Re: [gentoo-user] file .ogg
[EMAIL PROTECTED] schreef: Hi everybody. I think gentoo is driving me crazy. Since I found out the I can listen .wav file but not .ogg file, to-day I installed libvorbis, mpeg123 and vorbis-tools. But the result is the same . . . I cannot listen .ogg file. Hi, Emilio, What program are you trying to listen to said *.ogg files with? Was that program emerged with OGG/Vorbis support? An emerge -pv package will tell you what USE flags are available to the application you're trying to play the files with, and what support was compiled into it. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gcc upgrade, should it upgrade?
Teresa and Dale schreef Nagatoro wrote: Why don't upgrade? As far as I know upgrading gcc isn't a big deal. It was just the 3.3.x - 3.4.x that was due to that the api for c++ had changed. Oh, I thought it was a big deal. That's why I was wanting to wait. Funny thing is, it don't want to upgrade now so maybe it had a bug and they set it back again or something, or I just missed it and upgraded anyway. ;-) (other) Funny thing is, last I heard, you were planning to mask the upgrade versions of GCC. If you did that, of /course/ you are no longer offered upgrades, since that's the point of masking (to mark a package as unavailable to be installed on this computer). gcc-3.4.5-r1 is the most recent stable; current unstable (~x86) is 3.4.6, 4.0 is masked (hard-masked), so you wouldn't see it anyway. So I'm guessing you are running stable only, and masked the most recent stable version explicitly (3.4.5-r1)? If you masked only that version in /etc/portage/package.mask, like so =sys-devel/gcc-3.4.5-r1 you won't see an offer to update until 3.4.6 goes stable; if you masked all versions above your current version, as in | =sys-devel/gcc-3.4.5-r1 you won't see an offer to update ever, until you adjust the mask. Although when 4.0 makes it into the tree, it might use a different slot, so that might make you an offer. But as far as I know, 3.4.5-r1 is still alive and kicking in the tree. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] New To Gentoo and Emerge, No ACPI in Kernel
Lord Sauron schreef: Okay, here's where I've isolated the problem to. snip . I know that it correctly compiles the kernel. I put a new name for the new kernel (test1) to try and ID it snip # make install Sticks it into /boot. /boot now reads System.mapconfig.old System.map-2.6.15-gentoo-r1 grub System.map-2.6.15-gentoo-r1.old initramfs-genkernel-x86-2.6.15-gentoo-r5 System.map-2.6.15-gentoo-r1test1 kernel-genkernel-x86-2.6.15-gentoo-r5 System.map-2.6.15-gentoo-r1test1.old lost+found System.map.oldvmlinuz boot vmlinuz-2.6.15-gentoo-r1 configvmlinuz-2.6.15-gentoo-r1.old config-2.6.15-gentoo-r1 v mlinuz-2.6.15-gentoo-r1test1 config-2.6.15-gentoo-r1.old vmlinuz-2.6.15-gentoo-r1test1.old config-2.6.15-gentoo-r1test1 vmlinuz.old config-2.6.15-gentoo-r1test1.old Not terribly exciting. However, I went to /boot/grub/menu.lst and it reads as such: localhost boot # cat ./grub/menu.lst default 0 timeout 7 splashimage=(hd0,0)/grub/splash.xpm.gz title=Gentoo Linux root (hd0,0) kernel /kernel-genkernel-x86-2.6.15-gentoo-r5 root=/dev/ram0 init=/linuxrc ramdisk=8192 real_root=/dev/hda3 initrd /initramfs-genkernel-x86-2.6.15-gentoo-r5 The most concerning part is the last three lines. For any kernel, it appears to demand the kernel itself. If you'll refer back to # ls /boot then you'll notice that kernel-2.6.15-gentoo-r1test1 isn't there. Yes it is: vmlinuz-2.6.15-gentoo-r1test1 Nor is the initrd. I don't know where they might be, or if they're not there then how to generate them. No, the initrd isn't there; you apparently made your previous kernel with genkernel, which creates an initrd, and you further do not have splashutils installed (which would/can also make an initrd to contain the bootsplash images). Manual kernel installation using 'make install' copies the bzImage (the kernel itself) to /boot/, and names it vmlinuz-kernel.version-extra_version_if_used. Make install also copies the helpful but not strictly necessary .config and system.map files to /boot/ adding the version to the end to distinguish it from other supplemental files for other kernels, and creates two symlinks to the current and newly-installed kernel: vmlinuz -- links to the newly installed kernel vmlinuz.old -- links to the current kernel that the newly-installed kernel is replacing. Similar symlinks are also created for the .config and system.map files for the respective kernels. What this means is that you can just tell grub that the first item on the list should load 'vmlinuz' (which is going to be a link to the most recently installed kernel), and the second entry should load vmlinuz.old (which is a link to the previous kernel to the most newly installed, in case of problems). If you have even more old kernels, they can always be listed by the full kernel version. In any case, it seems to me that you basically need to create a new entry for the test kernel; just copy the current entry, paste it above the other one, then delete the following relevant portions: 1. Change 'kernel /kernel-genkernel-x86-2.6.15-gentoo-r5 root=/dev/ram0' to read kernel /vmlinuz root=/dev/ram0 (not sure about the 'root= part; /dev/ram0 seems a bit weird to me but perhaps this is in some way functional for your particular setup. I have no experience with genkernel, which this setting looks like to me, but maybe somebody can confirm that. If it is from genkernel, root= should be the partition of the root filesystem, on my system, this setting is root=/dev/hda5) 2. delete everything else, apparently. I can see that most of the entry is generated by/related to genkernel. Here, for reference is my grub entry for my manually compiled kernels (never used genkernel, as I said): # grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You have a /boot partition. This means that # all kernel and initrd paths are relative to /boot/, eg. # root (hd0,1) # kernel /vmlinuz-version ro root=/dev/hda5 # initrd /initrd-version.img #boot=/dev/hda default=0 timeout=10 splashimage=(hd0,1)/grub/grub-livecd2.xpm.gz title Gentoo_current (2.6.15-gentoo-r7mga) root (hd0,1) kernel /vmlinuz ro quiet root=/dev/hda5 video=matroxfb:vesa:0x11B,depth:32 splash=silent,theme:livecd-2005.1 CONSOLE=/dev/tty1 title Gentoo_prev (2.6.15-gentoo-r7) root (hd0,1) kernel /vmlinuz.old ro root=/dev/hda5 video=matroxfb:vesa:0x11B:ywrap,pmipal,mtrr:3,[EMAIL PROTECTED] splash=verbose,theme:emergence quiet CONSOLE=/dev/tty1 title Failsafe_current root (hd0,1) kernel /vmlinuz ro root=/dev/hda5 video=vesafb:ywrap,pmipal,mtrr:3,[EMAIL PROTECTED] emergency If you ignore all the video= settings, as you should, since they are
Re: [gentoo-user] GorillaTrades Tutorial
Mark Knecht schreef: Hi, I wonder if anyone else has this problem. The tutorial at this page starts and I get audio, but I don't get any video or slides. Config problem here or Linux multimedia issue? http://www.gorillatrades.com/tutorial/# Thanks, Mark Like JimD, I also was able to view as much of the tutorial as I could stand (the narrator's voice was pretty annoying to me) without problems. Unlike JimD, though, I do have gentoo-installed flash: Shockwave Flash Bestandsnaam: libflashplayer.so Shockwave Flash 7.0 r63 MIME-type BeschrijvingAchtervoegsels Actief application/x-shockwave-flash Shockwave Flash swf Ja application/futuresplashFutureSplash Player spl Ja eix netscape-flash * net-www/netscape-flash Available versions: 6.0.79 6.0.81 7.0.25 7.0.61 7.0.63 Installed: 7.0.63 Homepage:http://www.macromedia.com/ Description: Macromedia Shockwave Flash Player eix mozilla-firefox * www-client/mozilla-firefox Available versions: 1.0.7-r4 ~1.5-r9 ~1.5-r11 ~1.5.0.1-r2 ~1.5.0.1-r3 Installed: 1.5.0.1-r3 Homepage:http://www.mozilla.org/projects/firefox/ Description: Firefox Web Browser So, sorry, don't know what to tell you, Mark, but hopefully this information is in some way useful. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: mpg123: Can't open /dev/dsp!
Walter Dnes schreef:
On Tue, Mar 21, 2006 at 08:41:44PM +0100, Simon Kellett wrote
Walter Dnes [EMAIL PROTECTED] writes:
... mpg123 complains about not being able to open /dev/dsp ...
Does mpg123 -a /dev/sound/dsp work ?
Nope. The only change is that now I get Can't open
/dev/sound/dsp!.
Well, my understanding (which may not be correct, but I /think/ it is)
is that /dev/sound/dsp is the sound sequencer created by OSS or ALSA OSS
emulation-- if OSS is not enabled somehow, you won't have it, and if the
app uses ALSA directly (and not OSS emulation), you won't use it (so you
won't notice it's not there).
The ebuilds seem to suggest that mpg123 uses OSS (therefore ALSA OSS
emulation), whereas mpg321 uses ALSA natively:
# $Header:
/var/cvsroot/gentoo-x86/media-sound/mpg123/mpg123-0.59s-r10.ebuild,v
1.4 2006/03/08 15:46:03 flameeyes Exp $
inherit eutils
PATCH_VER=1.5
S=${WORKDIR}/${PN}
DESCRIPTION=Real Time mp3 player
HOMEPAGE=http://www.mpg123.de/;
SRC_URI=http://www.mpg123.de/mpg123/${PN}-pre${PV}.tar.gz
mirror://gentoo/${P}-gentoo-${PATCH_VER}.tar.bz2
LICENSE=as-is
SLOT=0
KEYWORDS=~alpha ~amd64 ~hppa ~ia64 ~mips ~ppc ~ppc-macos ~ppc64 ~sparc
~x86
IUSE=mmx 3dnow esd nas oss
RDEPEND=esd? ( media-sound/esound )
nas? ( media-libs/nas )
# alsa-1 b0rks and it's not a simple fix
#alsa? ( media-libs/alsa-lib )
Interestingly, the alsa USE flag is commented out, but the IUSE=
variable contains OSS.
More interestingly, the mpg321 ebuild contains no listed dependencies on
/either/ ALSA or OSS (or any other sound server)-- which strongly
suggests that it's native ALSA, since that is enabled in the kernel by
default (whereas OSS is not), indicating that if your soundcard's kernel
modules load, mpg321 will work, since it seems to have no explicit
additional ALSA or OSS dependencies.
less /usr/portage/media-sound/mpg321/mpg321-0.2.10-r3.ebuild
# Copyright 1999-2005 Gentoo Foundation
# Distributed under the terms of the GNU General Public License v2
# $Header:
/var/cvsroot/gentoo-x86/media-sound/mpg321/mpg321-0.2.10-r3.ebuild,v
1.1 2005/08/29 09:48:18 flameeyes Exp $
inherit eutils
IUSE=
DESCRIPTION=Free MP3 player, drop-in replacement for mpg123
SRC_URI=mirror://sourceforge/${PN}/${P}.tar.gz
HOMEPAGE=http://sourceforge.net/projects/mpg321/;
DEPEND=media-libs/libmad
media-libs/libid3tag
=media-libs/libao-0.8.0
SLOT=0
LICENSE=GPL-2
KEYWORDS=~alpha ~amd64 -mips ~ppc ~ppc-macos ~ppc64 ~sparc ~x86
PROVIDE=virtual/mpg123
src_unpack() {
In any case, do you have ALSA OSS emulation enabled? This requires:
1) ALSA OSS emulation elements to be enabled in the kernel, including:
M Advanced Linux Sound Architecture
M Sequencer support
Sequencer dummy client
M OSS Mixer API
M OSS PCM (digital audio) API
[*] OSS Sequencer API
2) Your soundcard to be configured correctly in /etc/modules.d/alsa
3) the alsa-oss package (at least; there may be other associated alsa-*
packages necessary, but since I install all of them except alsa-driver,
I've never investigated which packages are actually necessary to enable
each area of the ALSA functionality).
If so, it _ /does/ _ work:
lsmod |grep snd_
snd_seq_midi6176 0
snd_opl3_synth 12292 0
snd_seq_instr 6656 1 snd_opl3_synth
snd_seq_midi_emul 5504 1 snd_opl3_synth
snd_ainstr_fm 2176 1 snd_opl3_synth
==snd_pcm_oss41760 0
==snd_mixer_oss 14208 2 snd_pcm_oss
==snd_seq_oss26880 0
snd_seq_midi_event 5632 2 snd_seq_midi,snd_seq_oss
==snd_seq42256 8
snd_seq_midi,snd_opl3_synth,snd_seq_instr,snd_seq_midi_emul,snd_seq_oss,snd_seq_midi_event
snd_cmipci 27552 1
==snd_pcm69128 2 snd_pcm_oss,snd_cmipci
snd_page_alloc 7688 1 snd_pcm
snd_opl3_lib8064 2 snd_opl3_synth,snd_cmipci
snd_timer 18564 3 snd_seq,snd_pcm,snd_opl3_lib
snd_hwdep 6560 1 snd_opl3_lib
snd_mpu401_uart 5504 1 snd_cmipci
snd_rawmidi18336 2 snd_seq_midi,snd_mpu401_uart
==snd_seq_device 6284 6
snd_seq_midi,snd_opl3_synth,snd_seq_oss,snd_seq,snd_opl3_lib,snd_rawmidi
snd41316 13
snd_opl3_synth,snd_pcm_oss,snd_mixer_oss,snd_seq_oss,snd_seq,snd_cmipci,snd_pcm,snd_opl3_lib,snd_timer,snd_hwdep,snd_mpu401_uart,snd_rawmidi,snd_seq_device
eix alsa-oss
* media-libs/alsa-oss
Available versions: 1.0.10-r1 ~1.0.11_rc3
Installed: 1.0.11_rc3
Homepage:http://www.alsa-project.org/
Description: Advanced Linux Sound Architecture OSS
compatibility layer.
