app-arch/rpm2targz-9.0.0.3g', 'nomerge')
>app-arch/xz-utils required by ('installed', '/',
> 'sys-libs/gpm-1.20.5', 'nomerge')
>app-arch/xz-utils required by ('installed', '/',
> 'dev-libs/mpfr-2.4.1_p1', 'nomerge')
>(and 10 more)
> 
>  ('ebuild', '/', 'app-arch/lzma-utils-4.32.7', 'merge') pulled in by
>app-arch/lzma-utils required by ('installed', '/',
> 'app-portage/eix-0.17.0', 'nomerge')
>app-arch/lzma-utils required by ('installed', '/',
> 'app-arch/libarchive-2.7.0-r1', 'nomerge')
> 

You need at least eix-0.17.1 and libarchive-2.7.1. Older versions depend
on lzma-utils instead of xz-utils, thus the blocker.

-- 
Daniel Pielmeier



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools

[Daniel Pielmeier]

KH schrieb:

Hi,

after reading some emails from the list I have been trying to unmerge 
cdrkit and to emerge cdrtools. I aslo hat to unmerge dvd+rw-tools and 
kino. Anyway I am still not able to emerge cdrtools.


!!! Cannot write to '/usr/include/scsilib/scg'.
!!! Please check permissions and directories for broken symlinks.
!!! You may start the merge process again by using ebuild:
!!! ebuild /usr/portage/app-cdr/cdrtools/cdrtools-2.01.01_alpha34.ebuild 
merge

!!! And finish by running this: env-update


than there is a very long list I can post if needed.

ls -lah /usr/include/scsilib/scg
lrwxrwxrwx 1 root root 25 Jun 25 12:48 /usr/include/scsilib/scg -> 
/usr/include/scsilib/usal


on my xterm there is red highlighting under the output.

What can I do to install cdrtools correctly to a working standard?


Will I be able to emerge kino again, without cdrkit?

Thanks for any help

kh


So you successfully unmerged cdrkit, dvd+rw-tools and kino. The
installation of cdrtools fails because of a broken symlink. I guess the
/usr/include/scsilib/scg symlink to usr/include/scsilib/usal has not
been removed with cdrkit and is now broken and the cdrtools installation
can not overwrite the link. If the output is really red it is indeed a
broken symlink and you can safely remove it, then try to install
cdrtools again or use the proposed commands.

It should be no problem to install kino again as cdrkit provides the
same functionality like cdrtools, besides the different amount of
development that happens on both projects. But i think there was no need
to remove kino at all when switching to cdrtools, because it has no
dependency on both apps.

I find it a bit sad that cdrkit is the preferred cd burning application
because it satisfies the cdrtools virtual first. This means when you
have neither cdrkit nor cdrtools installed cdrkit is preferred. However
i could not find out why. There is nothing in the profiles which defines
how cdrkit is favored. Maybe someone can point me to the right location.

I think cdrtools should be preferred and not the fork cdrkit!

My make profile points to
/media/system/repositories/portage/profiles/default/linux/x86/2008.0.

Regards,

Daniel



--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools

[Daniel Pielmeier]

Volker Armin Hemmann schrieb:

On Samstag, 5. Juli 2008, Daniel Pielmeier wrote:


I find it a bit sad that cdrkit is the preferred cd burning application
because it satisfies the cdrtools virtual first. This means when you
have neither cdrkit nor cdrtools installed cdrkit is preferred. However
i could not find out why. 


k is before t in the alphabet?




Probably :-)

In the 2007.0 desktop profile it was the default application, but i can 
not find something similar for the 2008.0 profile


Here are the contents of 
/media/system/repositories/portage/profiles/default-linux/x86/2007.0/desktop/virtuals


# Copyright 1999-2007 Gentoo Foundation
# Distributed under the terms of the GNU General Public License, v2
# $Header: 
/var/cvsroot/gentoo-x86/profiles/default-linux/x86/2007.0/desktop/virtuals,v 
1.1 2007/04/13 22:02:31 wolf31o2 Exp $


virtual/cdrtoolsapp-cdr/cdrkit
virtual/fam app-admin/gamin


--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools

[Daniel Pielmeier]

Joerg Schilling schrieb:

Daniel Pielmeier <[EMAIL PROTECTED]> wrote:


So you successfully unmerged cdrkit, dvd+rw-tools and kino. The
installation of cdrtools fails because of a broken symlink. I guess the
/usr/include/scsilib/scg symlink to usr/include/scsilib/usal has not
been removed with cdrkit and is now broken and the cdrtools installation
can not overwrite the link. If the output is really red it is indeed a
broken symlink and you can safely remove it, then try to install
cdrtools again or use the proposed commands.


Mmmm, cdrtools may create a directory /usr/include/scg/ but the text /scsilib/
is nowhere in cdrtools.


In Gentoo scg is installed under /usr/include/scsilib

Here a snippet from the ebuild:

insinto /usr/include/scsilib/scg
doins include/scg/*.h

/usr/include/scsilib/scg -> /usr/include/scsilib/usal is a symlink used
by cdrkit. It was not removed when the package manager uninstalled
cdrkit. At least I think so, as this does not happen here but maybe with
older versions I don't use.

When installing cdrtools the package manager wants to create the
directory /usr/include/scsilib/scg which fails as the broken symlink is
still in place.


It should be no problem to install kino again as cdrkit provides the
same functionality like cdrtools, besides the different amount of


There is no active development in cdrkit, so cdrkit does not provide new 
features. If cdrkit did not change libscg in a way that prevents some usage, 
you could treat cdrkit like a 3 year old version of cdrtools.


I don't know how much work really happens with cdrkit as I don't follow
the progress there. It seems however you have done more work on cdrtools
than what happened on cdrkit besides finalizing star 1.5.

By the way I for myself say many thanks for cdrtools. I don't care much
about the license as long it is not closed source and I can use it
within my OS. I just care about the quality of the code.


development that happens on both projects. But i think there was no need
to remove kino at all when switching to cdrtools, because it has no
dependency on both apps.


This is correct.

I don't know kino but from looking at freshmeat, it seems that kino has no 
relation to cdrtools. Or does kino call "mkisofs"?


You are right, it depends on dvd+rw-tools maybe because of this the OP
has removed it from his system temporarily along with cdrkit and
dvd+rw-tools.

Regards,

Daniel


--
gentoo-user@lists.gentoo.org mailing list

SCG (was: Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools)

[Daniel Pielmeier]

Joerg Schilling schrieb:

Daniel Pielmeier <[EMAIL PROTECTED]> wrote:


Joerg Schilling schrieb:

Daniel Pielmeier <[EMAIL PROTECTED]> wrote:


So you successfully unmerged cdrkit, dvd+rw-tools and kino. The
installation of cdrtools fails because of a broken symlink. I guess the
/usr/include/scsilib/scg symlink to usr/include/scsilib/usal has not
been removed with cdrkit and is now broken and the cdrtools installation
can not overwrite the link. If the output is really red it is indeed a
broken symlink and you can safely remove it, then try to install
cdrtools again or use the proposed commands.

Mmmm, cdrtools may create a directory /usr/include/scg/ but the text /scsilib/
is nowhere in cdrtools.

In Gentoo scg is installed under /usr/include/scsilib

Here a snippet from the ebuild:

insinto /usr/include/scsilib/scg
doins include/scg/*.h

/usr/include/scsilib/scg -> /usr/include/scsilib/usal is a symlink used
by cdrkit. It was not removed when the package manager uninstalled
cdrkit. At least I think so, as this does not happen here but maybe with
older versions I don't use.


This is a really bad idea. 


/usr/include/scg/ is a planned directory that is known to be unique.

cdrkit does not deliver anything that is even approximately useful as a 
replacement for libscg.


Installing the includefiles from libscg into /usr/include/scsilib/scg makes them
unusable as there is no software that is aware of this location. It seems that 
the only result is that it makes it harder to install cdrtools instread of 
cdrkit.


Jörg



Out of curiosity I tried a manual install and /usr/include/scg/ was not 
created at all. The command i used was


./Gmake INS_BASE=/home/billie/cdrtools-test/ install
--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools

[Daniel Pielmeier]

Neil Bothwick schrieb:

On Sat, 05 Jul 2008 14:14:17 +0200, Daniel Pielmeier wrote:


I find it a bit sad that cdrkit is the preferred cd burning application
because it satisfies the cdrtools virtual first. This means when you
have neither cdrkit nor cdrtools installed cdrkit is preferred. However
i could not find out why. There is nothing in the profiles which defines
how cdrkit is favored. Maybe someone can point me to the right location.


It is defined in the virtuals file in your profile.


I think cdrtools should be preferred and not the fork cdrkit!


It depends on the profile, some default to cdrkit, some to cdrtools:

grep virtual/cdrtools /usr/portage/profiles/**/virtuals




grep virtual/cdrtools 
/media/system/repositories/portage/profiles/**/virtuals
/media/system/repositories/portage/profiles/base/virtuals:virtual/cdrtools 
app-cdr/cdrkit


That was what i am looking for, thanks! Why they did choose cdrkit in
favor of cdrtools is a miracle to me! Maybe I will ask the maintainer
about this!

Regards,

Daniel




--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Firefox 3 stability

[Daniel Beecham]

I agree with whoever said opera without mail is flying. It really is.
I cant see the major improvements in firefox as preposed - ie speed,
stability, memoryconsumption, etc. Opera still outdoes firefox imo.

On 7/4/08, Graham Murray <[EMAIL PROTECTED]> wrote:
> "Jason Messerschmitt" <[EMAIL PROTECTED]> writes:
>
>> I've noticed an odd visual glitch with blogspot pages- it superimposes
>> my desktop on some pages. It's a bit hard to explain so here's a
>> pic. As far as I know it's only this page, but might be good fodder
>> for some of you troubleshooters out there.
>
> There is something which lots of people have noticed. The suggested
> solution, which worked for me, was to change the X acceleration from XAA
> to EXA in /etc/X11/xorg.conf. This will only work if your video driver
> supports EXA acceleration.
> --
> gentoo-user@lists.gentoo.org mailing list
>
>

-- 
Sent from Gmail for mobile | mobile.google.com
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and large files

[Daniel Iliev]

On Fri, 04 Jul 2008 14:43:01 -0500
Dale <[EMAIL PROTECTED]> wrote:

> Hi again,
> 
> After getting a LOT of help on this, I got it sorted out.  This is
> what is installed:
> 
> [ebuild U ] app-cdr/cdrkit-1.1.8 [1.1.6] USE="unicode -hfs" 0 kB
> [ebuild U ] app-cdr/k3b-1.0.5-r1 [1.0.4] USE="alsa arts dvdr
> dvdread encode hal mp3 vorbis -css -debug -emovix -ffmpeg -flac
> -musepack -musicbrainz -sndfile -vcd -xinerama" LINGUAS="-af -ar -bg
> -br -bs -ca -cs -cy -da -de -el -en_GB -es -et -eu -fa -fi -fr -ga
> -gl -he -hi -hu -is -it -ja -ka -lt -mk -ms -nb -nds -nl -nn -pa -pl
> -pt -pt_BR -ru -rw -se -sk -sr [EMAIL PROTECTED] -sv -ta -tr -uk -uz -zh_CN
> -zh_TW" 0 kB
> 

Actually if you don't care about the GUI and compatibility with other
systems, you can do DVD burning with growisofs only (without
mkisofs/genisoimage respectively from cdrtools or cdrkit).

> I also noticed something else that was "funny".  I added the line to 
> fstab but sort of forgot something else.  This is one of my blonde 
> moments here, not blonde but anyway.  I forgot to make the mount
> point, you know, the hdd directory in /media.  < Dale slaps forehead >
> 

AFAIK this might cause problems if you use auto-mounting software and
forget to disable it before writing.


> The proof is in the puddin tho:
> 
> [EMAIL PROTECTED] / # ls -al /media/hdd/
> total 4194346
> dr-xr-xr-x 2 root root112 2008-07-04 14:22 .
> drwxr-xr-x 5 root root192 2008-07-04 14:21 ..
> -r--r--r-- 1 root root 4295007744 2008-07-04 12:32 
> Test-_2008.07.04-10.06.53_1.tar
> [EMAIL PROTECTED] / #
> 
> That's a BIG file.  Now I want a blue ray thingy.  o_O  Then I can
> make HUGE files.  lol



The good part is that you don't need isofs for DVDs. Actually
you don't need any fs at all. You could write files like this:

growisofs -Z /dev/dvd=Test-_2008.07.04-10.06.53_1.tar

Unfortunately I have I/O errors (ruined media) often with this method,
so I use:

tar cp path/to/files | \
pipebench -q -b 5000 | \
growisofs -Z /dev/sr0=/dev/fd/0

"pipebench" is in portage and provides two important advantages:
- you can check the speed and make adjustments it if necessary
- you can adjust the buffer (-b)


To restore the files from the above backup:

tar xpf /dev/dvd

It works perfectly for me.

You can burn this way images with your preferred FS.



-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] [SOLVED] f-prot

[Daniel Iliev]

On Fri, 04 Jul 2008 12:32:17 -0500
Dale <[EMAIL PROTECTED]> wrote:

>   It worked fine here. 


Thanks!

That was all I needed to know, that the problem is only mine. As Alan
proposed I sniffed the traffic and found there was a header
(User-Agent=MSIE) which was making the server say something like "this
OS is not supported". Now everything is fine.