# Alsa 0.9.X kernel modules' configuration file.
# $Header:
/var/cvsroot/gentoo-x86/media-sound/alsa-utils/files/alsa-modules.conf-rc,v
1.4 2004/11/16 01:31:22 eradicator Exp $
# ALSA portion
# OSS/Free portion
##
## IMPORTANT:
## You need to customise this section for your specific sound card(s)
## and then run `update-modules' command.
## Read
Re: [gentoo-user] autofs, supermount, submount... which is best for Gentoo?
[EMAIL PROTECTED] schreef: Very informative, thanks. I think I'll go with submount. These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild N] sys-fs/submount-0.9-r2 159 kB Total size of downloads: 159 kB localhost ~ # emerge submount Calculating dependencies ...done! emerge (1 of 1) sys-fs/submount-0.9-r2 to / snip 15:04:38 (64.11 KB/s) - `/usr/portage/distfiles/submount-0.9.tar.gz' saved [75476/75476] md5 files ;-) submount-0.9-r2.ebuild md5 files ;-) files/digest-submount-0.9-r2 md5 src_uri ;-) submount-2.4-0.9.tar.gz md5 src_uri ;-) submount-0.9.tar.gz * Determining the location of the kernel source code * Found kernel source directory: * /usr/src/linux * Could not find a usable .config in the kernel source directory. * Please ensure that /usr/src/linux points to a configured set of Linux sources. * If you are using KBUILD_OUTPUT, please set the environment var so that * it points to the necessary object directory so that it might find .config. !!! ERROR: sys-fs/submount-0.9-r2 failed. !!! Function linux-info_pkg_setup, Line 537, Exitcode 1 !!! Unable to calculate Linux Kernel version !!! If you need support, post the topmost build error, NOT this status message. I'm not sure what's meant by the topmost build error, but as it's not too large, I included everything. What is meant is the last output right before ERROR:; in this case, it is * Could not find a usable .config in the kernel source directory. * Please ensure that /usr/src/linux points to a configured set of Linux sources. This package compiles against the kernel, as you can see from * Determining the location of the kernel source code * Found kernel source directory: * /usr/src/linux However, the kernel source that the /usr/src/linux symlink points to has not been configured using make (menu/x)config. Therefore there is no .config file that the package can examine to ensure that the kernel source in question has/will be built with the support that the package requires. You don't have to build or install this kernel source, but you do have to configure it (properly for the submount package) before you attempt to install the submount package. I'd think that the wiki entry will detail the necessary kernel settings. Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] sudo echo
JimD schreef:
I have been using Linux for a number of years and the one trick I
have never read how to do is something like:
sudo echo app-portage/porthole ~* /etc/portage/package.keywords
Well this one I do with a set of revised command nicked from the list,
entered into ~/.bashrc, and requiring that
1) su is one of the commands that you are allowed to execute via sudo
2) you are exempted from needing to enter a password for 'sudo su':
addkey(){
sudo su -c echo $* /etc/portage/package.keywords
}
adduse(){
sudo su -c echo $* /etc/portage/package.use
}
addmask(){
sudo su -c echo $* /etc/portage/package.mask
}
addunmask(){
sudo su -c echo $* /etc/portage/package.unmask
}
The general idea being that a) sudo seems to be a bit weird; even though
it allows you to perform operations as if you are root, it doesn't do so
by pretending that you _are_ root, so you still couldn't write to the
/etc/portage/package.* files; b) su does pretend you are root, but su
alone only just re-logs you in, rather than actually allowing you to
execute a command-- unless you use the -c switch. su -c then says,
whatever follows this switch is a command that you should execute as
root. But of course, since echo $* (where $* stands for what I typed
after addkey) /etc/portage/package.* is a complex command,
containing spaces, the syntax of the command following sudo su -c needs
to be quoted.
Another one I always wanted to know if it is possible is:
sudo /var/log/foo.log
I'm sure it is, with a bit of creativity, though I honestly don't know
what your intention is in any case, since this looks to me like you're
logging the output of the sudo command to foo.log (but since there is no
output really to typing 'sudo', I have no idea what result you might
expect).
Anyway, hope this is to some degree helpful; what you most likely want
to do is read up on bash scripting to understand how to chain the
commands that do what you want to get done with sudo. Depending on your
goals, you might also consider aliasing (alias etc-update=sudo
etc-update), and fine-tuning your visudo to allow you to run specific
apps with sudo, preferably without a password, since if you have to type
the password everytime you want to do sudo emerge, you might as well
just su, imo.
Good luck,
Holly
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge java syntax
[EMAIL PROTECTED] schreef: I'm at http://www.gentoo.org/proj/en/java/tiger-faq.xml, which gives the syntax as: emerge =sun-jdk-1.5* but the pretend isn't working as I expect it to. localhost ~ # localhost ~ # localhost ~ # emerge -p =sun-jdk-1.5* These are the packages that I would merge, in order: Calculating dependencies !!! All ebuilds that could satisfy =sun-jdk-1.5* have been masked. !!! One of the following masked packages is required to complete your request: - dev-java/sun-jdk-1.5.0.06-r2 (masked by: package.mask) # [EMAIL PROTECTED] # Many things in the tree don't compile with 1.5 yet. # 1.5 defaults to -target 1.5, which makes downgrading to a 1.4/1.3 # impossible. See bug #69970 and bug 65937 for more information. # http://www.gentoo.org/proj/en/java/tiger-faq.xml For more information, see MASKED PACKAGES section in the emerge man page or refer to the Gentoo Handbook. This seems very clear. All ebuilds that could satisfy =sun-jdk-1.5* have been masked. For more information, see MASKED PACKAGES section in the emerge man page or refer to the Gentoo Handbook. This means that the package is masked (not available for installation unless explicitly unmasked), for the reasons noted in the error. If you want it nonetheless, read man emerge to find out how to unmask the package so it will install. I can assure you that if you unmask it, it will install (I'm using it, but I don't compile much, if anything, against Java, so the issue is not an issue for me). HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Lilypond version
Allan Spagnol Comar schreef: Hi, I don't know where should I post this kind of message, but I need to use lilypond as one of my working tools. I got stunned when I saw today that lilypond version available at portage tree is 2.0.3 the latest stable version is 2.6.5 and 2.0.3 is extreme old. Is there a reason that lilypond version is not up to date and should I build myself ( without portage ) if I want to used it ? eix lilypond * media-sound/lilypond Available versions: 2.0.3 ~2.2.4 ~2.2.6 ~2.4.2 ~2.5.2 2.7.9[2] ~2.7.27[2] Installed: none Homepage:http://lilypond.org/ Description: GNU Music Typesetter [1] /usr/local/portage [2] /usr/portage/local/layman/ecatmur So... still not up-to-date, /per se/, but there are more current versions in Portage (2.5.2), and development versions available in overlay, as well as on bugs.gentoo.org: http://bugs.gentoo.org/show_bug.cgi?id=97574 They're just not the /stable/ (in terms of Portage) version, but they are available; you must be keyworded ARCH, so all you see is the last stable (as per Portage) version. You might consider keywording lilypond in /etc/portage/package.keywords to get the ~ARCH version (I'm ~x86, so don't know if the versions are the same/available for other ARCHes), or learning how to create a PORTDIR_OVERLAY (overlay directory tree). See /man portage/ for information on keywording a specific package; see http://gentoo-wiki.com/Portage_Overlay for information on overlay directories and a listing of sources (other than b.g.o) for 3rd party ebuilds (such as ecatmur's overlay, listed above). HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Sharing file systems
Hemmann, Volker Armin schreef: On Sunday 19 March 2006 13:04, Sumeet Pal Singh wrote: HI This mail is not directly related to gentoo. I have gentoo,ubuntu 5.10 on my system and using FC3 since it was released. I tried to install FC4 on my system by sharing the swap and /home partition between FC4 and ubuntu. The installation went well. I did not install KDE in FC4. snip I booted in as root and created a new user with UID and GID same as on ubuntu. Now I booted in GNOME and it was completely configured Moreover gaim started and signed me in to all my accounts!!! This was first time login... My question is that can sharing /home okay for a long run or will it lead to problems. snip as long as you are using the same software versions on both systems (at least the major version number should be the same) it should be ok. That's what I would have thought, too. But frankly, it seems too much maintenance to me, since as soon as the versions go out of sync, then you're likely to have problems that are difficult to track down. I have no problem with two users from different distros having their /home folder on the same partition (in fact, my Gentoo install and my SuSE install share a /home partition), but I wouldn't myself have the two users merged that way across two Linux distros. I admit I did do something similar when I had a massive multiboot (5 Linux distros, 2 Windows installs), for relatively easy compatibility with the shared Windows partitions, but even then, every user had their own /home folder, they just had the same UID and a shared GID (which I made the same on all related distros). But then again, I don't store things in my /home folder; I store them on their own partitions that are linked into my home folder, so it's not as if it's a space saver for me to have two separate distros using the same /home/username, and since I see it as a relatively dangerous pain-in-the-butt, I don't do it. /usr and the like is right out, obviously-- your libs will get creamed pretty quick doing something like that, unless they're /exactly/ the same across both distros, and that seems unlikely to me (or unlikely to last long, if you're lucky temporarily). But why are you asking here and not at the ubuntu or fc mailing lists? Good question. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: QT 4
Dmitry S. Makovey schreef: there are quite a few slotted packages in portage so you might bump into this every now and then (KDE is slotted AFAIR). Even easier-- so are kernel sources. Install a new one, and it's always going to be [ NS ], not [ U ] . Looking at how kernel sources are handled on the system seems to me to be the easiest entry point to understanding SLOTs. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Gentoo USE flags
Goran Maksimović schreef: Hi! I said what will I do on my system and I value the answers of people who really answered what I have asked. Great, happy to hear it. Congratulations to our lucky winners, good luck in the future to everyone else, and can we please move on now? The 'discussion' would seem to be complete. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Gentoo USE flags
Neil Bothwick schreef: On Wed, 15 Mar 2006 21:18:25 +0100, Goran Maksimović wrote: If you read my first post about USE flags I wanted some recommendations and not links to sites where USE flags are explained. Did you read the second paragraph of my post before replying to it? I think he read it, but didn't understand it... after all, anybody who would say something like Goran Maksimović schreef: I don't know why you think that if you don't fine tune your system clearly doesn't know you from around here (or anywhere else :-) ) -- even a little bit. I'd LOL if I wasn't having a 'thing' today where I just don't want to watch half the list (including me, actually) come out swinging on somebody who's missing the point, shall we say, since apparently none of our attempts to explain why what Goran wants us to do is not really possible are getting through. But hey, have a good time, y'all. It's way past my bedtime anyway. :-D Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] how to kernel update when executing emerge -u world?
[EMAIL PROTECTED] schreef: Every time when I run emerge -u world, the kernel will be updated if it have some update. This is normal; you're asking for updates (-u), so Portage is offering you the available update you asked for. Why is this a problem? Are you running a different kernel? Not suspend2-sources, but something else and you simply don't want suspend2-sources present on your system anymore, or is there an issue with the specific version of suspend2-sources that's being offered, so the issue is that you don't want to upgrade? If the former (you used to use suspend2-sources, but you switched to another kernel variant like gentoo-sources or ck-sources, and you don't want to upgrade suspend2-sources anymore), the solution is to emerge -C (unmerge) the version of suspend2-sources you had installed (delete the source in /usr/src before unmerging to speed up the process). Then it won't be updatable, because it's no longer present (but since the source of the kernel variant you are using is still present, that will continue to be upgraded).. If the latter (you are still using suspend2-sources, but you don't want to upgrade it beyond a certain version), the solution is to mask the versions beyond what you want manually, using /etc/portage/package.mask: (not with quotes) =sys-kernel/suspend2-sources-2.6.15-r7 (since you're using -r6 and attempting to upgrade to -r8, this will mask everything including to and above the version beyond the one you're using). The downside of this is that you have to keep an eye on the package to see if the problem has been solved in order to get any further updates, as Portage will no longer tell you or offer you updates to this package. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Gentoo USE flags
Goran � schreef: Hi! I have already asked this but I will ask again. I am looking for who could write me a list of USE flags to setup in make.conf, so anyone? I need desktop system with KDE and not GNOME, support for DVD and CD ripping and burning, playing DVDs and DivX movies, MP3. I will do programming and testing of Apache server, mail server, MySQL. Apologies in advance; this is likely to be a rather testy response. Normally I'm more patient and polite. The most likely reason that you haven't gotten the response you're looking for (though you did get a list of suggestions from Richard Fish, among others)-- or at least the reason that I didn't respond-- is because USE flags are really very relative to the programs you're installing, and without knowing that (which is, for us, /wy/ too much information, so we don't *want* to know), it's fairly impossible to give a complete list of the USE flags you specifically should use for the use that you specifically are going to put your system to. Furthermore, in the five days since the last response in your previous thread, you could have read /usr/portage/profiles/use.desc and /usr/portage/profiles/use.local.desc (not to mention http://www.gentoo.org/dyn/use-index.xml ) any number of times and figured this out for yourself-- which you have to do anyway (figure it out for yourself, I mean). It's your system. You are in control. That means you have to know what you're doing, and to know what you're doing, you have to know what you want-- specifically-- and how to make the system do what you specifically want it to do (which means read the docs provided to tell you how to make the system do what you want). Heck, the longest part of a Gentoo install for me is the two hours I spend reading these docs and setting up my USE flags (obviously, compiling things takes more time, but I don't have to sit there and pay attention for that). I mean, really, we just can't know what you need. Fine, you're not going to use GNOME but KDE, so I'm sure that you've figured by now that you need -gnome and +kde. What about GTK (or GTK2)? Are you going to use Firefox, Mozilla, or neither? Abiword? OpenOffice? /Evolution/? These are GTK applications (though not necessarily GNOME applications, with the exeption of Evolution), so they will need GTK support-- which they'll get automatically without reference to USE flags but what about their dependencies (which may need to be compiled +gtk or +gtk2, which may then cause the app to bug out if you've compiled the dependencies without such support, because you thought gtk=gnome and I don't want GNOME). Not to mention that some of these apps spawn USE flags of their own, and that may not always be what you want. Examples: Lots and lots of unaffiliated apps use web browser support. There are separate USE flags for firefox and mozilla, but in a few rare cases, the mozilla USE flag covers Firefox as well. If you plan to install Liferea (the RSS news reader) you'll need to know which one of those two browsers you're planning to use, because enabling both flags would mean that one of those browsers is going to be installed unnecessarily-- and they both take a pretty long time to compile. Are you going to upgrade/try out Xorg 7? If we don't know your video card, or whether you want all the wizbang possible (evdev, composite) or whether you do or don't have a joystick/gamepad, how are we supposed to tell you what you should put in VIDEO_CARDS and INPUT_DEVICES? You're going to do programming and testing of Apache server, mail server, MySQL. Which apache? 1 or 2? Which mail server? Just because you're testing an Apache server, does that mean that any install of subversion or CVS that you might perform must also be compiled with apache(1/2) support? Maybe you don't want that, so how are we supposed to say oh, just enable it globally? I could go on like this forever, but hopefully you get the point. The reason why there is a USE= line in /etc/make.conf, in addition to default USE flags being set by /usr/portage/profiles/use.default, in addition to possible USE flag modifications from /etc/portage/package.use, as well as fairly copious documentation about what each USE flag generally indicates, as well as a --verbose flag in Portage to allow you to see what USE flags are in use before compiling, is because it's *your* system; its customization *is up to you.* No one but you is in a position to know that you need java support for your web browser, but you don't need it for OpenOffice.org. We can't say, enable it globally, or disable it globally, because *we have no way of knowing what you need your system to be capable of* *when you sit down in front of it*. You say, DVD ripping-- I say, Ripping to what? Do you need xvid/quicktime/ogg/theora support when you rip these DVDs? Maybe you do, maybe you don't-- but I have no way of knowing that, because I don't know what you need the final files to be capable of; only you do.