-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Re: DMRAID and complex setups

[Daniel Iliev]

On Sun, 6 Jul 2008 11:32:36 +0200
Florian Philipp <[EMAIL PROTECTED]> wrote:

> On Sun, 6 Jul 2008 11:03:45 +0200
> Francesco Talamona <[EMAIL PROTECTED]> wrote:
> 
> > On Saturday 05 July 2008, Florian Philipp wrote:
> > > - Could I disable the kernel auto-detection and instead add a
> > > custom init-script prior to checkfs and localmount?
> > 
> > It's impractical, check carefully all partitions with
> > 
> > mdadm -E 
> > 
> > If it's not automatically assembled but you can do it manually, it
> > means something is fooling mdadm.
> > 
> > Try also to comment ARRAY lines in /etc/mdadm.conf.
> > 
> > HTH
> > Francesco
> > 
>  What actually happens is that the kernel starts md1 (consisting of
> md0 and hdd7) BEFORE md0 (hda and hdb) is created. The most likely
> cause is that the kernel starts its autodetection on the disk where /
> (or /boot?) lie.
> 
> mdadm.conf has the following lines:
> 
> DEVICE /dev/hda1 /dev/hdb1
> DEVICE /dev/md0 /dev/hdd7
> 
> ARRAY /dev/md0 devices=/dev/hda1,/dev/hdb1
> ARRAY /dev/md1 devices=/dev/md0,/dev/hdd7
> 
> Since I can add kernel parameters for starting raids
> (see /usr/src/linux/Documentation/md.txt) I'm thinking about telling
> the kernel explicitly to start md0 in the hope that this happens
> before autodetection.
> 
> Stay tuned ;)
> 

You could also try something like this in local.start:


mdadm --stop /dev/md0
mdadm --stop /dev/md1

mdadm -A /dev/md0 --uuid  -c partitions
mdadm -A /dev/md1 --uuid  -c partitions


-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools

[Daniel Pielmeier]

2008/7/7, Joerg Schilling <[EMAIL PROTECTED]>:
> Also note that these people from Debian (whose claims have been repeated) have
> ZERO credibiltiy. In September 2006, when they started cdrkit, they claimed
> that there were exactly two problems:
>
> Claim 1:"The CDDL is not a free license"
>
> Reality:The CDDL was accepted by the _whole_ ODD community at the
>end of January 2005. Everybody had the change to send his
>remarks, Debian did not. Even Debian officially accepted the
>CDDL as a definitive free license in August 2006.
>
>
> Claim 2:"The build system for a GPLd program needs to be under GPL"
>
> Reality:The people around Bloch took the last GPLd cdrtools source
>and replaced the original build system by a build system
>that is definitely not under GPL (it is under a four clause
>BSDl). "All animals are equal but some animals are more
>equal than others"?
>
> You should be very careful when you repeat the claims from people who
> repeatedly published _obvious_ false claims in order to harm the cdrtools
> project.
>
> BTW: The claims from the people around Bloch are _not_ made by lawyers but by
> laymen.

Sorry, Jörg, cdrkit does not claim any of this above. The only claim
they have is that the CDDL is not compatible with the GPL [1]
_according to the FSF_. According to _cdrkits own document_ [1] they
do not claim that the CDDL is not a free license. I also don't think
the cmake build system can not be used with cdrkit as the 4clause BSD
licencse has been declared compatible with the GPL [2] _by the FSF_ in
contrary to the CDDL. So the only thing is that the debian people are
FSF oriented and thus have dropped cdrtools. But as far as i know
Debian was always a bit fanatic in such concerns and I don't give much
about this as I don't think the FSF is the ultimate source of all
truth.

[1] http://svn.debian.org/wsvn/debburn/cdrkit/trunk/FORK?op=file&rev=0&sc=0
[2] http://www.fsf.org/licensing/licenses/

Re: SCG (was: Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools)

[Daniel Pielmeier]

2008/7/5, Daniel Pielmeier <[EMAIL PROTECTED]>:
> Joerg Schilling schrieb:
> >
> > This is a really bad idea.
> > /usr/include/scg/ is a planned directory that is known to be unique.
> >
> > cdrkit does not deliver anything that is even approximately useful as a
> replacement for libscg.
> >
> > Installing the includefiles from libscg into /usr/include/scsilib/scg
> makes them
> > unusable as there is no software that is aware of this location. It seems
> that the only result is that it makes it harder to install cdrtools instread
> of cdrkit.
> >
> > Jörg
> >
> >
>
> Out of curiosity I tried a manual install and /usr/include/scg/ was not
> created at all. The command i used was
>
> ./Gmake INS_BASE=/home/billie/cdrtools-test/ install
>

Returning to the technical side, I would really want to know why
/usr/include/scg is not installed. I tried Gmake and also tried smake,
but in both cases the libscg includes are not installed to the target
directory! Am I doing something wrong here?

Regards,

Daniel

Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools

[Daniel Iliev]

On Mon, 07 Jul 2008 11:16:33 +0200
[EMAIL PROTECTED] (Joerg Schilling) wrote:

> Mark Kirkwood <[EMAIL PROTECTED]> wrote:
> 
> > > Do you know what defamation and slander is?
> > >
> > > If people did not believe in unproven and untrue claims, there
> > > was no problem. It therefore seems to be important to prevent
> > > underlying "messages"...
> >
> > Do you understand opinion? This is the heart of the matter. Your
> > mention of lawyers does not change this - as legal opinion is
> > merely an opinion that costs some money! (if your lawyer says !Z it
> > does not stop another one saying Z)...
> 
> 
> Short answer: in a democracy, your freedom ends where you may start
> to influence the freedom of others.
> 

How about stopping these reoccurring advertisements of cdrtools on
"gentoo-user"? They influence my freedom to use the list for what it is
meant to be used: Gentoo-only related stuff. 

> 
> Long answer:
> 
> If you repeat the "opinion" of other people, you make it _your_
> opinion and if your opinion may harm other people, you are not
> allowed to publish it unless you are able to definitely prove it!
> 

Here is a fact for you: Every mainstream binary distro dropped cdrtools.
It is their right to choose which packets they want to distribute and
they don't owe you an explanation.

You claim (unfortunately on this list) that it is an attack against you
and there is no problem with your precious package.
Care to "definitely prove it"?


> I am the author and I tell you that there is no problem. I am the
> only person who could sue you and I can't if I did tell you before
> that there is no problem.
> 

You may tell whatever you want but...
You are not the ONLY author. There is other people's copyrighted work
in cdrtools. Are you authorized to speak on their behalf?


-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools

[Daniel Iliev]

On Mon, 07 Jul 2008 13:11:59 +0200
[EMAIL PROTECTED] (Joerg Schilling) wrote:

> Daniel Iliev <[EMAIL PROTECTED]> wrote:
> 
> I know I should not feed trolls but


...otherwise, you'd starve to death.

> 
> > Here is a fact for you: Every mainstream binary distro dropped
> > cdrtools. It is their right to choose which packets they want to
> > distribute and they don't owe you an explanation.
> 
> Solaris (the only distribution where lawyers checked the license)
> happily distributes the original cdrtools.
> 
> Jörg
> 


Hurray! You proved you are right! Now will you stop, please, reopening
this topic on gentoo-user?


-- 
Best regards,
Daniel
--
gentoo-user@lists.gentoo.org mailing list

Re: SCG (was: Re: [gentoo-user] emerge -avC cdrkit && emerge -av cdrtools)

[Daniel Pielmeier]

2008/7/7, Joerg Schilling <[EMAIL PROTECTED]>:
> Daniel Pielmeier <[EMAIL PROTECTED]> wrote:
>
> > Out of curiosity I tried a manual install and /usr/include/scg/ was not
> > created at all. The command i used was
> >
> > ./Gmake INS_BASE=/home/billie/cdrtools-test/ install
>
> Is this intended to be a joke or do you really like to ask me why it did not 
> do
> something while you told it to do something else?

No this is no joke at all!

I should have said $INS_BASE/include/scg was not created at all.

What i meant with the first question is that the libscg headers are
not copied to the install destination.Sorry if I expressed this wrong
in my first question.

Regards,

Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] drive configuration changes on reboot; blkid.tab defeats UUIDs

[Daniel Pielmeier]

Kevin O'Gorman schrieb:

I had cause to reboot my gentoo box this morning, and it was an
unexpected disaster.
For some reason, my two PCI-X SATA controllers decided to switch
places in the /dev/sd* lists.
My /etc/fstab had explicit drive paths hard-coded, and they tried to
mount stuff that
didn't exist, and naturally failed.  I wound up in a root shell under
instructions to clean this
up.

I decided to go with UUIDs in /etc/fstab.  After a half-hour or so
pfutzing around with these
(how do you find the UUID of an unmounted partition when you're not
even really sure
what kind of filesystem it has), I got everything to mount with "mount
-a", and I rebooted.

The drives had changed names again, the sort of thing that UUIDs were
designed to
deal with, but the mount command was stubbornly using the old names.
Bootup failed and
I was back in a root shell.  Thank goodness my root directory is still
on an HDA drive.
But where did these names come from -- they weren't in /etc/fstab any
more.  I did a system
call trace on mount(1) to find out.

There's a file I never heard of or noticed before: /etc/blkid.tab, and
a backup, that seem to
override the UUIDs, putting us back in the world we were in before
Labels and UUIDs.

G.

I can get a good boot if I rm blkid.tab and its backup before I shut down.

So:

1) Can I disable blkid.tab?  In the presence of UUIDs this seems sensible.
2) Does anyone know if labels are also defeated?  I don't feel like
rebooting any more
today just to find out.
3) Can we just stop this madness somehow?

++ kevin



I have never heard of this before too. So i tried something like this:

blkid -c /dev/null -w /etc/blkid.tab

Which should drop the old blkid.tab and create new one from scratch. See
'man blkid' for reference. I want to do this because my blkid.tab has
also some entries which are obsolete. This did not work as expected,
after a search i found that the -w switch is not working for quite some
time now [1].

So i moved both the /etc/blkid.tab and /etc/blkid.tab.old to somewhere
else and simply run blkid and a new correct /etc/blkid.tab and
/etc/blkid.tab.old have been created.

I have not rebooted yet, but I don't expect much trouble with this changes!

Regards,

Daniel

[1] http://lists.openwall.net/linux-ext4/2008/03/02/2
--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] transcode dies on revdep-rebuild

[Daniel Pielmeier]

2008/7/8, Mick <[EMAIL PROTECTED]>:

http://bugs.gentoo.org/show_bug.cgi?id=229953

Try media-video/transcode-1.0.4-r3!
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DNS poisoning fix

[Daniel Pielmeier]

Mick schrieb:

Hi All,

Have you seen this?

http://uk.news.yahoo.com/afp/20080709/ttc-us-it-internet-software-crime-e0bba4a.html

and this?

http://www.doxpara.com/

Is it merely a matter of using the right version of bind (for those who run a 
bind daemon locally), or does it go further than that?


It was already announced on the planet [1] and there is already a bug 
[2]open about it.



[1] http://planet.gentoo.org/
[2] https://bugs.gentoo.org/show_bug.cgi?id=231201
--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] tar and huge tarballs used for back-ups

[Daniel Iliev]

On Wed, 09 Jul 2008 22:26:54 -0500
Dale <[EMAIL PROTECTED]> wrote:

> Hi,
> 
> Thought I would move this problem to a new thread.  It may not be 
> related to the DVD itself.  This is what I am currently testing.  I
> used Kbackup to create a tarball in my /backup directory.  I have the
> most basic setup for Kbackup at the moment.  It creates the tarball
> but does not compress the files themselves but creates a .tar file.
> I think it takes the files, places it in the tarball then compresses
> it or something.  I'm not real sure how Kbackup does its thing.  I
> have it set to create tarballs and the limit is 4.4Gbs.  I set it
> just short of a full 4.7Gb.
> 
> I then tried to untar the file to a directory to test if it will
> untar it or not.  It failed.  I guess you figured that since I am
> posting this problem.  I'm just going to post info as I go along
> here.  Feel free to ask for more if you need something else.  This is
> the files that Kbackup generates: 
> 
> -rw-r--r--  1 dale users 4295682560 2008-07-09 20:15 
> Data_2008.07.09-20.11.01_1.tar
> -rw-r--r--  1 dale users 4295361024 2008-07-09 20:52 
> Data_2008.07.09-20.11.01_2.tar
> 
> This is what I get when I try to untar the #1 tarball:
> 
> [EMAIL PROTECTED] / # tar -xvf /backup/Data_2008.07.09-20.11.01_1.tar -C 
> /backup/test/
> data/
> data/39.gif
> data/Address-book-2006-12-27.ldif
> data/alegis-group-sears8.21.03.sxw
> data/bookmarks.html
> data/Camera-pics/
> data/Camera-pics/2006-Christmas/
> data/Camera-pics/2006-Christmas/Christmas-0001.jpeg
> tar: Skipping to next header
> tar: Error exit delayed from previous errors
> [EMAIL PROTECTED] / #
> 
> I did not shorten that.  It is the whole thing.  It did not do much
> and that file is coming off the hard drive not a DVD.  Also note,
> there is a huge list of files that are missing from that tarball.
> Basically, that tarball is about worthless.  This is from the #2
> tarball:
> 
> [EMAIL PROTECTED] / # tar -xvf /backup/Data_2008.07.09-20.11.01_2.tar -C 
> /backup/test/
> data/Gentoo-stuff/livecd-i686-installer-2007.0.iso
> tar: Skipping to next header
> <  SNIP a long list of files >
> tar: Error exit delayed from previous errors
> [EMAIL PROTECTED] / #
> 
> I didn't see any other errors in the looong list of files.  Also, it
> did untar a lot of files but I really don't know how to tell if it
> did them all or not.  It appears they are all there but I'm not 100%
> sure.  It certainly did better than the first tarball for sure.
> 
> Since I didn't use a DVD at all in this process, is this a Kbackup 
> problem or a tar problem?  That is the only two programs I used so I 
> assume it would be one of those two.  Also note, I use ext3 on that 
> partition.  All my others are reiserfs but everything done was on
> ext3.
> 
> Other info that may matter:
> 
> [ebuild   R   ] app-cdr/cdrtools-2.01.01_alpha41  USE="unicode" 0 kB
> [ebuild   R   ] app-cdr/cdrdao-1.2.2-r2  USE="encode -debug
> -gcdmaster -pccts" 0 kB
> [ebuild   R   ] app-cdr/k3b-1.0.5-r1  USE="alsa arts dvdr dvdread
> encode hal mp3 vorbis -css -debug -emovix -ffmpeg -flac -musepack
> -musicbrainz -sndfile -vcd -xinerama" LINGUAS="-af -ar -bg -br -bs
> -ca -cs -cy -da -de -el -en_GB -es -et -eu -fa -fi -fr -ga -gl -he
> -hi -hu -is -it -ja -ka -lt -mk -ms -nb -nds -nl -nn -pa -pl -pt
> -pt_BR -ru -rw -se -sk -sr [EMAIL PROTECTED] -sv -ta -tr -uk -uz -zh_CN
> -zh_TW" 0 kB
> 
> [EMAIL PROTECTED] / # uname -r
> 2.6.23-gentoo-r8
> [EMAIL PROTECTED] / #
> 
> That is the latest kernel that I can get to work correctly.
> 2.6.24-r8 makes KDE slow as leap year for some reason.  No clue, just
> went back to what worked for me. 
> 
> Thoughts?  Opinions?  Suggestions?
> 
> Thanks
> 
> Dale
> 
> :-)  :-) 


http://www.kde-apps.org/content/show.php?content=44998


"NOTE:Due to a KDE3 limitation, it can only backup files each up to 4GB
in size (on 32bit platforms at least; therefore larger files are
skipped)"

If your platform happens to be a 32bit one...