Re: [gentoo-user] Gentoo USE flags
Goran � schreef: Thanks! I didn't only understand something about per packages which you mentioned in your mail. USE flags can be (and of course, are) set globally in /etc/make.conf (and /usr/portage/profiles/use.defaults), but USE flags can also be enabled or disabled for a particular package only using /etc/portage/package.use. You might do this for a number of reasons: 1) Every USE flag does not apply to every package, and some local USE flags only apply to one specific package. It keeps your /etc/make.conf neater and easier to read if it only contains the really global USE flags. Some of us care about that kind of thing :-) . But the main reason you'd use /etc/portage/package.use is because 2) You have a globally-enabled (or -disabled) USE flag that you want to disable (or enable), for a particular package. Example (from my own config): USE=*-kde* -arts -eds -esd -apache -apache2 16bit 3dnow aalib acpi -apm audiofile bash-completion -berkdb bigger-fonts bitmap-fonts bmp bzip2 caps cairo cddb cdr curl dbus dga directfb dri dts dv dvd exif expat extrafilters fam fbcon ffmpeg firefox flac font-server gamin ggi gimpprint glut gtkhtml gnutls -gstreamer gstreamer10 guile hal iconv idn imagemagick inkjar inotify -ipv6 jack *java* jikes kdeenablefinal kdexdeltas lcms ldap libcaca libnotify lua maildir math mmx mng -mozilla mpi nas nfs nptl nptlonly offensive openexr -pam pcre pic portaudio povray *samba* scanner slang sndfile socks5 sqlite sse -sse2 svg svga tcltk tetex theora threads tiff toolbar truetype-fonts type1-fonts unicode usb v4l wmf xml xprint Xaw3d v4l v4l2 xvid yv12 As you can hopefully see, the java and samba flags are enabled (globally), and the kde flag is disabled (globally). You can't see it here, but the default profile I'm using also disables the doc USE flag (which is extra documentation, not man pages and the like), as well as the symlink USE flag (which automatically re-targets the /usr/src/linux symlink when a new kernel source is installed. Now, the java USE flag is enabled globally, and it is a valid USE flag for OpenOffice, but I don't want it enabled for OpenOffice, because I don't need for OO.o to use Java, and it slows an already-slow application down. As for doc, well, I have Imagemagick installed, and it's very complex, so I want the extra documentation, but generally, I don't need it; --help and man pages are usually good enough. As for samba, I don't want Midnight Commander to have samba support, because if I'm using mc that extensively, my system is probably borked somehow (hey, I'm a GUI kinda girl, as great as mc is), or if somebody has compromised the system and has managed remote command-line access to mc, at least they can't just go right to the network shares. In either case, I want to limit access to network shares to try and minimize the ability to do damage to them. (Yes, I know it's not massively protective and perhaps not even effective, but at least I care enough to try-- and since Gentoo gives me the option, I took it ;-) .) As for kde, well I use it extremely rarely, but I want the OpenOffice to be correctly integrated with the environment if I do ever have to log into KDE rather than my preferred DE (KDE is my fallback DE of last X-resort, and it's much more likely that I'll need to print out a cv or something during a computer emergency that resulted in my having to use my last X-based fallback than that I'll have to read my mail-- that can wait, or I'll use webmail-- which is why you'll see that my fallback mail-client does not have kde support). So, in /etc/portage/package.use I have the following lines (among others): app-misc/mc 7zip -X -samba app-office/openoffice-bin kde -java media-gfx/imagemagick fpx inkjar plugin doc =sys-kernel/gentoo-sources-2.6.15-r7 symlink Thus, when I do an emerge involving those packages : emerge -pv openoffice-bin mc imagemagick mozilla-firefox kdebase-kioslaves sylpheed-claws =sys-kernel/gentoo-sources-2.6.15-r7 =sys-kernel/gentoo-sources-2.6.15-r5 (additional packages to show use of the default settings) | cfg-update-1.8.0-r3 : Building checksum index... (takes a few seconds) done! These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] app-office/openoffice-bin-2.0.2 USE=gnome -java 0 kB [ebuild R ] app-misc/mc-4.6.0-r14 USE=7zip gpm ncurses nls slang unicode -X -pam -samba 49 kB [ebuild R ] media-gfx/imagemagick-6.2.6.0 USE=X bzip2 doc fpx jpeg lcms mpeg perl png tiff truetype wmf xml2 zlib -graphviz -gs -jbig -minimal -nocxx 0 kB [ebuild R ] www-client/mozilla-firefox-1.5.0.1-r2 USE=gnome java xprint -debug -ipv6 -mozdevelop -xinerama 0 kB [ebuild R ] kde-base/kdebase-kioslaves-3.5.1 USE=hal kdeenablefinal kdexdeltas ldap openexr samba -arts -debug -xinerama 0 kB [ebuild R ] mail-client/sylpheed-claws-2.0.0 USE=clamav crypt dillo gnome ldap spamassassin spell ssl -doc -imap -ipv6 -kde -pda -xface 0 kB [ebuild R ]
Re: [gentoo-user] Reasons for testing or stable of kde packages
Marco Calviani schreef: Hi list, i would like to have clarification regarding the policy of switching packages from testing to stable. Is this policy due to particular bugs in the packages? No. Gentoo's stable and testing refers to the /ebuilds/, not the packages. I'm not a dev, but from my experience, if upstream (the developers of a particular package) release the package, then it is considered to be 'stable' (insofar as it's releaseable, and Gentoo does not include betas or development versions in the Portage tree). However, the ebuild script that allows the package to compile on Gentoo may contain errors, so it must be tested. That is what ~ARCH is about; making sure the provided ebuild compiles the source of the application correctly and successfully with relationship to the rest of a Gentoo system. ~ARCH packages/ebuilds are normally tested for (30? 90?) days, after which time if no bugs are filed, they generally move into stable. It is hoped that users who use ~ARCH are willing to file or comment on bugs on bugs.gentoo.org (b.g.o). The system only works if everybody helps. If a package itself contains serious errors (fairly easily discerned from a filed bug whether the problem is the ebuild or the package), you'll see it getting hard-masked pretty darn quick (that's what hard-masking is for/about-- it generally means the package itself is broken), while we wait for upstream to fix whatever is wrong. While 'testing' is just that, and therefore not specifically 'stable', in practice testing usually is pretty stable (all problems I've had I've been able to overcome myself, despite not being a developer or any kind of programmer who understands deeply what's actually going on). However, sometimes testing does expose upstream bugs that cause the package to break (or break on your system), so ultimately, it's not safe (as in safe as houses), but it's not like it's usually risking your system stability in a major way (i.e., you can't boot, or the system fails to operate, etc.) HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Reasons for testing or stable of kde packages
Marco Calviani schreef: Hi Holly, and thanks for your clear explanation, Hi list, i would like to have clarification regarding the policy of switching packages from testing to stable. Is this policy due to particular bugs in the packages? No. Gentoo's stable and testing refers to the /ebuilds/, not the packages. Well, that was my fault in explaining. i was referring to versions of a particular package ~ARCH packages/ebuilds are normally tested for (30? 90?) days, after which time if no bugs are filed, they generally move into stable. It is hoped that users who use ~ARCH are willing to file or comment on bugs on bugs.gentoo.org (b.g.o). The system only works if everybody helps. At the root of my question there was the need to understand why kde 3.5.1 packages are still testing even if there aren't critical bugs at bugs.gentoo.org (as far as i was able to find...). Possibly because KDE is a huge suite of interrelated packages, /all/ of which must be 'stable' (in ebuild terms) so that users who have expectations of what 'stable' means will be satisfied (especially for those who run ARCH and do not use ~ARCH packages). This may be why it takes a longer time for Gentoo to move all of those packages into ARCH; a single package is obviously easier to test than the many split ebuilds that KDE contains, especially when varied combinations are involved. I recognize that 'stable' users value... stability... over the latest and greatest, and that's fine. But it always annoys me somewhat when users then 'complain' about 'the latest and greatest' not being moved into stable in a timely (in their opinion) fashion. Either one wants guaranteed stability, or one is willing to possibly sacrifice some of that stability for new features or cool factor or whatever. The choice, like most things, Gentoo, is up to the user, and Gentoo users should always be aware that they need to explicitly choose and commit to their decision. Of course, that wouldn't be a reasonable position if one didn't trust the dev's judgement, but in fact, that's almost the first choice a Gentoo user needs to make; certainly about the status of the Portage tree, if one is not going to monitor or participate in the activities of the dev team (in this case, the KDE herd, I suppose). After all, if you think about it, the gentoo-*users* list is not really a logical place to seek answers as to why a specific arm of the development team has taken or not taken a specific action. Not that the dev list wants to hear anybody nagging them about this, mind you, but I would imagine that if any discussion is going on as to progress of migrating the ebuilds to stable, it would be there rather than here (so one could lurk and find out what was going on, if one cared to), or on an IRC channel (if one exists) or something like that. Possibly even on b.g.o., if someone has filed a bug to move KDE to stable (that happens), and a dev has responded to that bug with a reason why or why not, or with a we're doing it, buzz off! Gentoo is a very hands-on distro, and everyone can get involved to the extent that they're capable, and everyone is capable to some extent, because the extensive documentation enables one to know enough to do /something/ (write an ebuild, update an ebuild, file a bug, comment on a bug, etc). This is not some back-room, behind-closed-doors, we don't want to hear from you if you don't send a patch kind of dealie, so if you want to know the progress of the KDE herd in migrating KDE 3.5.x to stable, I'm sure it's possible to find out without a lot of difficulty. Just not necessarily from us, unless somebody who does monitor the KDE herd happens to be around (which is not a particularly efficient way of getting an answer to your question). :-) Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] mouse detected, but no /dev/mice
[EMAIL PROTECTED] schreef: hiya, I have two USB mice. Dmesg shows them being detected, but there is no /dev/mice nor /dev/mouse. Any suggestions? If you're using udev, the devices should be found in /dev/input/mice. At least that's where mine is. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] radeon and framebuffer
Aggelos schreef: Holly Bostick wrote: I myself have a 9800SE, and also use vesa-tng for fbsplash. Both fbsplash/framebuffer and fglrx work fine; aside from the fb console background, I can even play mplayer videos in the getty console, which is also a framebuffer operation as far as I know. I have heard that enabling the radeon framebuffer is a bad idea, though (it doesn't work well with the fglrx drivers). But stick to vesa or vesa-tng, and you should have no problems. I haven't, from 2.6.13 kernels up to my current (2.6.15-r7), and with fglrx versions from 8.8.whatever to my current 8.22.5. Could you tell me the options you are passing to the kernel at boot time (I am using grub) for the vesa-tng ? Sure, sorry it took so long to get back to you; I've been busy in RL. (from an old grub.conf backup; ignore the kernel version, as the options don't change-- I just change the title as I upgrade). Gentoo_current (2.6.12-gentoo-r10) root (hd0,1) kernel /vmlinuz ro quiet root=/dev/hda5 video=vesafb:ywrap,mtrr:3,[EMAIL PROTECTED] splash=silent,fadein,theme:livecd-2005.1 CONSOLE=/dev/tty1 It's pretty much exactly the same as the example on the Wiki (http://gentoo-wiki.com/HOWTO_gensplash#GRUB_Example), except that with the upgrade of splashutils, I now need mtrr:3 so that the splash comes up in a timely fashion (there's an einfo in the ebuild about this iirc), and of course changed for my system specs. The only thing that gave me a problem is that I had to set [EMAIL PROTECTED] as my default resolution for vesa-tng in the kernel config itself; you're not supposed to /have/ to, but if I didn't, I got errors when using the livecd themes (no 8bpp pictures, which is correct, since the livecd themes don't have 8-bit pictures, but this meant that the drivers were for some reason defaulting to 8-bit if I didn't specify 32-bit in the kernel. Which according to another thread on this subject is not correct behaviour-- if I specify 32-bit in the grub kernel line, it's supposed to use that instead-- but that's what it did on my system anyway. Specifying CONFIG_FB_VESA_DEFAULT_MODE=[EMAIL PROTECTED] solved the issue, since I don't like the Emergence theme, which does have 8-bit pictures, so worked correctly in all cases). Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Fairwell for now
A.R.S. KA9QLQ Alvin Koffman schreef: It recognises my Nvida card [fx 5200] but when you get to the stage where the icons dissapear as stuff loads, when the last one goes the monitor shuts off. as is't booting from a cd there's no error log and nofb didn't help. Like I said some day Gentoo will get there, but right now it's not stable. The graphical installer is not the distribution. The fact that the installer doesn't work for you is not the same as Gentoo is not stable. The graphical installer is indeed not stable, but we all knew that (or we would have if we read the release notes). Is there some reason that you can't use the current CD to install in the traditional manner (without X), or with an ncurses interface (again, without X), or conversely use the previous install CD (that doesn't have a graphical installer), to install Gentoo? Also, you might check the forums and bugs.gentoo.org to see if this is a known problem (you seem to be not the first having signal loss from the install CD on this list) and if there is a workaround. Or you could just go on as you are doing and blow us off as unstable. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Fairwell for now