Ah! One more thing. AFAIK Kbackup has nothing to do with GNU Tar. It
utilizes KDE's KTar class, so I wouldn't seek the problem in GNU Tar
(/bin/tar). Actually you can easyly make e test:

tar cpf my_huge_archive.tar --format=posix  "/path/to/many/big/files"
tar xpf my_huge_archive.tar --format=posix -C "/path/to/test/dir/"

I bet it would work.

More info: http://www.gnu.org/software/tar/manual/html_section/Formats.html


-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] tar and huge tarballs used for back-ups

[Daniel Iliev]

On Thu, 10 Jul 2008 11:01:02 +0200
[EMAIL PROTECTED] (Joerg Schilling) wrote:

> Daniel Iliev <[EMAIL PROTECTED]> wrote:
> 
> > > Since I didn't use a DVD at all in this process, is this a
> > > Kbackup problem or a tar problem?  That is the only two programs
> > > I used so I assume it would be one of those two.  Also note, I
> > > use ext3 on that partition.  All my others are reiserfs but
> > > everything done was on ext3.

Those are not my words. Please, quote correctly.

> 
> From other hints, this seems to be a problem on both sides:
> 
> - The built in tar archiver is far too dumb and does not
> support enough features for backups
> 
> - GNU tar often does not unpack POSIX compliant tar achives 
>   and even sometimes has problems with it's own non-standard
> archives.


Care to prove this claim?

> 
> You may like to check the POSIX compliance of the tar archive you
> created to verify at which side your problem is located:

[--snip--]

> 
> > Ah! One more thing. AFAIK Kbackup has nothing to do with GNU Tar. It
> > utilizes KDE's KTar class, so I wouldn't seek the problem in GNU Tar
> > (/bin/tar). Actually you can easyly make e test:
> 

May I ask why did you snipped the quote exactly where the correct
syntax was shown and replaced it with the incorrect one that follows!?


> An important reason not to use kbackup.
> 
> Some examples:
> /*--*/
> gtar -cf - /etc/passwd | tartest
> gtar: Removing leading `/' from member names
> 0.014r 0.000u 0.000s 0% 0M 0+0k 0st 0+0io 0pf+0w
> tartest 1.10 (i386-pc-solaris2.11)
(^^^  Are you using Gentoo at all or
you just love to abuse its mailing list?)
> 
> Copyright (C) 2002 Jörg Schilling
> This is free software; see the source for copying conditions.  There
> is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
> PARTICULAR PURPOSE.
> 
> Testing for POSIX.1-1990 TAR compliance...
> Warning: illegal character ' ' (0x20) found in field 't_magic[5]'
> Warning: illegal character ' ' (0x20) found in field 't_version[0]'

[--snip--]


Examples of what? Of your incompetence to use GNU Tar?
Check this out:

tar -cf - --format=posix /etc | tartest
tartest 1.12 (x86_64-unknown-linux-gnu)

Copyright (C) 2002 J�rg Schilling
This is free software; see the source for copying conditions.  There is
NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE.

Testing for POSIX.1-1990 TAR compliance...
tar: Removing leading `/' from member names
Archive uses POSIX.1-2001 extensions.
The correctness of the size field cannot be checked for this reason.
Found 1st EOF block at 266978
Found 2nd EOF block at 266979
No deviations from POSIX.1-1990 TAR standard found.
 It seems your own
tool disagrees with you.

Read this: http://www.gnu.org/software/automake/manual/tar/posix.html

...and stop spreading FUD!



-- 
Best regards,
Daniel
--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] tar and huge tarballs used for back-ups

[Daniel Iliev]

On Thu, 10 Jul 2008 13:06:06 +0200
Dirk Heinrichs <[EMAIL PROTECTED]> wrote:

> Am Donnerstag, 10. Juli 2008 schrieb ext Dale:
> 
> > Thought I would move this problem to a new thread.  It may not be
> > related to the DVD itself.  This is what I am currently testing.  I
> > used Kbackup to create a tarball in my /backup directory.  I have
> > the most basic setup for Kbackup at the moment.  It creates the
> > tarball but does not compress the files themselves but creates
> > a .tar file.  I think it takes the files, places it in the tarball
> > then compresses it or something.  I'm not real sure how Kbackup
> > does its thing.  I have it set to create tarballs and the limit is
> > 4.4Gbs.  I set it just short of a full 4.7Gb.
> 
> Hmm, never tried it myself so I don't know wether it works or not,
> but what about enabling CONFIG_CDROM_PKTCDVD in the kernel and let
> tar write to the device directly?
> 
> Like:
> 
> modprobe pktcdvd
> pktsetup backup /dev/cdrom
> tar -cMvf /dev/pktcdvd/backup myfiles # Change disc when prompted
> pktsetup -d backup
> rmmod pktcdvd
> 
> Bye...
> 
>   Dirk

AFAIK packet writing is intended for using RW media as a "normal"
block device. I have never tried it myself either.

If you just wanted to have a file system (other than isofs) on a
write-once *DVD*, you can do something like:

dd if=/dev/null of=test.fs bs=1M seek=4480 count=0
mkfs.ext2 test.fs
mkdir loopdir
mount -o loop test.fs loopdir 
cp -a "some files & dirs" loopdir/
umount test.fs
rm -r loopdir
pipebench -b 5000 < test.fs | growisofs -Z /dev/sr0=/dev/fd/0
mount /dev/dvd /mnt/dvd


-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] {OT} Audio CD pre-gap info? (cdrdao?)

[Daniel Iliev]

On Thu, 10 Jul 2008 07:30:54 -0700
Grant <[EMAIL PROTECTED]> wrote:

> Does anyone know of a tool that will provide info on audio CD
> pre-gaps?  I've read that cdrdao-utility will do it, but that doesn't
> seem to arrive with Gentoo cdrdao and I don't see a separate package
> for it.  It looks like 'cdparanoia -Q' should do it, but it outputs
> "pre no" regardless of the pre-gap.
> 
> - Grant


Perhaps: cdparanoia -Qvs


-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

[gentoo-user] gmail delays

[Daniel Iliev]

Hi,

Is anybody else observing big (several tens of minutes) delays when
receiving mail from gentoo-user to a Gmail account? 

-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] gmail delays

[Daniel Iliev]

On Thu, 10 Jul 2008 19:55:44 +0200
Volker Armin Hemmann <[EMAIL PROTECTED]> wrote:

> On Donnerstag, 10. Juli 2008, Daniel Iliev wrote:
> > Hi,
> >
> > Is anybody else observing big (several tens of minutes) delays when
> > receiving mail from gentoo-user to a Gmail account?
> >
> 
> I am seeing big delays with my non-gmail account.

Thanks for your reply.

I believe my problem is "Gentoo <-> Gmail" related.

I checked the headers of a message I've sent to the list and it appears
the list has sent it back to GMail with delay but OTOH the message
arrived at gmane.org in time. Strange.

-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] tar and huge tarballs used for back-ups

[Daniel Iliev]

On Thu, 10 Jul 2008 19:05:55 +0200
Dirk Heinrichs <[EMAIL PROTECTED]> wrote:

> Am Donnerstag, 10. Juli 2008 schrieb Daniel Iliev:
> > On Thu, 10 Jul 2008 13:06:06 +0200
> >
> > Dirk Heinrichs <[EMAIL PROTECTED]> wrote:
> > > Hmm, never tried it myself so I don't know wether it works or not,
> > > but what about enabling CONFIG_CDROM_PKTCDVD in the kernel and let
> > > tar write to the device directly?
> > >
> > > Like:
> > >
> > > modprobe pktcdvd
> > > pktsetup backup /dev/cdrom
> > > tar -cMvf /dev/pktcdvd/backup myfiles # Change disc when prompted
> > > pktsetup -d backup
> > > rmmod pktcdvd
> >

[-snip-]
 
> > If you just wanted to have a file system (other than isofs) on a
> > write-once *DVD*, you can do something like:
> 
> No, why would I? The idea behind the above is to let _tar_ write to
> the device directly, as it would do with tape devices. No fs
> involved.

Ah! Now I got it. "tar -M". I've missed that part. I'm sorry.

> OTOH, if I wanted a filesystem, that would be UDF, which I
> could still write to using the method above (+ create the fs with
> cdrwtool before running pktsetup and then mount it, of course).
> 
> > dd if=/dev/null of=test.fs bs=1M seek=4480 count=0

[-snip-]

> 
> And that would give me a multi volume backup if the data doesn't fit
> on one disc?
> 

No, no. It has nothing to do with multi-volumes. As I said I didn't see
the "-M" part from your message and merely showed that one can write
files directly without packet writing and those files can be FS images.

BTW there is a tool called "splipipe" [1] that can emulate the
behaviour of "tar -M" without packet writing. I've used it and it
worked perfectly.

[1] http://ds9a.nl/splitpipe/

Please don't get me wrong. I'm not advising against packet writing.
I'm merely sharing the methods for making backup on DVDs I'm aware of in
hope someone could benefit from the info.


-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] tar and huge tarballs used for back-ups

[Daniel Iliev]

On Fri, 11 Jul 2008 08:12:50 +0200
Dirk Heinrichs <[EMAIL PROTECTED]> wrote:


> 
> Didn't know that tool, looks nice. Would you mind creating a "new
> ebuild request" for it in b.g.o?
> 
> So we've two methods now, great.
> 

Actually I have already done this [1] a few days ago. Unfortunately it
is the first time I'm trying to create an ebuild and I have no
experience with the interface of B.G.O. therefore the result is quite
unsatisfactory. I used an ebuild from forums.gentoo.org as a template.
It installs the package in "/usr/local".

Any help will be much appreciated.

[1] http://bugs.gentoo.org/show_bug.cgi?id=230813

-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] how does Gentoo's mke2fs determine how many inodes to create?

[Daniel Iliev]

On Fri, 11 Jul 2008 13:16:53 +0200
Miernik <[EMAIL PROTECTED]> wrote:

> I installed Gentoo using the handbook, and the root partition has
> 4094951424 bytes (a 4 GB USB pendrive), and "mke2fs -j /dev/sda2" as
> on
> http://www.gentoo.org/doc/en/handbook/handbook-amd64.xml?part=1&chap=4#doc_chap4
> created me a partition with only 249984 inodes. That was REALLY SILLY
> of him, because:
> 
> przehyba ~ # df -i /dev/sda2
> FilesystemInodes   IUsed   IFree IUse% Mounted on
> /dev/sda2 249984  249739 245  100% /
> przehyba ~ #
> 
> I ran out of inodes! Now it cries that the disk ran out of space when
> I try to compile anything. I did already get rid of /usr/share/doc/ as
> there are many small files there, and purged /var/tmp/ too, but again
> emerge quickly fills all possible inodes, and fails.
> 
> Why did he did so little inodes? On another machine, which is Debian
> and about 5 GB disk, I have 752192 inodes and even on a 3.2 GB one
> 376096 inodes, so here why did I get so little by default?
> 
> What can I do now, besides starting the installation over again from
> the beggining, which would be 2 days of work lost!
> 
> Any other places besides /usr/share/doc and /var/tmp/ I can look to
> recover a lot of inodes? Can someone fix mke2fs so it doesn't do that
> to anyone again in the future?
> 


/etc/mke2fs.conf -  this file contains the default options plus several
presets.

The number of inodes can be specified at creation time with the "-N"
parameter. AFAIK it cannot be changed afterwards.



-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] how does Gentoo's mke2fs determine how many inodes to create?

[Daniel Iliev]

On Fri, 11 Jul 2008 13:35:02 +0100
Anthony Metcalf <[EMAIL PROTECTED]> wrote:

> Daniel Iliev wrote:
> > On Fri, 11 Jul 2008 13:16:53 +0200
> > Miernik <[EMAIL PROTECTED]> wrote:
> >
> >   
> > /etc/mke2fs.conf -  this file contains the default options plus
> > several presets.
> >
> > The number of inodes can be specified at creation time with the "-N"
> > parameter. AFAIK it cannot be changed afterwards.
> >
> >
> >   
> You can though, dd the partition to another drive, re-format, and dd
> it back
> 
> Anthony


No, not really. "dd" is not good for this case because it works on
lower level and will copy and restore the file system also. A normal
archive program like GNU Tar should do the trick.

-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] tar and huge tarballs used for back-ups

[Daniel Iliev]

On Sun, 13 Jul 2008 17:39:17 +0300
"Arttu V." <[EMAIL PROTECTED]> wrote:

> On 7/12/08, Arttu V. <[EMAIL PROTECTED]> wrote:
> > then see if I get around later to try with cdrtools instead of
> > wodim (cdrkit).
> 
> Ok, my final conclusion on splitpipe: it compiles but won't work for
> me.
> 
> I've produced about a dozen coasters while having tried some
> combinations of two different machines/arches (x86, amd64, both with
> stable core packages), two different burners (trusty last millennium
> HP CD-Writer+ 9100 series and newish el cheapo LG branded
> do-everything hyper-duper-combo thingie), two different tars (GNU tar
> and star) and two different pieces of burning software (cdrkit and
> cdrtools).
> 
> Consistently, the failure is at the end of the first volume (disc),
> where splitpipe must have been writing some broken bytes at the end of
> the volume (possibly miscalculating something first?) -- which it
> realizes when it does the reading (as joinpipe) and then there's the
> barf ("Fatal:" probably from splitpipe) and then bad input propagates
> to tar/zip/others which join in in the mass-barfing contest:
> 
> UUID of this session is '35a141883a219349 0c2b7efad2ffdaf7'
> joinpipe: volume was started on Sun Jul 13 16:38:38 EEST 2008
> joinpipe: found volume 1, as expected
> star: WARNING: skipping leading '/' on filenames.
> Fatal: during read of a stretch of input: Input/output error
> star: Tar file too small (amount: 101 bytes).
> star: Unexpected EOF on input.
> star: Cannot recover from error - exiting.
> star: 68552 blocks + 5733 bytes (total of 701978213 bytes =
> 685525.59k).
> 

Dirk, Arttu,

Sorry for the late reply. I was out for the weekend.

I've tested splitpipe with DVDs on stable amd64 only. I had no
problems. Later today I'll give the improved ebuild a try.
Currently I have several blank DVDs only, so I can't send any CD burning
results.

I have some questions:

   Dirk:
 - can you post the compile error, please?

   Arttu:
 - Can you, please, post the command you used to make the test backups?
 - Did you try splitpipe on DVDs or on CDs only?

   Dirk, Arttu:
 - Should we get off-list on this subject or at least open a new
thread? 


-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] [OT] Network chip always comes up eth1 on 1-year-old Dell Inspiron 530

[Daniel Iliev]

On Sun, 13 Jul 2008 11:30:56 +0100
Neil Bothwick <[EMAIL PROTECTED]> wrote:

> On Sat, 12 Jul 2008 19:55:56 -0400, Walter Dnes wrote:
> 
> > > Udev is doing this. If you have removed the second card,
> > > delete /etc/udev/rules.d/70-persistent-net.rules, otherwise edit
> > > the file to switch the assignments for the two NICs.  
> > 
> >   Thanks.  A "new and improved helpfull feature" that could've done
> > without.
> 
> It's hardly new, it's been around for some years. It is helpful if you
> have two NICs because it means they are named consistently, which is
> better than having your private network connected to the Internet
> because the kernel decided to load the modules in a different order.
> 
> 


IIRC:

1) You can explicitly tell the kernel the order in which load the
modules
2) If you build the the drivers in-kernel the the order is determined
by the PCI slot numbers



-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Problems compiling gcc 4.2.4 and 4.3.1-r1

[Daniel Pielmeier]

2008/7/14, John covici <[EMAIL PROTECTED]>:
>  > > Portage 2.1.4.4 (default/linux/x86/2008.0/developer, gcc-4.1.2, 
> glibc-2.6.1-r0, 2.6.24-gentoo-r5 i686)
>  >^^
>  > that's your major problem...
>  >
>
> Which of those is the problem -- I am running 32-bit operating
> systems.

Maybe not to use the developer profile :-)
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Uwe Thiem

[Daniel Pielmeier]

Alan McKinnon schrieb:
It's a sad day for me today. Uwe Thiem, who has been on this list longer 
than I have, passed away on Friday. My friend Alastair put up a little 
tribute here:


http://www.tectonic.co.za/?p=2574

What that page doesn't say is how much Uwe loved gentoo. A long-time 
subscriber around here, his posts must number in the 100s if not 1000s. 
I'm gonna miss him, even though he lived only 1000km away (a trivial 
distance in Africa) I never did get to collect that pitcher of 
Namibia's finest he owed me. I guess now I never will.





This are indeed very sad news. He was truly an enrichment to this list!
I count 554 messages from him to this list from the beginning of 2006 
when I joined! Google Groups lists 804 since 2004.

--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

Dale schrieb:


OK folks, still looking for a fix here.  Any ideas at all?



There is a bug [1] open about this problem, unfortunately still 
unsolved! Maybe the k3b command line will help to investigate on this 
further.


[1] http://bugs.gentoo.org/show_bug.cgi?id=213399
--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

Dale schrieb:

Daniel Pielmeier wrote:

Dale schrieb:


OK folks, still looking for a fix here.  Any ideas at all?



There is a bug [1] open about this problem, unfortunately still 
unsolved! Maybe the k3b command line will help to investigate on this 
further.


[1] http://bugs.gentoo.org/show_bug.cgi?id=213399


If you saw the post on the thread referenced on the bug, the last post 
was mine.  I'm dalek on the forums.  At least now I know it is not still 
just me.

I wonder if cdrkit has this issue?  Think it would be worth a shot?

Dale

:-)  :-)


I think it is also possible that there is something with k3b as 
everybody with this problem is using k3b. I burn my DVD's on the command 
line.


Here is some information from a video-dvd:

ISO-creation:
mkisofs -dvd-video -udf -R -iso-level 1 -v -volid DVD_VIDEO -sysid LINUX 
-no-cache-inodes -input-charset utf-8 -o ../dvd-video.iso ./


Burning:
cdrecord dev=0,1,0 speed=8 driveropts=burnfree fs=16m -sao -eject -v 
dvd-video.iso


Mount:
/dev/sr0 on /media/DVD_VIDEO type udf (ro,nosuid,nodev,uid=1000)

Mount Point:
dr-xr-xr-x  4 billie root   136 11. Jul 23:03 DVD_VIDEO

The DVD is auto mounted by xfce-extra/thunar-volman which uses hal.


When k3b finishes you can get the command line used by k3b, maybe you 
can post it.


Regards,

Daniel
--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Uwe Thiem

[Daniel Iliev]

On Mon, 14 Jul 2008 17:53:35 +0200
Alan McKinnon <[EMAIL PROTECTED]> wrote:

> It's a sad day for me today. Uwe Thiem, who has been on this list
> longer than I have, passed away on Friday. My friend Alastair put up
> a little tribute here:
> 
> http://www.tectonic.co.za/?p=2574
> 
> What that page doesn't say is how much Uwe loved gentoo. A long-time 
> subscriber around here, his posts must number in the 100s if not
> 1000s. I'm gonna miss him, even though he lived only 1000km away (a
> trivial distance in Africa) I never did get to collect that pitcher
> of Namibia's finest he owed me. I guess now I never will.
> 
> 


Terrible news :(

May he rest in peace.

-- 
Best regards,
Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

>
> You, or anybody else, notice anything strange, odd and/or just plain wrong?
> Would a different version of k3b help?  I'm not sure what is causing this
> problem just that I have one.  I would also be willing to try a command line
> if you can provide me with the 'correct' options to test with.
> Thanks.

I did a bit testing last night and found out that the files and
directories in the resulting images created by mkisofs all have
permissions 000. What I see from the man pages is that the -udf and -r
options are setting the user and group permissions to zero. When I
used -UDF and -R instead I also get zero values. I checked this by
using isoinfo to analyse the files within the image.

This however there seems to be no problem for me as the DVD's are
automounted with permissions 555 for folders and 444 for files.

So what is the problem here, mkisofs creating images with wrong
permissions, or this permissions are normal and some automounter or
mount in general are/is mounting the DVD with wrong permissions
probably with interference of the udf kernel driver.
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

2008/7/15, Joerg Schilling <[EMAIL PROTECTED]>:
> > I did a bit testing last night and found out that the files and
> > directories in the resulting images created by mkisofs all have
> > permissions 000. What I see from the man pages is that the -udf and -r
> > options are setting the user and group permissions to zero. When I
> > used -UDF and -R instead I also get zero values. I checked this by
> > using isoinfo to analyse the files within the image.
> >
> > This however there seems to be no problem for me as the DVD's are
> > automounted with permissions 555 for folders and 444 for files.
> >
> > So what is the problem here, mkisofs creating images with wrong
> > permissions, or this permissions are normal and some automounter or
> > mount in general are/is mounting the DVD with wrong permissions
> > probably with interference of the udf kernel driver.
>
> UDF permissions created by mkisofs have been verified by mounting the images 
> on
> Solaris. As I could not see any deviations from the original permissions, I
> suspect a problem in the Linux kernel.
>
> If you have no Solaris, it could help to make a mount test with a Rock Ridge
> mount. If the permissions are OK with Rock Ridge, then is is a bug in the UDF
> implementation in the Linux kernel.

I have opensolaris in a virtual machine but i don't think this is
enough to verify it correctly, so i will try the rock ridge mount.
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

2008/7/15, Dale <[EMAIL PROTECTED]>:
>
> LOL.  Since there is no fstab entry, you will have to enlighten me a bit.
> How do I tell hal to mount it that way instead of udf?
>
> Hmmm, guess I could turn off hal and try manually.  BRB  This is odd.
> [EMAIL PROTECTED] / # cat /proc/filesystems
> nodev   sysfs
> nodev   rootfs
> nodev   bdev
> nodev   proc
> nodev   binfmt_misc
> nodev   sockfs
> nodev   usbfs
> nodev   pipefs
> nodev   anon_inodefs
> nodev   futexfs
> nodev   tmpfs
> nodev   inotifyfs
> nodev   devpts
>   reiserfs
>   ext3
>   ext4dev
>   ext2
> nodev   ramfs
>   msdos
>   vfat
>   iso9660
> nodev   nfs
> nodev   nfsd
>   ntfs
> nodev   autofs
>   udf
>   xfs
> nodev   mqueue
> nodev   rpc_pipefs
> [EMAIL PROTECTED] / #
> I don't seem to have a Rock Ridge.  I couldn't find it in the kernel config
> to enable it and equery doesn't list one either.  Where does it come from
> exactly?
>
> Going to try to disable Rock Ridge in k3b and see what blows up.
>

Unmount the disk and then try to mount it manually.

mount -t iso9660 device mount-point
This mounts as iso9660 with rock-ridge extension.

mount -t udf device mount-point
This mounts as udf.

For further information it will be a good idea to look into the mount manpage.
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

2008/7/15, Joerg Schilling <[EMAIL PROTECTED]>:
> "Daniel Pielmeier" <[EMAIL PROTECTED]> wrote:
>
> > > UDF permissions created by mkisofs have been verified by mounting the 
> > > images on
> > > Solaris. As I could not see any deviations from the original permissions, 
> > > I
> > > suspect a problem in the Linux kernel.
> > >
> > > If you have no Solaris, it could help to make a mount test with a Rock 
> > > Ridge
> > > mount. If the permissions are OK with Rock Ridge, then is is a bug in the 
> > > UDF
> > > implementation in the Linux kernel.
> >
> > I have opensolaris in a virtual machine but i don't think this is
> > enough to verify it correctly, so i will try the rock ridge mount.
>
> This is sufficient, just copy the image into the virtual machine, set up a
> "lofi" device for the image and then call mount -F udfs 
>

I have set up the machine recently, and I think there is now way to
share files between Host and Guest OS in Virtualbox when the guest is
Solaris. So I will burn the Live CD and try it within the Live CD.
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

2008/7/15, Joerg Schilling <[EMAIL PROTECTED]>:

> >
> > I have set up the machine recently, and I think there is now way to
> > share files between Host and Guest OS in Virtualbox when the guest is
> > Solaris. So I will burn the Live CD and try it within the Live CD.
>
> Any virtualization solution I know supports at least scp or ftp to
> bring in files.