A.R.S. KA9QLQ Alvin Koffman schreef: As for insulting the developers I said I looked forward to when they had something I [a normal person] could use You know, it just occurs to me to question this often-heard assumption that non-geek=normal -- with geek being defined by these so-called normal people. Is it not normal to want to understand what you're doing when installing an OS? Is it normal to object (vociferously, usually) to having to read the instructions before attempting to perform a complex technical procedure? Is it normal to think that click a fancy button and everything just works is the way things are *supposed* to be, when dealing with a complex technical operation? I personally don't think so, but then again I'm a geek, apparently, and I don't know what normal is, anyway, since I'm apparently not. Not to be rude, but if I was normal, I suppose I would understand how one can be a normal person and severely disabled in the same paragraph, since these conditions either have no relationship to each other (in which case there was no need for the I'm disabled trump card), or contradict each other (in which case the speaker looks like a. person who smells of elderberries). But of course, I'm out of here; pointless flame-fests are not my idea of a fun Friday night. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Confused about latest gentoo-sources
Jerry Turba schreef: Willie Wong wrote: Do you have gentoolkit installed? If so, run # equery list -p gentoo-sources Results of equery: backup:~ $ equery list -p gentoo-sources [ Searching for package 'gentoo-sources' in all categories among: ] * installed packages [I--] [ ] sys-kernel/gentoo-sources-2.4.31-r1 (2.4.31-r1) [I--] [M ] sys-kernel/gentoo-sources-2.6.11-r3 (2.6.11-r3) [I--] [M ] sys-kernel/gentoo-sources-2.6.12-r9 (2.6.12-r9) [I--] [M ] sys-kernel/gentoo-sources-2.6.12-r10 (2.6.12-r10) [I--] [M ] sys-kernel/gentoo-sources-2.6.14-r2 (2.6.14-r2) * Portage tree (/usr/portage) [-P-] [M~] sys-kernel/gentoo-sources-2.4.32-r2 (2.4.32-r2) [-P-] [M~] sys-kernel/gentoo-sources-2.6.14-r7 (2.6.14-r7) [-P-] [M~] sys-kernel/gentoo-sources-2.6.15 (2.6.15) I do have gentoo-sources-2.6.12-r10 installed and running. I never had a 2.4 kernel installed on this system. Thanks for your help. Myself, I'm starting to wonder if this isn't a mirror issue (what mirror are you sync-ing with? Check /etc/make.conf), because mine looks like this: equery list -p gentoo-sources [ Searching for package 'gentoo-sources' in all categories among: ] * installed packages [I--] [ ~] sys-kernel/gentoo-sources-2.6.15-r5 (2.6.15-r5) [I--] [ ~] sys-kernel/gentoo-sources-2.6.15-r6 (2.6.15-r6) [I--] [ ~] sys-kernel/gentoo-sources-2.6.15-r7 (2.6.15-r7) * Portage tree (/usr/portage) [-P-] [ ] sys-kernel/gentoo-sources-2.4.31-r1 (2.4.31-r1) [-P-] [ ~] sys-kernel/gentoo-sources-2.4.32-r2 (2.4.32-r2) [-P-] [ ~] sys-kernel/gentoo-sources-2.6.14-r7 (2.6.14-r7) [-P-] [ ~] sys-kernel/gentoo-sources-2.6.15 (2.6.15) [-P-] [ ] sys-kernel/gentoo-sources-2.6.15-r1 (2.6.15-r1) [-P-] [ ~] sys-kernel/gentoo-sources-2.6.15-r2 (2.6.15-r2) [-P-] [ ~] sys-kernel/gentoo-sources-2.6.15-r3 (2.6.15-r3) [-P-] [ ~] sys-kernel/gentoo-sources-2.6.15-r4 (2.6.15-r4) If you've synced twice already, and you're still getting such old packages (2.6.*11*??), there's gotta be something wrong with your mirror or your cache, it seems to me. Cache I don't know how to fix, but changing mirrors is an easy way to fix the problem or eliminate one possible vector of trouble (if nothing changs, then we know the problem is something else, and there aren't too many things it could be, at that point). HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Confused about latest gentoo-sources
jerry schreef:
Holly Bostick wrote:
Myself, I'm starting to wonder if this isn't a mirror issue (what
mirror are you sync-ing with
If you've synced twice already, and you're still getting such old
packages (2.6.*11*??), there's gotta be something wrong with your
mirror or your cache, it seems to me.
Here is /etc/make.conf:
These settings were set by the catalyst build script that
automatically built this stage # Please consult
/etc/make.conf.example for a more detailed example CFLAGS=-O2
-march=pentium3 -fomit-frame-pointer -pipe CHOST=i686-pc-linux-gnu
CXXFLAGS=${CFLAGS}
GENTOO_MIRRORS=http://gentoo.osuosl.org/
http://mirror.datapipe.net/gentoo http://gentoo. seren.com/gentoo
SYNC=rsync://rsync.us.gentoo.org/gentoo-portage
snip
I am new to and learning how gnetoo works, but I don't quite follow
your line of thought; I download packages from osuosl.org (oregon
state?) but when I sync I seem to hit a random server somewhere. Last
time I synced it was in Milwaukee,Wisc.
Yes, that's correct. Syncing with the Portage tree (that's the SYNC
setting, above) is a different operation than downloading packages (from
the GENTOO_MIRRORS setting above).
You sync your Portage tree from a pool of Portage servers on your
continent. Which server is chosen is random; this decreases the load on
the individual members of the server group.
I take it you're in the U.S., so you're syncing with the US pool; I'm in
The Netherlands, so my SYNC variable looks like this:
SYNC=rsync://rsync.nl.gentoo.org/gentoo-portage
When I emerge --sync, I connect to one of the various Gentoo rsync
servers in this country randomly.
However, whenever I download tarballs, they come from one of these servers:
GENTOO_MIRRORS=ftp://ftp.easynet.nl/mirror/gentoo/
http://www.ibiblio.org/pub/Linux/distributions/gentoo
http://gentoo.osuosl.org;
And if the file is not found on any of them, the ebuild will usually
direct Portage to attempt to download directly from the homepage of the
project or Sourceforge, whichever is applicable.
This is all as it should be.
In any case, the problem I was trying to track is whether the reason
your local Portage tree was not current, even though you've synced,
might not be because the remote Portage tree you were syncing with was
not current (it can happen; servers change or are not often updated, and
if your profile was very old, you might be using an old server that
wasn't in sync with the times).
However, it's hard to imagine that the US pool might not be current.
[General question] What else might prevent a sync from actually updating
the local tree correctly?
U unless it actually has, and the problem is Portage not knowing
it somehow
Would you do an
ls /usr/portage/sys-kernel/gentoo-sources/
and post the output?
I'm just wondering if the ebuilds are physically there or not.
I have the uninformed naging thought that I am having this problem
because I have not upgraded gcc from 3.3.6. I posted this thought
earlier but no one commented on its likelihood.
I cannot imagine what gcc (your compiler) would have to do with the
contents of the Portage tree, or why you would think it did :-) .
Upgrading GCC or not doing so has nothing, as far as I know, to do with
rsync or the actual functionality of updating your Portage tree.
Certainly gcc
is not involved in any way in actually updating the Portage tree, which,
as far as we currently know (until you post the output of that ls,
anyway), is the issue at hand.
HTH,
Holly
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Confused about latest gentoo-sources
jerry schreef: Holly Bostick wrote: However, it's hard to imagine that the US pool might not be current. [General question] What else might prevent a sync from actually updating the local tree correctly? U unless it actually has, and the problem is Portage not knowing it somehow Would you do an ls /usr/portage/sys-kernel/gentoo-sources/ and post the output? I'm just wondering if the ebuilds are physically there or not. backup jerry # ls /usr/portage/sys-kernel/gentoo-sources/ ChangeLog gentoo-sources-2.6.15-r3.ebuild Manifest gentoo-sources-2.6.15-r4.ebuild files gentoo-sources-2.6.15-r5.ebuild gentoo-sources-2.4.31-r1.ebuild gentoo-sources-2.6.15-r6.ebuild gentoo-sources-2.4.32-r2.ebuild gentoo-sources-2.6.15-r7.ebuild gentoo-sources-2.6.14-r7.ebuild gentoo-sources-2.6.15.ebuild gentoo-sources-2.6.15-r1.ebuild metadata.xml gentoo-sources-2.6.15-r2.ebuild Well, wouldja look at that! So the ebuilds *are* there, but Portage is just not offering them. Have you: 1) checked /etc/portage/package.mask (maybe you masked them and forgot)? 2) checked your Portage version (maybe you need an upgrade)? 3) tried emerging one of the invisible versions explicitly (emerge -av =sys-kernel/gentoo-sources-2.6.15-r7)? Even an error or failure might tell us more about what's going on than we know now. HTH; talk to you after the gcc upgrade is done. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Monitor goes off on boot
A.R.S. KA9QLQ Alvin Koffman schreef: Well I'm stumped. Every time I boot the live cd it gets to where Gnome should start then the monitor goes off. Does any one know what video drivers the live cd uses? Alvin For the best jerky you've ever had go to http://alk.jerkydirect.com/ My home page http://ka9qlq.tripod.com This PC is windows free with Mepis Linux 3.4-3 http://www.mepis.org/ 1(747)632-4973 SIP Get Gizmo 1 cent per minuet calling http://www.gizmoproject.com/ What video card do you have and what drivers? I've had similar problems (not with this graphical live CD, since I installed Gentoo before it existed, but with X applications and X itself), because I have an ATI card. ATI cards do do that (just shut down the monitor) if 1) using the wrong drivers (radeon when card model is one above the 9(2/5?)50 (sorry, can't remember which model is the stopper for the Open Source drivers), and/or 2) DGA is enabled for the fglrx drivers (this will do exactly what you described; it has many times for me, and it is just one of the many PITAs with the fglrx drivers). The thing is-- in theory, I have no evidence to support this-- that GNOME (I am a GNOME user as opposed to a KDE user, though I don't use either of those DE's regularly or first during a new install due to their size) appears to require 3D support be working in order to load properly. Or at least, the hardware acceleration must be working if you're using drivers that supposedly support said acceleration. As I said, I have no evidence for this /per se/; it's just my theory based on experience. If you're using the 'vesa' drivers (which don't support 3D), I betcha GNOME will load fine (at least it always does for me), but as soon as you load drivers in your X config that are supposed to support hardware acceleration/OpenGL/3D, GNOME will break if that support is broken (even though, afaik, no basic operation of GNOME actually uses 3D-- that's why this is a theory and seemingly rather a crackpot one, but it's the only theory that fits my experience). So I would suggest first changing your xorg.conf to load the vesa drivers, which should load (that's what they're for, default drivers that should always be able to load and display), then editing your xorg.conf to resolve the obvious problem that it must have. Some option or driver causes your video card to stop sending a signal. I know that =9600 ATI cards do this when DGA is enabled on the fglrx drivers, and also that =9(2/5?)50 cards do strange things when using the radeon drivers which don't support these models for hardware acceleration, as opposed to the fglrx drivers which do-- but the LiveCDs will tend to (in my experience) recognize my 9800SE as an ATI card and load the radeon drivers incorrectly because the fglrx drivers that the card needs are not open source... and the radeon drivers don't actually work properly for my card. But you may have a different brand of card, or it could be a different option causing this. You might want to copy /var/log/Xorg.0.log to a backup location before you try to start X with the vesa drivers (since the vesa drivers should load correctly, it will overwrite the log with the errors and you need to know what they are, so backup the log with the errors first, is that I'm suggesting). But the first thing we'd have to know is what is the video card, and what are the loaded drivers; then we can work on things like what video options you set in the kernel, and what you've got for modelines and the like in xorg.conf (though I doubt that the issue is modelines, since that just knocks you out fo X completely with a no screens found error, not kills your signal between video card and monitor while leaving X actually running). Anyway, hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Monitor goes off on boot