Thanks I will try this first!
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Mail on multiple laptops

[Daniel Pielmeier]

2008/7/15, Michael Pobega <[EMAIL PROTECTED]>:
> I've always had just one laptop so I've been a POP user for years now,
> but I'm finally buying a second laptop for travels (Asus Eee PC) and I
> was wondering what gentoo-user would recommend as a good way to keep the
> mail synced on both of my machines.
>
> I was looking into IMAP but I can't really figure out how it works, and
> Google doesn't offer IMAP, and I'd like to keep this e-mail address. Can
> anyone make any suggestions?
>

IMAP is possible with gmail!

http://mail.google.com/support/bin/answer.py?hl=en&answer=75725
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

I think I don't need testing on an other OS. It is working flawless 
here. I have created to iso-images one as data and the other one as 
video with the following calls:


VIDEO:
mkisofs -dvd-video -UDF -R -iso-level 1 -v -volid DVD_VIDEO -sysid LINUX 
-input-charset utf-8 -o ../dvd-video.iso ./


DATA:
mkisofs -UDF -R -iso-level 1 -v -volid DVD_DATA -sysid LINUX 
-input-charset utf-8 -o ../dvd-data.iso ./


I created to mount-points where only root has permissions

ls -al ./

drwxr-xr-x  4 billie users72 15. Jul 18:29 .
drwx-- 16 billie users  8192 15. Jul 18:22 ..
-rw-r--r--  1 billie users 142530560 15. Jul 18:26 dvd-data.iso
-rw-r--r--  1 billie users 142587904 15. Jul 18:25 dvd-video.iso
drwx--  2 root   root  1 15. Jul 18:29 mount-data
drwx--  2 root   root  1 15. Jul 18:29 mount-video

Then I mounted both the data and the video image as udf and as iso9660

mount -o loop -t udf dvd-video.iso mount-video/
ls -al mount-video/

drwxr-xr-x 4 billie users 136 15. Jul 18:24 .
drwxr-xr-x 4 billie users  72 15. Jul 18:29 ..
drwxr-xr-x 2 billie users  40 15. Jul 18:24 AUDIO_TS
drwxr-xr-x 2 billie users 300 15. Jul 18:24 VIDEO_TS


mount -o loop -t iso9660 dvd-video.iso mount-video/
ls -al mount-video/

drwxr-xr-x 4 billie users 2048 15. Jul 18:24 .
drwxr-xr-x 4 billie users   72 15. Jul 18:29 ..
drwxr-xr-x 2 billie users 2048 15. Jul 18:24 AUDIO_TS
drwxr-xr-x 2 billie users 2048 15. Jul 18:24 VIDEO_TS


mount -o loop -t udf dvd-data.iso mount-data
ls -al mount-data

drwxr-xr-x 2 billie users92 15. Jul 18:26 .
drwxr-xr-x 4 billie users72 15. Jul 18:29 ..
-rw-r--r-- 1 billie users 141672448 15. Jul 18:21 001_cut.mpg


mount -o loop -t iso9660 dvd-data.iso mount-data
ls -al mount-data

drwxr-xr-x 2 billie users  2048 15. Jul 18:26 .
drwxr-xr-x 4 billie users72 15. Jul 18:29 ..
-rw-r--r-- 1 billie users 141672448 15. Jul 18:21 001_cut.mpg

Everything works perfect and the old permissions of the files and 
folders have been restored.


It was jst isoinfo which did not show the permissions:

isoinfo -l -i dvd-data.iso

Directory listing of /
d-   000   2048 Jul 15 2008 [266 02] .
d-   000   2048 Jul 15 2008 [266 02] ..
--   000  141672448 Jul 15 2008 [268 00] 001_CUT.MPG;1


isoinfo -l -i dvd-video.iso

Directory listing of /
d-   000   2048 Jul 15 2008 [274 02] .
d-   000   2048 Jul 15 2008 [274 02] ..
d-   000   2048 Jul 15 2008 [275 02] AUDIO_TS
d-   000   2048 Jul 15 2008 [276 02] VIDEO_TS

Directory listing of /AUDIO_TS/
d-   000   2048 Jul 15 2008 [275 02] .
d-   000   2048 Jul 15 2008 [274 02] ..

Directory listing of /VIDEO_TS/
d-   000   2048 Jul 15 2008 [276 02] .
d-   000   2048 Jul 15 2008 [274 02] ..
--   000   6144 Jul 15 2008 [281 00] VIDEO_TS.BUP;1
--   000   6144 Jul 15 2008 [278 00] VIDEO_TS.IFO;1
--   000  12288 Jul 15 2008 [  69466 00] VTS_01_0.BUP;1
--   000  12288 Jul 15 2008 [284 00] VTS_01_0.IFO;1
--   000  141672448 Jul 15 2008 [290 00] VTS_01_1.VOB;1

So I also guess this has to be a problem with hal, udev, k3b or even 
growisofs. The strange thing is that it works manually mounted as 
iso9960 and not as udf filesystem. So it does not look like a mounting 
problem.

--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

Joerg Schilling schrieb:

Daniel Pielmeier <[EMAIL PROTECTED]> wrote:


It was jst isoinfo which did not show the permissions:

isoinfo -l -i dvd-data.iso

Directory listing of /
d-   000   2048 Jul 15 2008 [266 02] .
d-   000   2048 Jul 15 2008 [266 02] ..
--   000  141672448 Jul 15 2008 [268 00] 001_CUT.MPG;1


isoinfo does not understand UDF and if you don't tell it to use Rock Ridge, 
it will be limited to the missing permissions from plain ISO-9660


Jörg



You are right something like this "isoinfo -l -R -i dvd-video.iso" did 
the trick. The -R flag prints the informations from the rock ridge 
extension.


By the way would you recommend those commands for creating dvd-video 
material, or should there be something changed. I mean it should be most 
compatible. Playable on PC and standalone players and different 
operating systems.


Iso creation:
mkisofs -dvd-video -UDF -R -iso-level 1 -v -volid DVD_VIDEO -sysid LINUX 
-input-charset utf-8 -o ../dvd-video.iso ./


Burning:
cdrecord dev=0,1,0 speed=8 driveropts=burnfree fs=16m -sao -eject -v 
dvd-video.iso


Regards,

daniel
--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

2008/7/16, Joerg Schilling <[EMAIL PROTECTED]>:
> Daniel Pielmeier <[EMAIL PROTECTED]> wrote:
>
> > By the way would you recommend those commands for creating dvd-video
> > material, or should there be something changed. I mean it should be most
> > compatible. Playable on PC and standalone players and different
> > operating systems.
> >
> > Iso creation:
> > mkisofs -dvd-video -UDF -R -iso-level 1 -v -volid DVD_VIDEO -sysid LINUX
> > -input-charset utf-8 -o ../dvd-video.iso ./
>
> Why do you like to do things like: -sysid LINUX that are default?

Okay so I leave this out.

> Why do you use things like: -input-charset utf-8 which is may be wrong
> and which is by default handled correctly by mkisofs?

I use utf-8 on my system, so i thought it is correct.

Can you please explain me what are $INS_BASE/lib/profiled
$INS_BASE/lib/siconv for. These two directories are not installed by
the gentoo installation. I ask this because I guess the siconv folder
has something to do with the locales. When I for example specify an
unknown locale mkisofs is complaining about a missing
$INS_BASE/lib/siconv folder.

> Why do you like to use iso level 1?

I thought it is enough for the DVD file structure.

From the manpage:
With level 1, files may only consist of one section and filenames are
restricted to 8.3 characters.

AUDIO_TS
VIDEO_TS
   VIDEO_TS.BUP
   VIDEO_TS.IFO
   VIDEO_TS.VOB
   VTS_01_0.BUP
   VTS_01_0.IFO
   VTS_01_0.VOB
   VTS_01_1.VOB
   VTS_01_2.VOB

For this file names ISO-Level 1 maybe sufficient, or am I wrong?

> > Burning:
> > cdrecord dev=0,1,0 speed=8 driveropts=burnfree fs=16m -sao -eject -v
> > dvd-video.iso
>
> If you only have one drive, you may omit the dev= parameter.
> Jörg

I have two DVD drives so I think it is needed!

Thanks,

Daniel

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

2008/7/16, Joerg Schilling <[EMAIL PROTECTED]>:
> >
> > I use utf-8 on my system, so i thought it is correct.
> >
> > Can you please explain me what are $INS_BASE/lib/profiled
> > $INS_BASE/lib/siconv for. These two directories are not installed by
> > the gentoo installation. I ask this because I guess the siconv folder
> > has something to do with the locales. When I for example specify an
> > unknown locale mkisofs is complaining about a missing
> > $INS_BASE/lib/siconv folder.
>
> This is correct, your installation is incomplete.
>
> Mkisofs cannot work correctly in all cases if siconv data is missing.
>
> When I replaced the old libunls code by libsiconv more than a year ago,
> there was a verbose instruction in the announcement file. I thought that
> any maintainer would read this and update his install procedures...
>
> If you omit -input-charset, mkisofs will chose the right values.

I have a bug open addressing this and a few other things. Do you
remember the Announcement file, so I can add this information to the
bug! If not i will delve through the files myself.

Can you give me information about the $INS_BASE/lib/profiled folder
too. I mean the contents beside the size of the files are the same as
in $INS_BASE/lib/.

> > > Why do you like to use iso level 1?
> >
> > I thought it is enough for the DVD file structure.
> >
> > From the manpage:
> > With level 1, files may only consist of one section and filenames are
> > restricted to 8.3 characters.
> >
> > AUDIO_TS
> > VIDEO_TS
> >VIDEO_TS.BUP
> >VIDEO_TS.IFO
> >VIDEO_TS.VOB
> >VTS_01_0.BUP
> >VTS_01_0.IFO
> >VTS_01_0.VOB
> >VTS_01_1.VOB
> >VTS_01_2.VOB
> >
> > For this file names ISO-Level 1 maybe sufficient, or am I wrong?
>
> Well, level 1 is default too ;-)

okay :-)

> > > > Burning:
> > > > cdrecord dev=0,1,0 speed=8 driveropts=burnfree fs=16m -sao -eject -v
> > > > dvd-video.iso
> > >
> > > If you only have one drive, you may omit the dev= parameter.
> > > Jörg
> >
> > I have two DVD drives so I think it is needed!
>
> Correct (but you may create /etc/default/cdrecord with the right content to
> default to the drive you usually like to use).
>

I think I prefer passing it at the command line.

Regards,

Daniel
z�b�� z{h���x%��

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

Dale schrieb:


This appears to be a udf problem I would assume with the kernel.  If I 
mount the DVD as iso9660 then it works fine but if it mounts with udf 
the permissions are wrong.  Is there a way to tell hal, ivman and 
friends to always use iso9660?  At least until there is a fix in the 
kernel anyway.




So it would be a good idea to add this information to the bug. If it is 
not known already and nobody reports it, there will never be a fix! :-)


Regards,

Daniel
--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] DVD and changing permissions

[Daniel Pielmeier]

2008/7/17, Joerg Schilling <[EMAIL PROTECTED]>:
> >
> > I have a bug open addressing this and a few other things. Do you
> > remember the Announcement file, so I can add this information to the
> > bug! If not i will delve through the files myself.
>
> Mmmm, don't you know "grep"?

s/delve/grep/ :-)
I searched for it yesterday evening, when asking for the file I had no
cdrtools-sources at hand.

> > Can you give me information about the $INS_BASE/lib/profiled folder
> > too. I mean the contents beside the size of the files are the same as
> > in $INS_BASE/lib/.
>
> in profiles/ you find developer libs in case you like to use the BSD code
> profiler "gprof". Whether you like to install them depends on what you like to
> do.
>

Thanks for the clarification!

Daniel
-- 
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] Video DVD Creation

[Daniel Pielmeier]

sean schrieb:

Jorg,

Another update. I had made two attempts burning my movie to a 
no-brand DVD-R.


Not wanting to waste another DVD-R I instead used a brand name DVD+RW 
for another test.
It worked, the movie played. When I get the chance I will pick up a 
brand name DVD-R and try the burn again.


Thanks for the help,
Sean


I had problems with burning DVD-VIDEO to DVD-R too. Now i use DVD+R and 
it works without problems at least at the PC (had no chance to test on 
standalone players yet).


When burning to DVD-R the burn process starts, it burns 1MB or at least 
the progress shows 1MB then it stucks for a few seconds after that the 
burn process resumes and the rest is burned normally.


Regards,

Daniel
--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] mount: "special device /dev/hdc does not exist". What does this mean?

[Daniel Pielmeier]

Alan Mackenzie schrieb:

Hi, Gentoo?

I've a newly installed system, now working with my own special
optimiesed keyboard layout.  :-)

However, I can't access my DVD drives.  I know at least one of them
works, because I installed Gentoo from it.

When I do

   mount -tiso9660 /dev/hdc /cdrom

, it comes back with "special device /dev/hdc does not exist".  And yes,
there was a CD in the drive, and /cdrom exists.


Do you mean

mount -t iso9660 /dev/hdc /cdrom


What does "special device" mean here?  Does it mean the physcial
hardware, the controller chip, the directory entry /dev/hdc, the driver
in the kernel, or what?  What is "special" about my DVD writer?


It means the directory entry /dev/hdc.


Well, to answer some of my questions, I was missing a /dev/hdc, so I
made one with

# mknod /dev/hdc b 22 0

.  This didn't help one iota.  I had a look at dmesg, but there was no
mention of hdc in it.  (It did mention hdg, hdh, where my main hard
drives are (don't ask!)).


Can't resist what is on hd{a-f}


My kernel is an up to date linux-2.6.25-gentoo-r6.  I _think_ it's got
all the needed options set in the configuration.  Can anybody suggest
how to get my system to recognise my DVD drives?

Thanks in advance!



What kind of DVD writer do you have maybe it is sata or scsi, and it 
wiil appear under /dev/srX or /dev/sgX. Or if you use the new libata 
library in the kernel even IDE devices are under /dev/srX or /dev/sgX.

--
gentoo-user@lists.gentoo.org mailing list

Re: [gentoo-user] mount: "special device /dev/hdc does not exist". What does this mean?

[Daniel Pielmeier]

Alan Mackenzie schrieb:

When I do



  mount -tiso9660 /dev/hdc /cdrom



, it comes back with "special device /dev/hdc does not exist".  And yes,
there was a CD in the drive, and /cdrom exists.



Do you mean



mount -t iso9660 /dev/hdc /cdrom


Maybe.  Is that different?



Yeah, there is a space between -t and iso9660 :-)



What kind of DVD writer do you have maybe it is sata or scsi, and it 
wiil appear under /dev/srX or /dev/sgX. Or if you use the new libata 
library in the kernel even IDE devices are under /dev/srX or /dev/sgX.


No, the box is no longer young, and contains no SATA or SCSI bits at all.
I'm just going away to see if I've got any /dev/s[gr]X on the box. 