Ghaith Hachem schreef: On 3/8/06, Holly Bostick [EMAIL PROTECTED] wrote: What video card do you have and what drivers? I've had similar problems (not with this graphical live CD, since I installed Gentoo before it existed, but with X applications and X itself), because I have an ATI card. ATI cards do do that (just shut down the monitor) if 1) using the wrong drivers (radeon when card model is one above the 9(2/5?)50 (sorry, can't remember which model is the stopper for the Open Source drivers), and/or 2) DGA is enabled for the fglrx drivers (this will do exactly what you described; it has many times for me, and it is just one of the many PITAs with the fglrx drivers). i know ati drivers suck but i booted the live cd on my x300se card so i guess it's not really the problem.. I wouldn't bet on that... just because the liveCD boots and loads what seems to be X doesn't mean that the settings that you ultimately wound up with for X are sufficient and correct to load a full desktop environment. The installer has limited needs (so really doesn't need more than the failsafe vesa drivers), and a full DE has many more requirements. Also, the x cards are particularly problematic in terms of X drivers at this time. maybe the screen don't support the resolution set by default (1024 x 768 was the resolution the live cd set by default on my pc and it worked great) Well, you might have a point there (no way to know without more detail), but again, the symptoms of that don't necessarily fit the conditions you're experiencing. It's true that most modern monitors will shut off to save themselves if told to operate in parameters out of their range (for example if I told my monitor to run at 1600x1200 when it only does 1280x1024 with difficulty). But if the resolution is set by default, then auto detection should have told the x setup what the correct parameters for your monitor are and the default setting should be within the correct range. Did you change something, and now the default resolution is *not* 1024x768 (which ought to work for almost any monitor, certainly a current flatpanel)? And secondly, as I said, the usual result of X asking a monitor to display at a higher resolution than the monitor says it can do is *not* signal loss-- it's either a no screens found error (if the higher resolution is the only setting available), or display at a much lower resolution than you asked for (if lower settings are available, X will try them all until it finds one that fit within the monitor's stated parameters of vsync and hres). Signal loss, in my experience (which is admittedly limited, but since most people never experience signal loss at all-- which is as it should be-- even limited experience seems valuable), indicates a very specific range of conditions that might cause it, and they are almost all related to the video card and its drivers and settings, rather than anything to do with the monitor itself (after all, what do you think sends or doesn't send a signal to the monitor :-) ?). What does your Xorg log say (/var/log/Xorg.0.log), and what are your settings for the monitor and video device in /etc/X11/xorg.conf? Also, bizarre as it sounds. is this a new card, or have you been swapping things around inside your case before this install? You might check the seating and connections of the card. Although your situation is slightly different from mne-- you have a flat panel which presumably uses the DVI connector natively, whereas I have a CRT and must use the included DVI-to-VGA converter plug-- I have noticed that several times after modifying something internal to the box itself (whether related to the video card or not), I often had to open the box a second time and reseat the card and/or the monitor connection, which resolved errors I would suddenly be getting for no reason. So you might want to check that as well, though perhaps the connectors have been redesigned since my 9800SE came out (and so are not loosened at the slightest little thing). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Monitor goes off on boot
Daniel da Veiga schreef: The livecd uses the VESA drivers, that should support most if not all cards Yes, I'm sure that's true-- but what I'm wondering is, does the LiveCD use the VESA drivers *no matter what* is returned by autodectection of the video hardware? My question here is because I have installed a lot of distros, and the thing is that as soon as autodetection got an ATI product ID back from the video hardware, I automatically got the radeon drivers. And for me, the radeon drivers are bad news. So I would always have to manually specify vesa anyway (after X broke). I would actually suspect that the same thing happens to nVidia users under most circumstances (they load the nv driver rather than vesa), but of course the nVidia open-source drivers actually work properly, so this is likely not a problem for them. So I'm originally asking if the card might be subject to this condition (since I have experienced signal loss errors due to misconfiguration of an ATI card, which misconfiguration is in some cases due to hardware autodetection mishandling the card when found). so, what you are getting is probably a VSync problem, Xorg tries to probe the monitor for supported resolutions and H/VSyncs, what happened must be that it is using a Vsync not supported by your monitor. That certainly would explain things, but how broken would autodetection then be (or the monitor's EDID?? support) if it couldn't even detect/report its own Vsync range? Not that you might not be right, but that's really scary-- scarier than my theory of autodetect working properly, but mishandling what it detects. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: modules built post kernel install (on the fly)
Harry Putnam schreef: Peter [EMAIL PROTECTED] writes: That is correct. Unless you alter bzImage, modprobe newmodule should work just fine. If your new module is built in, you will need to reload the kernel (reboot). Ok, this is confusing to me... What do you mean by `built in'. I'm thinking the very nature of a module is that it isn't built in. Or do you just mean I'd chose `*' instead of `m' and move bzImage into place in /boot? (Most) kernel modules can be either built into the kernel, or separately from the kernel. Only a very few can only be builtin or only loadable. But whichever they are, they're really all modules-- the kernel is a modular framework, after all, which is why you have to configure it-- to say which kernel modules you want to build, and how you want them built (as builtin to the kernel, or as separate loadable modules). If they're built into the kernel body (*), they're called built-in, in which case they are an integral part of the bzImage, and increase the size of the kernel. Builtins will also always be loaded by the kernel just because they're part of the kernel; this is why you must build certain modules (like for filesystems) as builtins and not as modules, so the kernel has them loaded before it needs them to read the relevant filesystem. If the modules built as dynamically loadable modules (M. which produces little chunks of code-- *.ko files, I think-- in /usr/lib/modules/kernel_version when you run make modules_install), they are called modules (or loadable modules, or dynamic modules). In this case, they 1) do not increase the size of the kernel (because they're not in the bzImage that is the kernel), 2) they are dynamically loadable (modprobe) and removeable (modprobe -r), and may or may not exist at all (because you have the ability to pick and choose which loadable modules you actually want to build in your kernel config). So what Peter meant was that if you add the modules as loadable modules (by choosing M), you wouldn't have to do anything other than make and make modules_install to install the new module (you need to do a make so that the kernel config knows that there's a new module *to* make), but because the body of the kernel has not actually changed (since loadable modules are not compiled into the bzImage like the builtin * modules are), you don't actually have to install the new bzImage, because it's exactly the same as the one you had previously installed. You should be able to modprobe the new module and go right on without rebooting. However, if you compiled the new modules directly into the kernel (by choosing *, or compiling a module that has a sub-function that requires a *), then you would have to install the new bzImage and reboot, because the bzImage (the kernel) has actually changed. Hope this helps explain things, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] radeon and framebuffer
Aggelos schreef: Is it possible to have framebuffer on console with the ATI RV280 graphics card, while also getting 3D hardware acceleration when switching to X? I am asking this because according to http://gentoo-wiki.com/HOWTO_ATI_Drivers we must remove Support for framebuffer devices from the kernel. Aggelos. Looks like you didn't read the entry thoroughly; this quote is below the screenshot of the kernel config specifying no framebuffer support: I have a Radeon 9500/9600 (not sure which) in a Dell Inspiron 8600 and was able to compile support for the framebuffer devices. I used the vesa-tng framebuffer device for getting fbsplash to work. I am running kernel 2.6.14-gentoo-r2. The program 'fglrxinfo' reports that I am getting 3D hardware acceleration. I myself have a 9800SE, and also use vesa-tng for fbsplash. Both fbsplash/framebuffer and fglrx work fine; aside from the fb console background, I can even play mplayer videos in the getty console, which is also a framebuffer operation as far as I know. I have heard that enabling the radeon framebuffer is a bad idea, though (it doesn't work well with the fglrx drivers). But stick to vesa or vesa-tng, and you should have no problems. I haven't, from 2.6.13 kernels up to my current (2.6.15-r7), and with fglrx versions from 8.8.whatever to my current 8.22.5. Hope this helps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Sound... a known method to track down problems
Harry Putnam schreef: cat /proc/asound/cards 0 [V8237 ]: VIA8237 - VIA 8237 VIA 8237 with AD1980 at 0xc800, irq 201 I did install alsa-utils and ran alsaconf. /etc/modules.d/alsa looks like this: alias /dev/mixer snd-mixer-oss alias /dev/dsp snd-pcm-oss alias /dev/midi snd-seq-oss # Set this to the correct number of cards. # --- BEGIN: Generated by ALSACONF, do not edit. --- # --- ALSACONF version 1.0.11rc2 --- alias snd-card-0 snd-via82xx alias sound-slot-0 snd-via82xx # --- END: Generated by ALSACONF, do not edit. --- Should the oss stuff be there? Yes, it should (imo); many apps use ALSA's OSS emulation to play sounds. So it's wise to have it unless you want to investigate every sound app you use to make sure it's ALSA-capable. I've put alsasound into run level 3 and rebooted. On starting kde I still get a dialog message about the sound server being unavailable. No play tools work. (lsmod output at the end) Running the command line util `play' on a wav file shows this: reader play WestMemphisBlues_SonnyBoyWilliamson2.wav ALSA lib confmisc.c:672:(snd_func_card_driver) cannot find card '0' snip ALSA lib conf.c:3960:(snd_config_expand) Evaluate error: No such device ALSA lib pcm.c:2107:(snd_pcm_open_noupdate) Unknown PCM default sox: Failed writing default: cannot open audio device OK, here's the thing. Forgive me if I'm going too far back to basics, but The VIA 8237 is an onboard sound chip. I have one myself. As such, it's controlled by the BIOS. Since you have previously not used sound on this machine, it's within the realm of possibility (imo) that you disabled the sound chip in the BIOS, which -- if that was the case-- would possibly explain why the card is not being found when you actually try to use it. Have you checked, or would you check your BIOS and confirm that the onboard sound is enabled there? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] opengl (?) weirdness in wine
Robert Persson schreef: I am finding that with one particular windows application running under wine the graphics are incredibly slow. My question is: Is this something to do with wine that I just have to live with, or could it be connected to other things on my system, such as the xserver? snip The problems are: 1. Typed text gets to the screen pretty slowly - 200 characters per second or so. However pasted text appears more or less instantly. 2. Windows are drawn very slowly - which is pretty annoying because scribe opens and shuts a lot of windows. 3. when scribe starts, a considerable part of the screen (both inside and outside the application windows) ends up covered in black rectangles. These disappear when something is dragged over them or a button underneath them is clicked. 4. There is a slowdown in opengl which persists even when scribe and all wine-related processes have terminated. Before starting scribe glxgears will clock around 960 fps, whereas afterwards it will only clock up 750 fps or so. I am using the following versions of things: x11-base/xorg-x11-6.8.99.15-r4 app-emulation/wine-0.9.8-r1 x11-drivers/ati-drivers-8.22.5 Robert, I'm sorry to have to tell you this, but there's a high chance that you have just run into one of the reasons that people say the ATI drivers suck. I am also an ATI user, and to be flatly honest, there are some things (effects, usually bad ones) that you *will* get when using an ATI card under Wine or Cedega that you *won't* get while using another brand of video card (not even specifically nVidia, but really just any other brand with decent drivers). I have several programs that I run under Wine/Cedega that behave oddly or run significantly slower than reports of that same program running with nVidia cards. Because the drivers really are not good, and don't do what they are supposed to/expected to in many respects, for one, and because the driver team has not yet reached the point where they're optimizing for speed (they're trying to get the drivers stable first), for another. So if you have another brand of video card available to you, I would suggest you give it a whirl and see if the problem persists. It may not. That said, I suspect you do have an issue with your X server as well... the 6.8.99-blahblahblah series is buggy and unmaintained. I tried it myself and had even more problems than I did with 6.8.2-r6, which I reverted to. So the xserver version you're using may well be excaberating the problems the drivers already have. I am (at this very moment, actually) upgrading to Xorg 7.0, after which I'm going to enable the r300 drivers (I have a 9800SE), and if those don't help, I'm going to see if the fglrx drivers work better under 7.0 than under 6.8.2. If not, I'm buying an nVidia card (my birthday is coming, and I am just sick of this after nearly two years; I've had enough of the does not work with ATI cards list on the Cedega Release notes). It's an option to take under serious consideration (upgrading to 7.0 and switching to the open source drivers). I'll keep you posted on my progress -- meaning, if Wine seems to work better under these conditions. Also consider that Wine is doing a lot of work as they close in on 1.0 (which might be as soon as April if the release schedule holds up), and the devs are specifically working on ATI issues (I lurk on wine-dev, and some of the devs working on wine3d and direct x do have ATI cards). Now I'm sure that scribe doesn't use d3d or OGL (though it might, for all I know), but surely any work on the directx backend will have some effect on things like draw speed and rendering. So there's hope from that end as well. In any case, hope this helps to some extent. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] mysterious segfaults
[EMAIL PROTECTED] schreef: Recently, programs on my computer have been victims of abrupt segfaults. snip The weird thing is, after waiting a while (say two or three hours), the problem went away---everything just started working. This sounds awfully like a heat problem, especially if the problem goes away after waiting a significant amount of time (the system cools down), or only appears after a long time of use (the system heats up). I did have a similar problem (but I knew it was heat, because I have the gkrellm temp monitors displayed), and it was so bad that I could not even run the computer without an external fan pointing into the box (the sides have been off all the time). If I turned the fan off, I got immediate segfaults, and I was unable to even boot the machine. Even with the fan, it was running *extremely* hot (I'm talking about 70 to over 100 degrees Celsius). Finally, I realized that I was insane to even attempt to run like that, and took the machine apart, to discover that the CPU fan was absolutely clogged with dust and stuff (I have a cat, and her hair gets into *everything*. A LOT of hair). Cleaned out the CPU fan, and suddenly the box was running at like 48 degrees C. Without the external fan. It could be some kind of intermittent hardware issue in your case, but heat is easy enough to check, and very well might be the problem. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Firefox OOo oddness