No, I've got no /dev/s[gr]X at all.

Could it be that the kernel has looked at hd[ab], found nothing there,
and therefore decided "it's not worth the bother even looking at
hd[cd]"?



Hmm, maybe the output of `dmesg`, `lspci -v` and `ls -al /dev` could be 
helpful. Probably also your kernel configuration.

Re: [gentoo-user] Set "max locked memory" to unlimited

[Daniel Iliev]

On Thu, 24 Jul 2008 10:01:17 +0200
Justin <[EMAIL PROTECTED]> wrote:

> Hi all,
> 
> I tryed to set the "max locked memory" for a user to unlimited.
> I did this in that way,
> opened shell
> su
> ulimit -l unlimited
> exit
> ulimit -a
> 
> And the unlimited was gone and reset to 32k. As root it was set right.
> 
> What did I do wrong, or where else I have to change things?
> 
> 
> Thanks,
> justin
> 

Assuming your system is PAM-enabled, you can set the limits by editing
"/etc/security/limits.conf".

The new limits will not affect sessions which are already started. In
other words you need to log off/on for the new settings to take effect.

More info: man(5) limits.conf


-- 
Best regards,
Daniel

Re: [gentoo-user] Set "max locked memory" to unlimited

[Daniel Iliev]

On Fri, 25 Jul 2008 18:10:03 +0200
Justin <[EMAIL PROTECTED]> wrote:

> 
> >
> > Assuming your system is PAM-enabled, you can set the limits by
> > editing "/etc/security/limits.conf".
> >
> > The new limits will not affect sessions which are already started.
> > In other words you need to log off/on for the new settings to take
> > effect.
> >
> > More info: man(5) limits.conf
> >
> >
> >   
> 
> That works, but how can I set it to unlimited?
> 


If you have no entry there should be no limit. Alternatively, I
believe, a dash ("-") can be used to specify "unlimited".



-- 
Best regards,
Daniel

Re: [gentoo-user] media-video/totem-2.20.3 Access violation

[Daniel Pielmeier]

All the problems above are related to python byte code, and the way it
is handled in gentoo. It would be to exhaustive to explain the
problems here but maybe this bug [1] helps to understand the problem.
Also if you want to learn more about python in gentoo you can read
this [2].

In the case of scons and docbook the problem is the same as mentoned
in the bug where python bytecode is installed into the wrong
filesystem location.

Other problems can occur when the python eclasses which handle the
python byte code are not used at all or improperly used with python
applications. For instance when python byte code is not handled by the
eclasses the py{c,o} files are created at runtime when using the
program as root. At deinstallation this files are not removed from the
filesystem because the package manager is not aware of them.

[1] http://bugs.gentoo.org/show_bug.cgi?id=194343
[2] http://www.gentoo.org/proj/en/Python/developersguide.xml

Re: [gentoo-user] Re: new timezone data requires setting a symlink by hand

[Daniel Pielmeier]

2008/7/31, Anno v. Heimburg <[EMAIL PROTECTED]>:
> Huh? I always thought the proper place was the TIMEZONE variable
> in /etc/conf.d/clock:

It was he proper place in pre baselayout-2 and openrc times! Now the
proper place for gentoo is /etc/timezone!

Re: [gentoo-user] Sync'ing and compiling pkgs for multiple PCs

[Daniel Iliev]

On Wed, 30 Jul 2008 20:59:02 -0400
Simon <[EMAIL PROTECTED]> wrote:

> 
> > distcc is, IMO, a bit more elegant than (for instance) trying to 
> > manually emerge binary packages for machine A on PC B. You can tell
> > it to share the work or just unload it to the most powerful
> > machine. There may be concerns about using a binary package if USE
> > flags are different between the two machines, but distcc ensures
> > that the package is built using those defined in make.conf of the
> > machine on which you're running emerge.
> 
> Yes!  I was actually trying distcc today for the first time and got
> it working from the perspective of my fastest computer, I got some
> trouble though (see below).  What you mentioned about running the
> `emerge -uDN world` on each individual machines + sharing built
> packages is absolutely awesome. Best of all worlds if i could say!
> 


Perhaps you have tried it already, but I didn't see it mentioned
in this thread, so: for additional speed up of the compilation at the
expense of disk space you may try ccache.

ccache, gentoo.org: http://tinyurl.com/6o2ql

ccache & distcc, gentoo-wiki.com: http://tinyurl.com/85ngj


-- 
Best regards,
Daniel

Re: [gentoo-user] Numpad keys behaving strange after system-wide update

[Daniel Pielmeier]

Mark David Dumlao schrieb am 02.08.2008 18:16:

Hi, I noticed my numpad keys stopped working after doing a system update.
The thing is I can't pinpoint what exacly I changed since I did an emerge
-uDnav world. I'm sure its some file somewhere...

Pressing numpad keys in X doesnt seem to produce results or nothing at all.
However, presisng numpad keys seems to work fine on the (non xterm) command
line. Did anyone else's numpad appear to break? What do you think I should
look up?



Maybe it is the update to hal-0.5.11-r1.

http://bugs.gentoo.org/show_bug.cgi?id=233184

Re: [gentoo-user] Numpad keys behaving strange after system-wide update

[Daniel Pielmeier]

Roy Wright schrieb am 03.08.2008 08:10:

Mark David Dumlao wrote:

Hi, I noticed my numpad keys stopped working after doing a system update.
The thing is I can't pinpoint what exacly I changed since I did an emerge
-uDnav world. I'm sure its some file somewhere...

Pressing numpad keys in X doesnt seem to produce results or nothing at all.
However, presisng numpad keys seems to work fine on the (non xterm) command
line. Did anyone else's numpad appear to break? What do you think I should
look up?



This is the symptom I see when I try "evdev" driver and am using a MS
Natural Pro keyboard.  Briefly the keyboard is handled as two USB
devices.  The keys that work are on the first device, the ones that
don't on the second device.  Following gentoo-wiki howtos it looks like
you have to hack the kernel.  At that point I simply reverted to using
the "kbd" driver.  Maybe some year evdev will mature...

Quick check, look in your xorg.conf "ServerLayout" section, identify
which keyboard InputDevice, then check to see which driver it is using.

HTH,
Roy



The new hal version converts the InputDevice section for the keyboard
from xorg.conf to a fdi rule file (/etc/hal/fdi/policy/10-x11-input.fdi)
for hal. Even if you use the kbd driver in xorg.conf it is converted to
use evdev in the fdi file. Now when you use >=xorg-server-1.4 and have
the hal use-flag enabled it will use the evdev driver for the keyboard.

However a bug in the hal-0.5.11 ebuild prevents the conversion to the
fdi file. The ebuild states it has done the conversion but actually has
done nothing, and recommends to remove the InputDevice section in
xorg.conf. So there are chances that there is no rule file and no entry
in xorg.conf. Maybe evdev then does some guessing and guesses wrong.

Regards,

Daniel

Re: [gentoo-user] Cleaning out my world file

[Daniel Pielmeier]

Dale schrieb am 04.08.2008 00:18:

Hi,

I was wondering if their is a easy way to clean out the cruft in my 
world file?  Is their a script?  I tried saving a cope then running 
regenworld but that didn't do much.  Maybe it is old and out of date?


The reason why I want to do this?  Well, sometimes I forget to use the 
-1 option when working on getting a dependency to install.  It adds up 
after a while.


Ideas?

Dale

:-)  :-)




emerge udept

dep -w

This will find unneeded entries in the world file.

Re: [gentoo-user] what qt package?

[Daniel Pielmeier]

Matthew R. Lee schrieb am 04.08.2008 00:28:
I'm doing a clean up of my system, getting rid of stuff I never use. Some time 
ago, for reasons I can't remember, I installed a qt development package which 
provided Qt linguist, Qt designer and Qt assistant. I want to unmerge them 
but I can't remember, or figure out which package they're from. I've tried 
variations on the names, I've looked at the gentoo-portage web page 
and 'searched' the database, I've been to the trolltech pages. But I can't 
work it out! I think they all came in one package together. Does anyone know 
what it's called. It will be obvious when you tell me :-)

Cheers
Matt


Linguist, Designer and Assistant are part of x11-libs/qt and I don't 
think you can get rid of them by use flag. You have to unmerge qt but 
this is surely not what you want or can. Only if no package from your 
system depends on qt anymore.

Re: [gentoo-user] what qt package?

[Daniel Pielmeier]

2008/8/4, Matthew R. Lee <[EMAIL PROTECTED]>:
> > Linguist, Designer and Assistant are part of x11-libs/qt and I don't
> > think you can get rid of them by use flag. You have to unmerge qt but
> > this is surely not what you want or can. Only if no package from your
> > system depends on qt anymore.
>
> Is that true? I could of sworn I didn't always have them, and I've always used
> kde. But then again it might explain why I can't find specific packages for
> them.
> Thanks
> Matt
>

There are many tools out there which let you find out to which ebuild
a file belongs or which files belong to a specific ebuild.

gentoolkit:
equery belongs assistant
equery list qt

portage-utils:
qfile assistant
qlist qt

udept:
dep -F assistant
dep -f qt

Re: [gentoo-user] Cleaning out my world file

[Daniel Pielmeier]

2008/8/4, Dale <[EMAIL PROTECTED]>:
>
> There are three versions, all masked.  Could you share which version you
> use?  I ask since it works for you, then it may work for me too.  ;-)
> - app-portage/udept-0.5.99.0.2.95-r1 (masked by: ~x86
> keyword)
> - app-portage/udept-0.5.99.0.2.95 (masked by: ~x86 keyword)
> - app-portage/udept-0.5.96.2 (masked by: ~x86 keyword)
>

I suggest you to use the ebuild from bug 172611 [1]. Unfortunately
there has not been any progress of udept upstream in the last months.
So this ebuild fixes a few issues.

There are other things where udept will not work properly, like it
does not know anything about EAPI's so far, which causes the
dependency and reverse dependency calculation to report false
positives.

dep -w works reliable though. If you are unsure you can always backup
the world file before, but when emerge -avuND world goes through
without any errors I think you are safe.

[1]] http://bugs.gentoo.org/show_bug.cgi?id=172611

Re: [gentoo-user] make oldconfig

[Daniel Pielmeier]

2008/8/4, James <[EMAIL PROTECTED]>:
> Hello,
>
> It seems like I remember that 'make oldconfig' is not
> needed any more, to pass the current (booted) kernel
> option  to the .config for building a new kernel.
>
> Of is 'make oldconfig' still a good idea?
>
>
> James

At least in the kernel Makefile there is no hint about /proc/config.gz
which contains the running kernel configuration, so I think make
oldconfig or your favourite kernel configuration tool is still needed.
If there is no .config or .config.old it will load a default
configuration which is probably not what you want.

Re: [gentoo-user] gentoo-wiki.com - whom to contact?

[Daniel Pielmeier]

2008/8/5, Helmut Jarausch <[EMAIL PROTECTED]>:
> Hi
>
> nearly all links on
> http://gentoo-wiki.com/Index:MAN
> are broken.
>
> Does anybody know someome to inform about that?

http://gentoo-wiki.com/Gentoo_Linux_Wiki:Mailing_Lists
http://gentoo-wiki.com/Gentoo_Linux_Wiki:Irc
http://gentoo-wiki.com/FAQ_Who_are_the_admins

Regards,

Daniel

Re: [gentoo-user] Re: make oldconfig

[Daniel Pielmeier]

2008/8/6, James <[EMAIL PROTECTED]>:
> So what I gleen is that you run on
> a kernel, say version linux-2.6.24-gentoo-r8
>
> You down load newer sources, say version
> linux-2.6.25-gentoo-r7
>
> cd /usr/src
>
> rm linux
>
> ln -sf /usr/src/linux-2.6.25-gentoo-r7 linux
>
> make oldconfig  
> make menuconfig
>
> cp System.map /boot/System.map-2.6.25-gentoo-r7
> cp arch/x86_64/boot/bzImage /boot/kernel-2.6.25-gentoo-r7
> cp .config /boot/config-2.6.25-gentoo-r7
>
>
> edit grub apppropriately
> and reboot to the new kernel?

In the case you run menuconfig oldconfig is not needed. I did so
myself in the past. Now i simply run just oldconfig, because you have
a better control of what has changed between the versions. After
oldconfig menuconfig is not needed one of them will suffer.

Regards,

Daniel

Re: [gentoo-user] gcc not downward compatible?

[Daniel Pielmeier]

2008/8/11, Peter Wood <[EMAIL PROTECTED]>:
> I don't really mind using the ~x86 version of these packages, but wonder why
> the old, stable packages won't compile with a newer gcc. Isn't gcc supposed
> to be downward compatible?
>

No, take a look here:
http://bugs.gentoo.org/show_bug.cgi?id=198121

Re: [gentoo-user] Re: unicode losting

[Daniel Pielmeier]

Rev. Ferris schrieb am 14.08.2008 19:26:

#cat /etc/locale.gen
it_IT UTF-8
de_DE ISO-8859-1
[EMAIL PROTECTED] ISO-8859-15
de_DE UTF-8
it_IT ISO-8859-1
[EMAIL PROTECTED] ISO-8859-15
pl_PL ISO-8859-1
[EMAIL PROTECTED] ISO-8859-15
pl_PL UTF-8


try this:

de_DE.UTF-8 UTF-8
de_DE ISO-8859-1
[EMAIL PROTECTED] ISO-8859-15

it_IT.UTF-8 UTF-8
it_IT ISO-8859-1
[EMAIL PROTECTED] ISO-8859-15

pl_PL.UTF-8 UTF-8
pl_PL ISO-8859-1
[EMAIL PROTECTED] ISO-8859-15

Regards,

Daniel

Re: [gentoo-user] Cleaning out my world file

[Daniel Pielmeier]

Dale schrieb am 17.08.2008 05:06:

Dale wrote:
 
Will report back later. Dale


:-)  :-)




OK.  I ran it but not real sure what the output is.  Text file 
attached.  Looks like a error or two and not sure if that matters or not.


Info:

portage-utils-0.1.29
portage-2.2_rc8

Thanks

Dale

:-)  :-)


Try the attached file, putting the script directly into the mail 
probably broke it!


Regards,

Daniel
#!/bin/bash

# Print out orphan files in specified directories

find /etc -xdev -type f -print | xargs qfile -o
find /usr -xdev \( -path /usr/src -prune \) -o -type f -not -name '*.pyc' -not 
-name '*.pyo' \
-not -name .keep  -print | xargs qfile -o
find /lib -xdev \( -path /lib/modules -prune \) -o -type f | xargs qfile -o

Re: [gentoo-user] Cleaning out my world file

[Daniel Pielmeier]

Dale schrieb am 17.08.2008 05:35:
 > OK.  Did that.  No errors.  It is a HUGE list of files.  Here is a
SMALL

snippet:


snipping the snippet :-)



They all look about the same as those listed above.  What would it look 
like if one was orphaned?

Dale

:-)  :-)



Detecting real orphan files is not a trivial task, it requires some
knowledge about the installed packages, packages which have been
installed before and which are now removed. Files which are essential to
the system and which are not part of any packages (/etc/make.conf,
/etc/passwd). There are also files which are generated by portage but
can not be tracked by portage.

The script already tries to exclude some false positive like *.pyc,
*.pyo, .keep files. The pyc and pyo files are bite compiled python
modules which are generated after the files are installed to the file
system and thus are not known by portage. .keep files are normally
generated by portage to prevent empty directories from being removed.

It also excludes the directories /usr/src (there are kernel and other
sources stored) and /lib/modules (location of the kernel modules) both
locations contain files not known by portage too. But there are many
other files not excluded by the script which are essential to the
system. So you should know what you do when deleting any of the files
you get from the script.

Regards,

Daniel

Re: [gentoo-user] Re: ffmpeg (update)

[Daniel Pielmeier]

2008/8/21, Peter Ruskin <[EMAIL PROTECTED]>:
>
> Instead of running emerge --depclean, better IMHO is:
> emerge --update --newuse --deep --oneshot @system @world -vat,
> followed by
> emerge -C arts and emerge @preserved-rebuild if necessary.
>

But you can not expect that everybody is running unstable portage so
this commands are not available for those running stable.

Re: [gentoo-user] Re: ffmpeg (update)

[Daniel Pielmeier]

Neil Bothwick schrieb am 21.08.2008 18:02:

On Thu, 21 Aug 2008 17:27:08 +0200, Daniel Pielmeier wrote:


Instead of running emerge --depclean, better IMHO is:
emerge --update --newuse --deep --oneshot @system @world -vat,
followed by
emerge -C arts and emerge @preserved-rebuild if necessary.
 

But you can not expect that everybody is running unstable portage so
this commands are not available for those running stable.


True, but you should run emerge -uavDN world at least.




I just want to state this particular commands are not available if using 
unstable portage so I would recommend this for stable after cleaning 
make.conf, package.use and world:


emerge --update --newuse --deep --oneshot world
emerge -p --depclean
revdep-rebuild

Sorry if this caused some confusion!

Regards,

Daniel

Re: [gentoo-user] Re: ffmpeg (update)

[Daniel Pielmeier]

2008/8/22, James <[EMAIL PROTECTED]>:
> At which version of portage do these feature/options become available?
>
> I think I'm going to upgrade to one of the unstable (~) versions of portage
> to test out these features
>
> Since there are several choices:
>
> ~2.1.5.6 ~2.2_rc1 ~2.2_rc5 ~2.2_rc6
>
> does it matter which one I use? Use the latest (~2.2_rc6) ?
>

Then you should sync your portage tree. The only version supporting
the mentioned features which is currently in the tree is
portage-2.2_rc8.

Regards,

Daniel

Re: [gentoo-user] Debugging X

[Daniel Beecham]

Dale wrote:

forgottenwizard wrote:

I'm having a problem getting X to work. It is seg faulting on me, and
despite countless revdep-rebuilds and emerge -e world, it still doesn't
work.

It dies after the cursor shows up, spitting this backtrace and output.
Sorry if the formatting sucks. The last line is probably refering to the
fact I tried to run it from within screen, so if that could cause a
problem say so, and tell me how the heck to get a log of this output
(since startx > log.txt doesn't work)

#--- startx output ---#

X Window System Version 1.3.0
Release Date: 19 April 2007
X Protocol Version 11, Revision 0, Release 1.3
Build Operating System: UNKNOWN
Current Operating System: Linux localhost 2.6.25-gentoo-r7 #1 SMP
PREEMPT Fri Aug 1 21:56:38 CDT 2008 x86_64
Build Date: 22 July 2008
Before reporting problems, check http://wiki.x.org
to make sure that you have the latest version.
Module Loader present
Markers: (--) probed, (**) from config file, (==)
default setting,
(++) from command line, (!!) notice, (II)
informational,
(WW) warning, (EE) error, (NI) not
implemented, (??) unknown.
(==) Log file: "/var/log/Xorg.0.log",
Time: Wed Aug 20 00:11:37 2008
(==) Using config file:
"/etc/X11/xorg.conf"
(WW) NVIDIA: No matching Device section
for instance (BusID PCI:0:1:3) found
(II) Module already built-in
The XKEYBOARD keymap compiler (xkbcomp)
reports:
> Warning:  Multiple names for
> keycode 211
>   Using ,
>   ignoring 
Errors from xkbcomp are not fatal to the
X server
Backtrace:
0: X(xf86SigHandler+0x6d) [0x49690d]
1: /lib/libc.so.6 [0x7fae2c0a4430]
2: X(NumMotionEvents+0x12) [0x447822]
3: X(CreateConnectionBlock+0x53) [0x439623]
4: X(main+0x658) [0x43a168]
5: /lib/libc.so.6(__libc_start_main+0xf4) [0x7fae2c091b74]
6: X(FontFileCompleteXLFD+0x229) [0x439259]

Fatal server error:
Caught signal 11.  Server aborting

waiting for X server to begin accepting connections
giving up.
xinit:  Connection reset by peer (errno 104):  unable to connect to X 
server

xinit:  No such process (errno 3): Server error.
Couldnt get a file descriptor referring to the console
   
#--- end ---#


I've brought this to #x (or xorg, whichever the X support channel in
freenode is), #linux, #gentoo, and the forums. I'm at a bit of a loss as
to what the problem is, or how to go about trying to find out what is
the problem.

  


If I read this correctly, it appears that it can not find the keyboard 
or something.  This is what makes me think that:  "The XKEYBOARD 
keymap compiler (xkbcomp) reports: > Warning: Multiple names for > 
keycode 211".


In your make.conf, do you have a line that is something like this:

INPUT_DEVICES="keyboard mouse"

Also make sure you have something like the following in your xorg.conf 
file:


Section "InputDevice"
   Identifier "Keyboard0"
   Driver "kbd"
EndSection

Section "InputDevice"
   Identifier "Mouse0"
   Driver "mouse"
   Option "Protocol" "auto"
   Option "Device" "/dev/input/mouse0"
   Option "ZAxisMapping" "4 5 6 7"
EndSection

This may not have anything to do with the problem but it is something 
that didn't look right to me.


Dale

:-)  :-)


Nah, that's not the problem, aas you can see here; "Errors from xkbcomp 
are not fatal to the X server"


/ Daniel.

Re: [gentoo-user] emerge kde-meta troubles

[Daniel Pielmeier]

2008/8/28, Dale <[EMAIL PROTECTED]>:
>
> Also, what this heck is this about?  MailScanner has detected a possible
> fraud attempt from "mirror.ovh.net" claiming to be  That shouldn't be in
> your make.conf file.
>

Just take a look in the signature of thie message. :-)

"This message has been scanned for viruses and dangerous content by
MailScanner, and is believed to be clean."

Don't know what's the scanners problem with the mirror adress though.

Regards,

Daniel

Re: [gentoo-user] emerge kde-meta troubles

[Daniel Pielmeier]

2008/8/28, Suman Chakrabarty <[EMAIL PROTECTED]>:
> Now my question is: why? "man emerge" says that the -1 option is
> equivalent to "--oneshot" which is "Emerge as normal, but do not add the
> packages to the world file for later updating." But how does it help
> emerge something which was not emerging without -1?! May be I am missing
> something obvious, but please explain.

I guess Dirk suggested to use "--oneshot" because you probably don't
want kdelibs in world. The main issue was that you had to re-emerge
kdelibs to be built against the newer qt version. So "emerge kdelibs"
or "emerge --nodeps kdelibs" is enough but then kdelibs is added to
your world file.

Regards,

Daniel

Re: [gentoo-user] emerge kde-meta troubles

[Daniel Pielmeier]

2008/8/28, Suman Chakrabarty <[EMAIL PROTECTED]>:
>
> Thanks, but I don't think my confusion was addressed fully. Let me explain.
> The following command did not work as reported (and suggested) before:
>
>   ~ # emerge -D --newuse kdelibs
> Calculating dependencies... done!
> >>> Auto-cleaning packages...
>
> >>> No outdated packages were found on your system.
>
>
> But, "emerge -1v kdelibs" worked as suggested. I don't understand why it
> works with -1 option added, but not without. Even if I had included kdelibs
> in the world file, it should have been re-emerged through the previous
> command, right? I didn't see through the additional magic done by -1!
>

As I said -1 is not the problem here. The interesting and different
parts of this two commands are:

emerge kdelibs

This just re-emerges kdelibs.

emerge -D --newuse kdelibs or
emerge --deep --newuse kdelibs

Here all dependencies and changed use flags of all dependencies
including kdelibs will be re-emerged. The dependencies are updated
before kdelibs and thus in this case kdescreenseaver fails as it needs
kdelibs re-emerged first.

Regards,

Daniel

Re: [gentoo-user] virtualbox-ose build failed

[Daniel Pielmeier]

pat schrieb am 31.08.2008 13:35:

Hello,

I'm trying to build virtualbox-ose, but it fails :-( I've googled but without
luck.

Error message is:

Checking for Qt4 uic3: 
  ** not found!
 * 
 * ERROR: app-emulation/virtualbox-ose-1.6.4 failed.

 * Call stack:
 *   ebuild.sh, line   49:  Called src_compile
 * environment, line 2756:  Called die
 * The specific snippet of code:
 *   ./configure --with-gcc="$(tc-getCC)" --with-g++="$(tc-getCXX)"
${myconf} || die "configure failed";
 *  The die message:
 *   configure failed
 * 
 * If you need support, post the topmost build error, and the call stack if

relevant.
 * A complete build log is located at
'/var/tmp/portage/app-emulation/virtualbox-ose-1.6.4/temp/build.log'.
 * The ebuild environment file is located at
'/var/tmp/portage/app-emulation/virtualbox-ose-1.6.4/temp/environment'.
 * 


Any suggestions?

Thanks

 Pat




Which qt version do you use? My guess is 4.4.1.
Try building it with with qt3support USE flag enabled.

Regards,

Daniel

Re: [gentoo-user] virtualbox-ose build failed

[Daniel Pielmeier]

pat schrieb am 31.08.2008 14:03:

make.conf
~
USE="a52 accessibility -apm acpi alsa -arts audiofile browserplugin dbus
divx4linux -directfb -doc dts dvb dvd dvdr dvdread -eds -emacs -emboss flac
gnome -kde kerberos mad mime mmx mozilla nptl nptlonly nsplugin opengl posix
-qt -qt3 -qt3support -qt4 -qt-static -samba slang sse -svga unicode usb v4l
v4l2 videos win32codecs X xinerama -xmms xsl xvid"

package.use
~~~
app-emulation/virtualbox-ose additions alsa pulseaudio qt qt3 qt3support qt4

emerge --pretend qt
~~~
[ebuild   R   ] x11-libs/qt-4.3.3

What I have to change or emerge?


Do you have /usr/bin/uic3 installed?

Re: [gentoo-user] virtualbox-ose build failed

[Daniel Pielmeier]

pat schrieb am 31.08.2008 14:21:

On Sun, 31 Aug 2008 14:18:56 +0200, Daniel Pielmeier wrote

pat schrieb am 31.08.2008 14:03:

make.conf
~
USE="a52 accessibility -apm acpi alsa -arts audiofile browserplugin dbus
divx4linux -directfb -doc dts dvb dvd dvdr dvdread -eds -emacs -emboss flac
gnome -kde kerberos mad mime mmx mozilla nptl nptlonly nsplugin opengl posix
-qt -qt3 -qt3support -qt4 -qt-static -samba slang sse -svga unicode usb v4l
v4l2 videos win32codecs X xinerama -xmms xsl xvid"

package.use
~~~
app-emulation/virtualbox-ose additions alsa pulseaudio qt qt3 qt3support qt4

emerge --pretend qt
~~~
[ebuild   R   ] x11-libs/qt-4.3.3

What I have to change or emerge?

Do you have /usr/bin/uic3 installed?


No. Which package contains it?





x11-libs/qt-qt3support-4.4.1

But as you can see I am on qt-4.4.1 already where the qt package has 
been split up. In qt-4.3.3 it should be part of x11-libs/qt when 
qt3support is enabled. But you should verify if uic3 is in a different 
location for 4.3.3.