John J. Foster schreef: On Sat, Feb 25, 2006 at 02:59:19AM +0100, Holly Bostick wrote: John J. Foster schreef: Good evening, I'm running a KDE desktop and in the File Open dialog boxes on both Firefox OOo I can't seem to find where to tell them to display hidden folders and/or files. For firefox, if a Save dialog, select Browse for other folders, then right-click inside the file list on the right -- not the folder list on the left-- and select Show hidden files. For the Open dialog, you don't have to browse, obviously. Thanks, that works fine. Even for the Open File, which I'm not sure why you wouldn't think I'd have to browse. Sorry, it was late for me and I was typing fast. What I meant is that for the Save dialog, you have to open the Browse section of the dialog window by opening Browse for other folders, so showing hidden files is two steps, but the Open dialog begins with a Browse window, so you only have one step. However, if you do this a lot, you might consider making a symlink from the hidden folder or file to a link of the same name, but without the preceding ., that way the symlink won't be hidden. Yeah, but that's really lame that I'd have to do that. OOo can be so advanced on some things, and so incredibly stupid on others. Lame? I dunno. Is it 'lamer' than having to type a path or filename in the dialog? Or 'lamer' than Joseph's suggestion (which doesn't work for me using OO.o-2.02_rc2): Joseph schreef: Yes, it is there for OO as well. When you click on Open File, the window will have few icons, click on the available icons and you will find it :-) In my Open dialog (OOWriter), there are 3 button icons: Up one level in the filetree, Make new folder, and Home. This is exactly what I would expect, and these buttons do exactly what I would expect. There is no context menu for them, and if the menu exists on the main window, I don't see it, and I didn't find it in the Preferences screen last night. I don't know about you, but to me, one 10-second task (making a symlink to a folder which is likely not hidden by my choice but by default, such as config files in my /home folder which I often edit), and afterwards knowing that it's /done/ and I don't have to worry about it any more, is much less 'lame' than having to search for 10 minutes for a stupid setting in the program in question (Firefox, as a GNOME program, puts the setting one place, Krusader, as a KDE program, puts the setting another place, and OO.o, as an independent program, puts it $DEITY knows where, if the setting exists at all). But perhaps you have different needs (as a single home user, I don't particularly have any need to hide any files from possible abuse, so while I may work with hidden files often, they aren't mission-critical, voluntarily hidden files). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Gentoo LVM Newbie Question
Alexander Skwar schreef: Okay, so the following fs are online resizable, according to you: - reiser3 - xfs - jfs - ext2, ext3 That's pretty much every FS, isn't it? Yes, but as far as I know from the docs, jfs and xfs can only be *grown* online, not shrunk, which could be a problem depending on your needs. Unless that's changed recently. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge error
Michael Smith schreef: John Fawcett wrote: I'm new to gentoo and I wonder if someone can point me in the right direction to resovle the following kind of problem. I have run emerge mysql and get the following output: Calculating dependencies /!!! Cannot resolve a virtual package name to an ebuild. !!! This is a bug, please report it. (virtual/perl-Storable-2.13) What should I do, is it really a bug, and who should it be reported to? thanks Hmm, I can't find any open bugs on this in bugs.gentoo.org, so I guess it's new. You should register a new account at bugs.gentoo.org and report it. Before you do, make sure you run emerge --sync and see if the problem goes away. It very well might: eix perl-Storable * virtual/perl-Storable Available versions: 2.07 2.13 2.15 Installed: none Homepage:http://www.gentoo.org/proj/en/perl/ Description: Virtual for Storable The point being, that there *is* no other ebuild *but* virtual/perl-Storable, so the fact that it cannot be found/resolved seems very likely to be a sync issue of some sort rather than a bug, /per se/, especially since this is the middle available version. What version of mysql are you attempting to install, John, and what ARCH are you running (emerge info)? Holly -- gentoo-user@gentoo.org mailing list
[gentoo-user] Help with ALSA bug report
Hi all-- For some days now, I've been having a problem with almost all media players and one emulator. Xine 1.1.1-r4 opened, but crashed when closing the splash screen. Totem 1.3.91 (bmg, gstreamer backend) opened, and then crashed immediately. mplayer was OK (but I prefer xine, and naturally was concerned about the other two crashing like that). FakeNES 0.3.1 with Allegro 4.1 (yes, I tried all three versions) crashed with a signal #6 in Allegro (but 0.1.5 with Allegro 4.0.3 worked fine, so I just masked everything above that combination). I ultimately found that the error was the same for all three programs: totem: conf.c:3144: snd_config_iterator_first: Bewering `node-type == SND_CONFIG_TYPE_COMPOUND' mislukt. The same line number, even, but in what program or library's conf.c, was the question. Goolgling revealed that the problem was ALSA, but the only solution I found was to downgrade ALSA. So I did that, masking =media-libs/alsa-lib-1.0.11_rc3 =media-libs/alsa-oss-1.0.11_rc3 =media-sound/alsa-firmware-1.0.11_rc3 =media-sound/alsa-headers-1.0.11_rc3 and then doing an emerge -uaDNtv world to downgrade those packages (to 1.0.10, for the most part). Lo and behold, Totem and Xine suddenly worked again. I unmasked Allegro and FakeNES-0.3.1, and that works too (got FakeNES 0.3.1 open right now). Hurrah!! The problem is, this seems very much to be a bug in some ALSA ebuild (the person whose advice I followed on the ALSA-user's list was also a Gentoo user, also using 1.0.11_rc3), but I don't know which part of ALSA (which package of the four I downgraded I should file the bug against), nor what precisely to file as a bug; can I get away with Alsa-* fails with multiple applications? I mean, basically, all I know is that downgrading some part of ALSA allowed these programs to run when they stopped before-- which is weird enough in itself, since I upgraded ALSA on the 5th of February, and this issue just manifested itself, afaik... I *think* xine and Totem were running up to a couple of days ago. So it could be something else I upgraded that broke ALSA, but that takes me completely out of my depth in terms of bug-testing. I seem to have found a bug, and it seems to be related to ALSA, but may be triggered by something else. Is anybody else seeing this kind of issue, first of all, and second of all, does anybody know how I can determine where the bug actually is so I can submit it? Thanks for any help, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Firefox OOo oddness
John J. Foster schreef: Good evening, I'm running a KDE desktop and in the File Open dialog boxes on both Firefox OOo I can't seem to find where to tell them to display hidden folders and/or files. For firefox, if a Save dialog, select Browse for other folders, then right-click inside the file list on the right -- not the folder list on the left-- and select Show hidden files. For the Open dialog, you don't have to browse, obviously. Setting will remain until you revoke it, though possibly not across browser sessions (not sure). There does not seem to be a global setting for OO.o. Of course, if you know the filename, you can just type it once in the correct folder; this also works for folders. In the case of a folder, this will take you to the folder, where you can then select the file. However, if you do this a lot, you might consider making a symlink from the hidden folder or file to a link of the same name, but without the preceding ., that way the symlink won't be hidden. Also, in Firefox the only columns available in the File dialog boxes are Name and Modified. I know size was there in 1.0.7. Yes, apparently Name and Modified are the only columns available. Things change. 1.5 is a long way from 1.0.7. Myself, I don't need the filesize in a web-browser save dialog (I find Modified date much more useful), but everybody's different. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] alsactl store wont save alsamixer
Allan Spagnol Comar schreef: Hi all, I am having a bit of trouble trying to save alsamixer settings. I had tried every little thing I know and could make it to work, if I reboot my system alsamixer settings go all to mute :( rc-update add alsasound default The 'alsasound' daemon saves your settings on exit (poweroff), and more importantly, reloads them at boot. 'Alsactl store' doesn't help much if you don't somehow account for 'alsactl restore' :-) . Alsasound is the easiest way to do this, but you can just get your session manager to run the command for you, if you prefer. Hope this helps, Holly P.S. also-- if you're using KDE and KMix, make sure that you don't have KMix overriding the alsa mixer settings (or unmute the KMix settings). H. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Re: Problems with GRUB in the installation of Gentoo
Neil Bothwick schreef: On Thu, 16 Feb 2006 23:33:15 +0100, Holly Bostick wrote: This functionality is not in any way a hack: equery belongs /sbin/installkernel [ Searching for file(s) /sbin/installkernel in *... ] sys-apps/debianutils-2.15 (/sbin/installkernel) It's part of system. # emerge -C debianutils !!! 'sys-apps/debianutils' is part of your system profile. !!! Unmerging it may be damaging to your system. I suspected that (since it would be eminently logical for the package that provides the 'make install' command to be on the system by default), but as I said in an unquoted part of that mail I don't know how to check what's in system. Is there another way to get a list of packages that comprise the system profile (other than just trying to unmerge one and getting a warning)? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Kernel not updating...
gentuxx schreef: Daniel da Veiga wrote: On 2/16/06, gentuxx [EMAIL PROTECTED] wrote: Forgive me if this ends up being the stupid question of the day. But, I haven't been able to upgrade my kernel for some reason. (Obviously I could download it from kernel.org and go that route, but I would like to keep as much as possible in portage.) The vanilla kernel is in portage as well (vanilla-sources). So I run an `emerge --sync`, and `emerge -av gentoo-sources`. And it wants to rebuild my 2.6.13-r3 Hmmm, shouldn't it be emerge -uav gentoo-sources ? I mean, u for update? emerge -uav gentoo-sources snip? Nothing to merge; do you want me to auto-clean packages? [Yes/No] n Quitting. Have you checked /etc/portage/package.mask? About the only reason left is that you manually masked versions above the one you're now using, and forgot about it (hey, it happens). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Problems with GRUB in the installation of Gentoo
Michael Kintzios schreef: -Original Message- From: Neil Bothwick [mailto:[EMAIL PROTECTED] Sent: 16 February 2006 16:10 To: gentoo-user@lists.gentoo.org Subject: Re: [gentoo-user] Re: Problems with GRUB in the installation of Gentoo make install does exactly the same, and sets up the vmlinuz and vmlinuz.old symlinks to point to your new and previous kernel respectively, so you don't need to edit grub.conf. Hmm, it doesn't on my two boxen. :-( I do not have a vmlinuz, System.map and config links. Do I have to first set up the symlinks manually? Looking as /sbin/installkernel, it doesn't appear that you have to create the links. Of course, you do have to make sure /boot is mounted first :) Yep, /boot is always mounted (just to be sure I won't forget it, I always mount it before I even cd into /usr/src/linux). Running make make modules_install does *not* create any links in my /boot directory, Neither make nor make modules_install is *supposed* to create any links in your /boot directory. Make compiles the kernel and static modules, make modules_install compiles the dynamic modules and installs them in /lib/modules/kernel-version/wherever_they_belong. This is why you then have to either copy the compiled kernel (and supplemental files like .config and system.map, if you want) to /boot. make install (which is a different command than make modules_install) does this automatically, and additionally makes symlinks called vmlinuz and vmlinuz.old (if a previous kernel exists), and updates lilo (if available). If you use grub, if grub.conf is set to load vmlinuz as the current kernel, and vmlinuz.old as the previous kernel, you don't then have to update grub in any way when you have used make install to install your newly-compiled kernel, because make install changes the symlink for the previous kernel to vmlinuz.old, and makes the symlink for the newly installed kernel vmlinuz. Kernels older than the current and most-recent-previous can be added manually to the menu directly. My impression is that you haven't yet run make install. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Problems with GRUB in the installation of Gentoo
Frino Klauss schreef: On 2/16/06, Holly Bostick [EMAIL PROTECTED] wrote: My impression is that you haven't yet run make install. Can't you use genkernel instead ? I have no idea; I've never used genkernel, and am unlikely to ever do so. Since it is a mostly automated process (though you can use some sort of switch to go to a manual menuconfig), I suppose it would compile and install the kernel to /boot as part of its functioning, but I have no idea what it actually does, or how to make it do what it does. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Re: Problems with GRUB in the installation of Gentoo