What gives "emerge -pv qt" and "emerge -pvNDt virtualbox-ose"

Regards,

Daniel

Re: [gentoo-user] virtualbox-ose build failed

[Daniel Pielmeier]

pat schrieb am 31.08.2008 14:49:


Well, I've enabled qt3support in make.conf and the my Qt (4.3.3) library
contains the uic3 source, so I'll see if the Qt rebuild helps :-)

I'll inform if it helped or not.



Normally it should be enough if it is enabled in package.use for 
x11-libs/qt. You already have this setting so I thought there are 
packages which need rebuilding. Maybe it is a good idea to rebuild qt 
and also give revdep-rebuild a go.


Regards,

Daniel

Re: [gentoo-user] Weird df listing

[Daniel Pielmeier]

Michael Sullivan schrieb am 02.09.2008 19:05:

On Tue, 2008-09-02 at 15:39 +0200, Volker Armin Hemmann wrote:

On Dienstag, 2. September 2008, Michael Sullivan wrote:

Can someone please explain to me what's going on with /dev/sda6?  I
couldn't log into GNOME after my reboot yesterday, and when I asked for
a df listing in the console, I got this.  Shouldn't there be 4GB
available?

camille ~ # df -h
FilesystemSize  Used Avail Use% Mounted on
/dev/sda6  78G   74G  0G  100% /
udev   10M  184K  9.9M   2% /dev
/dev/sda7  52G   40G   12G  78% /mnt/store
shm   247M 0  247M   0% /dev/shm
catherine:/backup  44G   34G  8.5G  80% /backup/catherine

you have space left, but the inodes are all used up.

Typical problem for fs like extX.




What fs should I use instead?  For future reference what's the current
standard?





First I would verify if the inodes on that particular partition are 
really all used by "df -i". I don't think this is an inodes problem as 
the partition is quite big and ext3 tends to create to less inodes only 
on small partitions. 4Gigabyte of a 78 Gigabyte partition sounds like 
the default 5 percent reserved by root on an ext3 filesystem.


Regards,

Daniel

Re: [gentoo-user] Dealing with MS Access files (.mdb)

[Daniel Pielmeier]

[EMAIL PROTECTED] schrieb am 04.09.2008 00:41:

I have a file which may or may not have useful info in it (I am trying
to translate US/Canada NPANXX phone numbers into latitude/longitude or
at least city without buying a commercial db).  I know almost nothing
of Access otehr than it being some kind of bastardized database.  I
guess it runs some form of SQL but I don't even know that.

Does anyone know of a program to convert this .mdb file into something
useful, other than access to Access on a Windows machine?  If it truly
is an SQL database, I suppose it would be some kind of dump complete
with table creation , foreign keys, etc.  But I can write a script to
deal with that.




app-office/mdbtools http://mdbtools.sourceforge.net/ ??

Regards,

Daniel

Re: [gentoo-user] Automatic network configuration for USB connected device

[Daniel Pielmeier]

Hong Hao schrieb am 06.09.2008 15:29:

Dan Johansson wrote:

I have bought me a new toy - an Openmoko Freerunner.
Each time I connect it to my Gentoo-box I must configure the usb0 
device on the Gentoo-box as follows:
  ifconfig usb0 192.168.0.200 netmask 255.255.255.248 After that I can 
without problem ssh to the Freerunner. Now I want the above to be 
automatically done when I connect the Freerunner. I have found the the 
following description for Ubunto:


  Add the following lines to your /etc/network/interfaces :
  allow-hotplug usb0
  auto usb0
  iface usb0 inet static
  address 192.168.0.200
  netmask 255.255.255.0
  network 192.168.0.0
  up iptables -A POSTROUTING -t nat -j MASQUERADE -s 192.168.0.0/24 &
  up echo 1 > /proc/sys/net/ipv4/ip_forward &
  up iptables -P FORWARD ACCEPT &
  down iptables -D POSTROUTING -t nat -j MASQUERADE -s 192.168.0.0/24

What would be the corresponding file(s) on Gentoo?
Regards,
  

/etc/conf.d/network




I would try this: /etc/conf.d/net :-)

Re: [gentoo-user] user & group

[Daniel Pielmeier]

2008/9/12 Hinko Kocevar <[EMAIL PROTECTED]>:
> Hi,
>
> When I create a new user on my machine the user belongs to its own group - 
> and not the group 'users'. Also when creating directory the user creates it 
> with its group name instead of users group. How can this be fixed, so that 
> user belongs to users group by default and hence all the files it created 
> belong to users group as well?

You may take a look at the useradd man page especially the -g and the -n option.

Regards,

Daniel

Re: [gentoo-user] Network configuration: looking up URLs is very slow; how can I fix this?

[Daniel Beecham]

On Fri, 2008-09-12 at 18:47 +, Alan Mackenzie wrote:
> Hi, Gentoo!
> 
> I bought a new hard drive on Wednesday, and am seriously getting Gentoo
> installed (after doing a trial installation in July).
> 
> However  looking up URL's is very, very slow.  This is most
> noticeable when running emerge.  It is very also noticeable running
> Firefox; the looking up is _much_ slower than on my existing Debian
> sarge system.
> 
> Presumably, I need to configure some sort of DNS cache, or proxy, or
> whatever it might be called.  I've looked in
> , but couldn't find a network configuration
> manual there.
> 
> Would somebody give me a pointer, please?  Thanks!
> 

Where are your DNS-server, and how does your routes look like?
I'd like to see your /etc/resolv.conf aswell. :-)

Re: [gentoo-user] Network configuration: looking up URLs is very slow; how can I fix this?

[Daniel Beecham]

As long as speedport.ip is 192.168.2.1, the routes are okay, and the
resolv-file is okay aswell.

It's most likely either DNS-cacheer, i.e. your own router, or the
DNS-server your ISP is giving you beeing the bottleneck.

You could go ahead and dig(1) both those servers and perhaps time(1)
that. If you can find any logs in your router that would help aswell.

--
Daniel.

On Fri, 2008-09-12 at 20:59 +, Alan Mackenzie wrote:
> Hi, Daniel,
> 
> On Fri, Sep 12, 2008 at 08:50:13PM +0200, Daniel Beecham wrote:
> 
> > > However  looking up URL's is very, very slow.  This is most
> > > noticeable when running emerge.  It is very also noticeable running
> > > Firefox; the looking up is _much_ slower than on my existing Debian
> > > sarge system.
> 
> > > Presumably, I need to configure some sort of DNS cache, or proxy, or
> > > whatever it might be called.  I've looked in
> > > <http://www.gentoo.org/doc/>, but couldn't find a network
> > > configuration manual there.
> 
> > > Would somebody give me a pointer, please?  Thanks!
> 
> > Where are your DNS-server, and how does your routes look like?
> > I'd like to see your /etc/resolv.conf aswell. :-)
>  
> /etc/resolv.conf:
> #
> # Generated by dhcpcd for interface eth0
> search Speedport_W_700V
> nameserver 192.168.2.1
> #
> [translation: A router/DSL modem (called "Speedport") at local address
> 192.168.2.1.]
> 
> I do # route (as root), and get this:
> #
> Kernel IP routing table
> Destination Gateway Genmask Flags Metric RefUse Iface
> 192.168.2.0 *   255.255.255.0   U 0  00 eth0
> link-local  *   255.255.0.0 U 0  00 eth0
> loopback*   255.0.0.0   U 0  00 lo
> default speedport.ip0.0.0.0 UG0  00 eth0
> #
> 
> Er, where is my DNS-server?  That's the entry in resolv.conf, isn't it,
> i.e. the router at 192.168.2.1?
> 

Re: [gentoo-user] Network configuration: looking up URLs is very slow; how can I fix this?

[Daniel Beecham]

On Sat, 2008-09-13 at 14:59 +0100, Stroller wrote:
> On 12 Sep 2008, at 21:59, Alan Mackenzie wrote:
> > ...
> > I do # route (as root), and get this:
> > ## 
> > ###
> > Kernel IP routing table
> > Destination Gateway Genmask Flags Metric Ref 
> > Use Iface
> > 192.168.2.0 *   255.255.255.0   U 0   
> > 00 eth0
> > link-local  *   255.255.0.0 U 0   
> > 00 eth0
> > loopback*   255.0.0.0   U 0   
> > 00 lo
> > default speedport.ip0.0.0.0 UG0   
> > 00 eth0
> > ## 
> > ###
> 
> I'd also specify 192.168.2.1 as the default route explicitly, rather  
> than by name.
> 
> Stroller.
> 

I think that route rDNS'es the host, and is not set.
I could be wrong though.

--
Daniel.

Re: [gentoo-user] SSH fixed; now su doesn't work

[Daniel Beecham]

On Sat, 2008-09-13 at 23:36 -0500, Michael Sullivan wrote:
> On Sat, 2008-09-13 at 23:27 -0500, Michael Sullivan wrote:
> > I can ssh over to the old box now, but my su command doesn't work:
> > 
> > 
> > [EMAIL PROTECTED] ~ $ su -
> > su: Authentication failure
> > 
> > >From /var/log/messages:
> > 
> > Sep 13 23:23:07 bullet sshd[24134]: pam_unix(sshd:session): session
> > opened for user michael by (uid=0)
> > Sep 13 23:23:10 bullet su[24142]: FAILED su for root by michael
> > Sep 13 23:23:10 bullet su[24142]: - pts/1 michael:root
> > 
> > At first, I thought I might not be in the wheel group, so I checked:
> > 
> > bullet log # grep wheel /etc/group
> > wheel:x:10:root,michael,amy
> > 
> > What have I done wrong this time?  (I know somebody's going to gripe at
> > me about using the root account this way; don't bother.  There are far
> > too many tasks I do every day that require root privileges to configure
> > them all for sudo...)
> > 
> > 
> 
> Please disregard what I said about sudo.  I was misinformed.  I was told
> that every command I wanted to use as a regular user that required root
> privileges had to be listed in /etc/sudoers.  Now I see that that is not
> true.  I apologize for my ignorance...
> 
> 

Or you could do sudo -i or sudo -s...
But that's not a good solution, up to fix su!
(Although, I can't help you. The info you gave us is not much, but I
can't help getting any more...)
:-)

--
Daniel.

Re: [gentoo-user] Emerge takes a long time calculating dependencies...

[Daniel Pielmeier]

2008/9/18 BRM <[EMAIL PROTECTED]>:
> And syncing portage is done nightly using a cron job to do "emerge --sync"; 
> however, I run "emerge world -vuDNp" manually.

How often does this cron job run and what are the contents of your
cron script. Do you use "emerge --sync" or probably "eix-sync". I am
asking this because if you sync to often and you use eix-sync the
portage cache is deleted in advantage of the sync and is not recreated
if it fails because of identical timestamps between the local machine
and the rsync server.

So when calling emerge after this the cache has to be created before
and thus it seems dependency calculation takes very long.

> (The system also hosts portage via rsync for my other gentoo system on my 
> network too, which I manually sync against it; I might be adding another 
> system too in the same manner, not sure yet.)
>
> Ben

Re: [gentoo-user] Emerge takes a long time calculating dependencies...

[Daniel Pielmeier]

2008/9/18 BRM <[EMAIL PROTECTED]>:
> Any ideas on what might be making it go so slow?
> I've thought about doing some clean-ups with e-merge (e.g. emerge 
> --deep-clean), but haven't risked it yet.

Did you mean "emerge --depclean"?

> In all other respects, the systems seems to be performing fine. It runs 24-7, 
> hosts my internet connection and network services for my other systems.

How do you sync your portage tree with this machine? How often? Do you
use some kind of script?

Regards,

Daniel

Re: [gentoo-user] Emerge takes a long time calculating dependencies...

[Daniel Pielmeier]

2008/9/18 Ricardo Saffi Marques <[EMAIL PROTECTED]>:
>
> In my case, my cron runs "eix-sync" every night at 00h and I just did an 
> "time emerge -DNuvp world"
> here, getting:
>
> Total: 129 packages (103 upgrades, 18 new, 8 reinstalls), Size of downloads: 
> 187,761 kB
>
> real0m21.074s
> user0m18.913s
> sys 0m0.724s
>
> So I assume once a day (let's say, every night) is a good choice for syncing, 
> right? :-)
> Never ever had a problem with slow dependency calculation on emerge.

As far as I know the rsync servers are syncing against the master
server(s) about every half an hour which obviously also updates the
timestamp. So what I described above only happens if you sync within a
shorter period. Running eix-sync every day shouldn't be a problem.

Syncing twice an hour is also considered bad practise because of the
heavy traffic involved. It just happens to me sometimes if I test
stuff and had to sync within a shorter period.

I just asked this because it could be possible that the script calls
eix-sync twice or after an emerge --sync which is not necessary or for
some other reason it is called twice within a too short period.

Regarding the fragmentation. I use ext3 and reiserfs for the portage
tree and did not recognize a significant slowdown. There is a slowdown
but not from 4 to 20 minutes. At least in my case.

Regards,

Daniel

Re: [gentoo-user] Emerge takes a long time calculating dependencies...

[Daniel Pielmeier]

2008/9/18 BRM <[EMAIL PROTECTED]>:

> And I am only referring to the time specified in the "calculating 
> dependencies" stage when one first runs "emerge world -vuDNp".

Is the second run significantly faster, if yes chances are high that
there is a problem with the portage cache.

Re: [gentoo-user] Sending RSS feeds via email

[Daniel Pielmeier]

2008/9/19 Justin <[EMAIL PROTECTED]>:
>>
>> Looks like what I am looking for.. Thank you!
>>
>>
>
> I just saw that the latest version is 2.63. Perhaps you need to bump the
> ebuild.
>

Just bumping the ebuild is not enough if you read the bug report carefully.
The 2.55 version ebuild uses a custom tarball created by the reporter
of the bug.
So you need to create a new tarball, plus you probably need to revisit
the gentoo.patch which is applied.

Regards,

Daniel