Mick schreef: John Jolet wrote: On 2/16/06 11:05 AM, Michael Kintzios [EMAIL PROTECTED] wrote: From: Neil Bothwick [mailto:[EMAIL PROTECTED] make install does exactly the same, and sets up the vmlinuz and vmlinuz.old symlinks to point to your new and previous kernel respectively, so you don't need to edit grub.conf. Hmm, it doesn't on my two boxen. :-( I do not have a vmlinuz, System.map and config links. Do I have to first set up the symlinks manually? Yep, /boot is always mounted (just to be sure I won't forget it, I always mount it before I even cd into /usr/src/linux). Running make make modules_install does *not* create any links in my /boot directory, ever. Could it be that there's something wrong with my system(s) - at least three installations have always behaved like this . . . No, make makes the bzImage under /usr/src/linux.arch/.boot And make modules_install installes the modules in /lib, I think. Neither of them touches /boot. I THINK there's a make install that's supposed to do some stuff for you, but I prefer to do those steps by hand. I am confused: how many 'make install's are there? Don't they 'all' do the same? Are we talking about a customised (hacked) make install here? There is only one make install. It is the optional command out of the set of commands used to compile and install a kernel: make (which now includes make modules, whereas under 2.4-series kernels you had to run make modules as a separate command) -- compiles the kernel and modules make modules_install -- installs dynamic (loadable) modules to /lib/modules/kernel-version/ and you can then either copy the kernel manually to /boot, or use the make install command to do so. This functionality is not in any way a hack: equery belongs /sbin/installkernel [ Searching for file(s) /sbin/installkernel in *... ] sys-apps/debianutils-2.15 (/sbin/installkernel) Programs That Depend On debianutils app-admin/sysklogd app-portage/gentoolkit app-text/noweb dev-util/bazaar dev-util/larch dev-util/pbuilder dev-util/tla mail-filter/clamassassin sys-apps/apmd sys-boot/arcboot x11-misc/sux x11-wm/fvwm And it's not in my world file, so it was probably installed by gentoolkit or sux, which are the only two RDEPENDS of this list I have installed (other than fvwm, but that's recent and I'm sure that I had the ability to make install long before that). So I suppose it's possible that some users might not have this, unless it's part of system, which I don't know how to check. But any utility ported from Debian certainly can't be considered a hack by any stretch of the imagination Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] lm_sensors problem
Jarry schreef: OT Is it possible to install lm_sensors with gentoo-kernel too Yes, it is. I tried with gentoo-sources, but got this error: _ * Determining the location of the kernel source code * Found kernel source directory: * /usr/src/linux * Found sources for kernel version: * 2.6.15-gentoo-r1 * getfilevar requires 2 variables, with the second a valid file. * getfilevar VARIABLE CONFIGFILE * Could not find a usable .config in the kernel source directory. * Please ensure that /usr/src/linux points to a configured set of Linux sources. * If you are using KBUILD_OUTPUT, please set the environment var so that * it points to the necessary object directory so that it might find .config. !!! ERROR: sys-apps/lm_sensors-2.9.2 failed. !!! Function linux-info_pkg_setup, Line 537, Exitcode 1 !!! Unable to calculate Linux Kernel version !!! If you need support, post the topmost build error, NOT this status message. This error means that your /usr/src/linux symlink points to unconfigured kernel sources. Many, if not all, external packages that compile against the kernel have this requirement. You don't have to make (compile) or install the kernel *right now* (although you will obviously want to do that asap), but you do have to run some type of config (make menuconfig, make xconfig, whatever) and save a configuration with proper support for the sensors before lm_sensors will compile the modules. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] ALSA + CMI8738 card set up
Rohit Sharma schreef: Hi list, This is about my struggle to make my Trust 514DX soundcard work with Linux. Well, I don't have this specific card, but I do have a Typhoon Acustic 6, which is the same chipset: 00:0e.0 Multimedia audio controller: C-Media Electronics Inc CM8738 (rev 10) and it works fine, so I'll do my best to help. This is expected to be supported in Linux [alsa-project.org] and I have read other testimonials on the net. Yes, this chipset is supported for a long time now (it's very common, and by no means new). I admit that I have never had to manually set anything alsa/oss as all my previous soundcards worked very easily with default settings of the modules. Well the most likely problem is that your settings are incorrect in the kernel, rather than anywhere else. The kernel settings are the foundation; if they're right, everything will Just Work, if they're wrong, nothing will work. That's just the long and the short of it, pretty much. But we'll get to that shortly. Alsa is started as a service on my comp. I thought it may just be that a newer-than-in-kernel code may help. The error that I have, from say Mplayer, is as follows. . Checking audio filter chain for 48000Hz/2ch/s16le - 48000Hz/2ch/s16le... AF_pre: 48000Hz/2ch/s16le [AO OSS] audio_setup: Can't open audio device /dev/dsp: No such file or directory alsa-init: 1 soundcard found, using: default ALSA lib pcm_dmix.c:819:(snd_pcm_dmix_open) unable to open slave alsa-init: playback open error: No such file or directory Just for comparison: mplayer Battlestar\ Galactica\ 01x02-xvid.avi MPlayer 1.0pre7try2-3.4.5 (C) 2000-2005 MPlayer Team CPU: Advanced Micro Devices Athlon MP/XP Thoroughbred (Family: 6, Stepping: 1) Detected cache-line size is 64 bytes CPUflags: MMX: 1 MMX2: 1 3DNow: 1 3DNow2: 1 SSE: 1 SSE2: 0 Gecompileerd voor x86 CPU met de volgende uitbreidingen: MMX MMX2 3DNow 3DNowEx SSE 85 audio 196 video codecs Bezig met het afspelen van Battlestar Galactica 01x02-xvid.avi Cache fill: 16.80% (1409024 bytes)AVI bestandsformaat gedetecteerd! VIDEO: [XVID] 640x352 16bpp 25.000 fps 1031.8 kbps (126.0 kbyte/s) Clip info: Software: Nandub v1.0rc2 == Bezig met het openen van de audio decoder: [mp3lib] MPEG layer-2, layer-3 mpg123: Can't rewind stream by 291 bits! AUDIO: 48000 Hz, 2 ch, s16le, 112.0 kbit/7.29% (ratio: 14000-192000) Selected audio codec: [mp3] afm:mp3lib (mp3lib MPEG layer-2, layer-3) == vo: X11 running at 1280x1024 with depth 24 and 32 bpp (:0.0 = local display) It seems there is no Xvideo support for your video card available. Run 'xvinfo' to verify its Xv support and read DOCS/HTML/en/video.html#xv! See 'mplayer -vo help' for other (non-xv) video out drivers. Try -vo x11 == Bezig met het openen van de video decoder: [ffmpeg] FFmpeg's libavcodec codec family Selected video codec: [ffodivx] vfm:ffmpeg (FFmpeg MPEG-4) == Checking audio filter chain for 48000Hz/2ch/s16le - 48000Hz/2ch/s16le... AF_pre: 48000Hz/2ch/s16le ==AO: [oss] 48000Hz 2ch s16le (2 bps) Building audio filter chain for 48000Hz/2ch/s16le - 48000Hz/2ch/s16le... Start afspelen... VDec: vo config aanvraag - 640 x 352 (csp voorkeur: Planar YV12) VDec: using Planar YV12 as output csp (no 0) Film-Aspect is 1.82:1 - voorscalering naar het correcte film-aspect. VO: [x11] 640x352 = 640x352 Planar YV12 SwScaler: using unscaled Planar YV12 - BGRA special converter [mpeg4 @ 0x860ed30]frame skip 8t: 0.000 1/ 1 ??% ??% ??,?% 0 0 9% [mpeg4 @ 0x860ed30]frame skip 8t: 0.004 2/ 2 ??% ??% ??,?% 1 0 99% No bind found for key MOUSE_BTN0 0.9% 6 0 91% A: 9.4 V: 9.4 A-V: -0.000 ct: 0.029 235/235 23% 15% 0.9% 6 0 89% Bezig met afsluiten... (Stop) Just so that you know that it does work when your sound modules are properly configured. Your problem is here: [AO OSS] audio_setup: Can't open audio device /dev/dsp: No such file or directory You have apparently not enabled OSS audio emulation in alsa (kernel) driver, and this is what creates the /dev/dsp device (or rather the sound sequencer device; /dev/dsp is a symlink to /dev/sound/dsp la /dev/dsp lrwxrwxrwx 1 root root 9 feb 9 01:27 /dev/dsp - sound/dsp which is the actual sequencer device). At the same moment, other config details and system's state are as below. [EMAIL PROTECTED] ~ $ lsmod | grep snd snd_cmipci 20484 0 snd_opl3_lib 6688 1 snd_cmipci snd_hwdep 6116 1 snd_opl3_lib snd_mpu401_uart4416 1 snd_cmipci snd_rawmidi 15620 1 snd_mpu401_uart You seem to be missing a lot of modules; compare: lsmod | grep snd snd_seq_midi6176 0 snd_opl3_synth 12292
Re: [gentoo-user] ALSA + CMI8738 card set up
Rohit Sharma schreef: Quick questions - Holly. When I configure alsa support in my kernel and build modules using menuconfig and tell kernel to load modules automatically [or do manual loading using modules.autoload.d] 1. Do I still have to enable alsasound service using rc-update add alsasound default ? Yes; the service does the important job of storing your mixer levels on shutdown and restoring them on boot, as well as actually loading the modules for use. The idea of a loadable module is that the module is loaded when something needs it, rather than just loading because it is available. This is why you have to, for example, compile the modules for your / filesystem statically-- because they have to be already loaded before you need them in the boot process (statically-compiled modules are loaded in the very first initializing kernel instructions). Loadable modules, on the other hand, are loaded when something asks for them, like an initialization script (network initialization script to 'need' the network card drivers, X to need the fglrx drivers, firestarter to need the various ip_filter modules, and so forth). Otherwise, you'd have to modprobe the modules before attempting to start the program that needed them to be already in operation before the program could work (like the boot process, which has to read filesystems on a drive and can't stop to load the modules for the filesystem-- which modules, if compiled as loadable modules rather than statically, are themselves on the filesystem that the boot process cannot read because the modules aren't loaded-- before it can read the filesystem to manage the files it needs on that filesystem in order to continue the boot process). So you need something to need the sound modules to load, or else they won't, and that's what alsasound does, too. 2. Is it a good idea to let the kernel automatically load modules or use /etc/modules.autoload.d/kernel-2.6 for it? [ Currently I am using the file instead of kernel [I disabled my onboard i8x0 card in BIOS to install this CMIPCI one and was skeptic that kernel would pick up the onboard one. I guess for a correctly configured host, it woudn't matter]. Well, if you want to do more work when it isn't necessary, you're free to use modules.autoload.d to manage the modules, but you then have to maintain that file. For no reason. Alsasound contacts the kernel to load the modules just fine as long as you've given it a correct range of modules to access. As for your onboard sound, I was afraid of it loading as well when I disabled it and added a new sound card, but the kernel is pretty smart, and just ignored it, especially since I removed all reference to the driver/modules in my ALSA configuration files, and stopped compiling modules for it in the kernel. After all, the kernel can't load a module that doesn't exist :-) . If for some reason that doesn't work for you, you can add the onboard sound modules to /etc/hotplug/blacklist, but that shouldn't be necessary. Really, the kernel does know what it's doing rather well these days; you can definitely give it the benefit of the doubt and trust it first, assuming that it's going to do what it's supposed to, rather than suspecting it of not being competent and assuming you have to manage it doing its job. To say the least your response was encouraging. I am in the middle of an emerge after two months of disconnectivity Glad to hear it. Good luck with the next steps. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Problems with GRUB in the installation of Gentoo
Gilberto Martins schreef: Hi again !!! Perhaps you should post the output of: #ls -l /boot #cat /boot/grup/grub.conf There it goes: livecd / # ls -l /boot total 2231 lrwxrwxrwx 1 root root 1 Feb 11 09:22 boot - . -rw-r--r-- 1 root root 32414 Feb 12 16:06 config-gentoo-2.6.12-gentoo-r10 drwxr-xr-x 2 root root1024 Feb 12 13:16 grub -rw-r--r-- 1 root root 2227076 Feb 12 16:01 kernel-gentoo-2.6.12-gentoo-r10 drwx-- 2 root root 12288 Feb 10 22:44 lost+found Ummm... isn't there supposed to be a system.map for the kernel as well? I myself don't manually copy my kernels after compiling it; I use make install to do so, and I have the following files in /boot for all my kernels: la /boot totaal 22M -rw-r--r-- 1 root 512 apr 27 2005 14:23 backup_mbr lrwxrwxrwx 1 root1 apr 27 2005 18:52 boot - . lrwxrwxrwx 1 root 20 jan 07 2006 02:44 config - config-2.6.15-gentoo -rw-r--r-- 1 root 60K mrt 24 2005 01:31 config-2.6.11.4-20a-default -rw-r--r-- 1 root 33K okt 13 2005 15:06 config-2.6.13-gentoo-r4 -rw-r--r-- 1 root 35K nov 07 2005 18:40 config-2.6.14-gentoo -rw-r--r-- 1 root 35K jan 06 2006 21:34 config-2.6.14-gentoo-r7 -rw-r--r-- 1 root 36K jan 07 2006 02:44 config-2.6.15-gentoo lrwxrwxrwx 1 root 23 jan 06 2006 21:34 config.old - config-2.6.14-gentoo-r7 drwxr-xr-x 2 root 1,0K jan 15 2006 06:06 grub lrwxrwxrwx 1 root 27 apr 27 2005 14:23 initrd - initrd-2.6.11.4-20a-default -rw-r--r-- 1 root 1,4M apr 27 2005 14:23 initrd-2.6.11.4-20a-default -rw-r--r-- 1 root0 apr 29 2005 00:47 .keep drwx-- 2 root 12K apr 27 2005 14:13 lost+found -rw-r--r-- 1 root 135K apr 27 2005 14:23 message -rw-r--r-- 1 root 121 mrt 24 2005 01:31 README.vmlinux-2.6.11.4-20a-default.gz -rw-r--r-- 1 root 67K mrt 24 2005 01:32 symvers-2.6.11.4-20a-i386-default.gz lrwxrwxrwx 1 root 24 jan 07 2006 02:43 System.map - System.map-2.6.15-gentoo -rw-r--r-- 1 root 693K mrt 24 2005 01:26 System.map-2.6.11.4-20a-default -rw-r--r-- 1 root 948K okt 13 2005 15:06 System.map-2.6.13-gentoo-r4 -rw-r--r-- 1 root 736K nov 07 2005 18:40 System.map-2.6.14-gentoo -rw-r--r-- 1 root 736K jan 06 2006 21:34 System.map-2.6.14-gentoo-r7 -rw-r--r-- 1 root 748K jan 07 2006 02:43 System.map-2.6.15-gentoo lrwxrwxrwx 1 root 27 jan 06 2006 21:34 System.map.old - System.map-2.6.14-gentoo-r7 drwx-- 2 root 1,0K apr 29 2005 15:52 .Trash-root lrwxrwxrwx 1 root 21 jan 07 2006 02:43 vmlinuz - vmlinuz-2.6.15-gentoo -rw-r--r-- 1 root 1,4M mrt 24 2005 01:26 vmlinuz-2.6.11.4-20a-default -rw-r--r-- 1 root 3,6M okt 13 2005 15:06 vmlinuz-2.6.13-gentoo-r4 -rw-r--r-- 1 root 3,7M nov 07 2005 18:40 vmlinuz-2.6.14-gentoo -rw-r--r-- 1 root 3,7M jan 06 2006 21:34 vmlinuz-2.6.14-gentoo-r7 -rw-r--r-- 1 root 3,4M jan 07 2006 02:43 vmlinuz-2.6.15-gentoo lrwxrwxrwx 1 root 24 jan 06 2006 21:34 vmlinuz.old - vmlinuz-2.6.14-gentoo-r7 (ignore the 2.6.11.4-20a kernels, those are from my dual-boot SUSE install.) In any case, for each available kernel, make install copies 3 files (and makes 3 symlinks): config-kernel.version system.map-kernel.version vmlinuz-kernel.version the config file is just a convenience, but the vmlinuz and system.map are, afaik, required to boot the kernel. I don't know where system.map is copied from (the original compiled file name), but if I was you (which I of course am not), I would cd to /usr/src/linux and run make install to copy the kernel files properly to boot, because I suspect this is not happening however you're doing it. Just my €0.02 Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OpenOffice / Evolution display settings
Franta schreef: Hi all snip Openoffice has now display settings for a (nearly) blind person. The icons in the taskbar aren't big. They are huge. I'd like to change some font settings in Evolution too. . Can anyone tell me which part of Gnome to install to be able to change the display of the Not a part of GNOME, but a loose gtk application: gtk-chtheme (which allows for changing the GTK2 default font, unlike some other similar loose applications). You're probably thinking of control-center (the GNOME Control Center), but that would install most if not all of GNOME, so there's no point in that, when gtk-chtheme will do the job. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Problems with GRUB in the installation of Gentoo
John Jolet schreef: Ummm... isn't there supposed to be a system.map for the kernel as well? I myself don't manually copy my kernels after compiling it; I use make install to do so, and I have the following files in /boot for all my kernels: I've never done anything with a system.map. I manually copy it myself to allow me to name them whatever I want. Well, that's my point, sort of... what exactly do you copy, and has that file been copied to Gilberto's /boot folder? From my /boot listing previously, you can see that even SUSE creates a system.map in the /boot folder, and that's a precompiled kernel (so it's not like it's copying manually or via make install). So I kinda suspect that it's a needed file across all distros, whatever it may be called and, looking in /usr/src/linux, it is a separate file from the bzImage file, which is the actual compiled kernel. The fact that the make install command also finds it necessary to copy this file from /usr/src/linux to /boot is not to be sneezed at either, imo. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Problems with GRUB in the installation of Gentoo
CapSel schreef: On 2/13/06, Holly Bostick [EMAIL PROTECTED] wrote: Ummm... isn't there supposed to be a system.map for the kernel as well? snip OT?? AFAIK system.map is not needed for lilo and not for grub. I don't have it and all works without any errors or warnings about it. No, it's not OT, it was a question (is the absence of system.map in Gilberto's /boot folder possibly relevant to the problems booting the kernel?) which you, among others, have answered (No). Thanks. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OpenGL X desktop
Daevid Vincent schreef: Whoa. Check out this demo of a new OpenGL X desktop. IMHO it blows away the Mac OS/X http://www.freedesktop.org/~davidr/xgl-demo1.xvid.avi Anyone have any other info on when this will be available? How does it work? It looks like they're just using Gnome. Will it work with KDE too? I want it now! Oh. My. God. (thanks for the heads-up, Daevid, certainly that video was worth seeing!) More questions... what kind of a rig is that (will it run on my fairly ordinary box? Will it run with the *%%^ ati drivers?)? Oh. My. Heck, I'd run GNOME again if that was the only way to get that, I don't bloody care. I like GNOME anyway. *Finally*, eye-candy worthy of the name!!! Oh, how many people could I begin converting to Linux if they saw my desktop doing *that*...?!!? Bite me, Vista!!! Holly, off to the forums to read the giant thread. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Stuch on k3b at dep install of Transcode
Harry Putnam schreef: I'm trying to install k3b. One of its dependancies is transcode. The emerge stopped at transcode with specific errors, When checking bugzilla I see this is a known problem. I remember having this problem way back when (I mean, several k3b versions ago), but haven't had it in at least 3 upgrades: motub - emerge -pv transcode k3b These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild R ] media-video/transcode-1.0.2-r2 USE=3dnow X a52 dv dvdread extrafilters fame ffmpeg gtk imagemagick jpeg mjpeg mmx mp3 mpeg network ogg quicktime sdl sse theora truetype v4l2 vorbis xml2 xvid -lzo -sse2 0 kB [ebuild R ] app-cdr/k3b-0.12.10 USE=alsa css dvdr encode ffmpeg flac hal mp3 musepack musicbrainz sndfile vcd vorbis -arts -debug -kde -xinerama 0 kB Istr that the issue was related to the encode USE flag, which I used to disable, since I don't actually re-encode things in K3b anyway, but as you see, I now have it enabled, and it works fine (meaning that both K3b and transcode are properly installed, since I still don't re-encode things with K3b). What are the specific errors that you are receiving? Perhaps the problem is your transcode USE flags; what are those? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Getting udev to create tun devices
Robin Atwood schreef: I need a device /dev/net/tun to use with hercules. tun is defined in the kernel and the traditional mknode method works fine but I loose it after a reboot. After a bit of research I added: # tun device for hercules KERNEL==tun, NAME=net/tun to /etc/udev/50-udev.rules but it doesn't seem to hack it. :( Anybody any ideas? TIA -Robin. Hey, Robin, I don't know what, if anything, is wrong with your rule, but I'm almost sure you're adding it to the wrong file. (from a Linux Format article about udev): - Making up the rules The rules are contained in files in /etc/udev/rules.d. The default file is usually called 50-udev.rules. Don’t change this file as it could be overwritten when you upgrade udev. Instead, write your rules in a file called 10-udev.rules. The low number ensures it will take priority over any definitions in the default file. - Basically, the 50-udev.rules file is the system default rules (managed by Portage, in that if udev is updated, this file will also be updated), local changes should go in 10-udev.rules (which is unmanaged by Portage, I believe). Anyway, it's possible that not using a higher-priority rules file to create this rule is part of your problem, either because another rule in the 50 file overrides it, or because your rule is being removed everytime you update udev, which has been happening a lot lately. Hope this helps until somebody with more familiarity with that particular device and custom udev rules comes along. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Remove?
andrew turner schreef: Quoting osv [EMAIL PROTECTED]: Hi all! It's at least one week that I try to unsubscribe from the list, but it seems they're not considering me at all! I sent various empty mails (and even one or two with something written in) to [EMAIL PROTECTED] I seem to remember a post a while back saying that you needed to send an email to lists.gentoo.org, not gentoo.org. That was me in that thread, saying /maybe/ that was the problem. But Andrea Barsini (list admin) came by and said that the regular address (as given above) should work fine. So, no idea what the problem might be. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: kdm 3.5.1, broken ebuild?
Andrei Slavoiu wrote: --- Steven S. [EMAIL PROTECTED] wrote: Ah-hah! Thank you for pointing that out. I had -fomit-frame-pointers, I'm surprised nothing died before kdm. That's strange, there shouldn't be any way that -fomit-frame-pointers could break compilation. I've been using it for years without any problem (and compiled kde 3.5.1 using it too). Andrei, I thought Steven meant that he originally had -fomit-frame-pointers and for whatever reason removed it-- but not all of it (he left the 's' of 'pointers' in error; one too few deletes, as it were). My understanding was that he was surprised that the remaining, invalid, 's' in his CFLAGS didn't cause anything to fail to compile before kdm. And that is surprising, but hey, a lucky break is a lucky break. But that was just my understanding of what Steven said. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] php 4.4.1-r3 wont emerge
Michael George wrote: I am trying to update my system, and it seemed like a good time to go from the old php ports to the dev-lang ports. But I'm having trouble building php-4.4.1-r3. I have masked php-5 and up, as I'm not quite ready to move to that yet. When I try to emerge php-4.4.1-r3, it stops with this error: QA Notice: USE Flag 'nis' not in IUSE for dev-lang/php-4.4.1-r3 I get the same error with php-5.0.5. I don't have nis set as a use flag anywhere in my stuff, though, and I cannot find it set for php in the profiles. I have put -nis into package.use for php and I have tried removing nis and setting -nis into /etc/make.conf, but nothing helps. Did you do an emerge -uaDNtv world after you put -nis in make.conf, to recompile anything with the USE flag previously active? From looking at the ebuild, it looks like this is due to php-4 inheriting the php-4_4.sapi eclass; contained in the eclass is this line: # Recode is not liked confutils_use_conflict recode mysql imap nis Now, I don't know what this means, but it clearly is intended in some way. I also would suspect that the nis flag is not related to PHP itself, but one of its dependencies, which was compiled previously with this flag set, which has now become 'illegal' (that happened to me the other day with another package, the dependency had to be recompiled without a particular flag set for the upgrade of the main package to install). Here's the list of packages that have this flag: equery hasuse -p nis [ Searching for USE flag nis in all categories among: ] * installed packages * Portage tree (/usr/portage) [I--] [ ] x11-libs/qt-4.1.0-r2 (4) [-P-] [ ~] x11-libs/qt-4.1.0-r1 (4) [-P-] [ ] net-mail/lbdb-0.30 (0) [-P-] [ ~] net-mail/lbdb-0.31 (0) [-P-] [ ] mail-mta/postfix-2.2.5 (0) [-P-] [M~] mail-mta/postfix-2.2.4 (0) [-P-] [M~] mail-mta/postfix-2.2.7 (0) [-P-] [ ~] mail-mta/postfix-2.2.8 (0) [-P-] [ ~] x11-libs/qt-embedded-3.3.4-r1 (3) [-P-] [ ] sys-libs/pam-0.78-r2 (0) [-P-] [ ] sys-libs/pam-0.78-r3 (0) [-P-] [ ~] sys-libs/pam-0.78-r4 (0) [-P-] [M~] mail-mta/postfix-2.2.8-r1 (0) [-P-] [ ~] www-servers/thttpd-2.25b-r3 (0) [-P-] [ ~] www-servers/thttpd-2.25b-r2 (0) [-P-] [M~] mail-mta/postfix-2.2.7-r1 (0) [-P-] [M~] mail-mta/postfix-2.2.5-r1 (0) [-P-] [M-] x11-libs/qt-4.1.0 (4) [-P-] [ ] mail-mta/exim-4.43-r2 (0) [-P-] [ ~] mail-mta/exim-4.60 (0) [-P-] [M~] mail-mta/exim-4.50-r999 (0) [-P-] [ ] mail-mta/exim-4.54 (0) [-P-] [ ~] mail-mta/exim-4.52 (0) [-P-] [ ~] mail-mta/exim-4.50 (0) [-P-] [ ~] x11-libs/qt-4.0.1 (4) [-P-] [M~] x11-libs/qt-embedded-3.3.5 (3) [-P-] [ ] x11-libs/qt-embedded-3.3.4 (3) [-P-] [ ~] mail-mta/exim-4.50-r2 (0) [-P-] [ ] mail-mta/exim-4.50-r1 (0) I don't have anything to do with PHP myself, but if one of these is associated with your install of PHP, I would consider recompiling it without the nis USE flag, and then see if PHP compiles. Hope this is helpful, despite my ignorance of this specific package. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Crossover Office demo?
Alan E. Davis schreef: I wish to install crossover office so I can use various animations, primarily, that rely on plugins, as well as the odd Windoze software package. Noting there is an ebuild for amd64, I had attempted to install a demo copy to check if it will work, without success. I don't want to pay for a copy that I won't know whether it will work, if it won't work. Is there a way to do this? Alan Davis I would say, just install the demo as it is and don't worry about Portage. The demo is time-limited (30 days, after which it stops working). If in that time you decide you want to buy it, then do so, uninstall the demo and reinstall the full version (which your purchase will enable you to download, if it is in fact a different download) with Portage. If you don't want it, then uninstall the demo-- no harm, no foul. Both you and Portage can handle being a bit out-of sync on one package, for a month or less, I'm sure :-) HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Viewing SVG/SVGZ files.
Abhay Kedia schreef: Thanks for the ksvg idea and will be reall greatful if firefox worked as well. The only thing I can thinl about this is: what version of Firefox are you using (literally, which ebuild). The early versions of Firefox (1.0.x to approx 1.5-r4) used the mozsvg USE flag to enable firefox to load SVG images. However, the Changelog says: *mozilla-firefox-1.5-r4 (23 Dec 2005) 23 Dec 2005; Jory A. Pratt [EMAIL PROTECTED] +mozilla-firefox-1.5-r4.ebuild: canvas and svg now default and indeed, the mozsvg USE flag is no longer available to firefox, but my firefox seems to render svg images fine. I tested by going to http://www.linuxrising.org/svg_test/test.html and opening one of the images in a new tab. http://www.w3.org/Graphics/SVG/Test/20030813/svggen/styling-css-01-b.svg I'm using Firefox 1.5-r9 (self-compiled, not the -bin). So I'm not quite sure what your problem is, but it seems that it must relate in some fashion to the specific version of FF (maybe you're using a version that needed the USE flag, but you don't have it enabled). HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Xine-ui masked? [Solved]
Daniel Pielmeier schreef: Thank you, in one portage overlay i use there are three ebuilds snip and they are all marked ~x86. Ebuilds from overlay are always ~arch, afaik. Overlays are not considered stable for what seem to me to be obvious reasons; namely that only the Portage tree itself can mark ebuilds as stable, and since these ebuilds are not part of the legal Portage system, who is supposed to judge? and secondly, that overlay ebuilds, which could come from anywhere, could never out of the box be considered stable -- as in working-- by the official Portage tree (which is really just another way of saying the first part). Anyway, I have removed the ~x86 media-video/xine-ui-0.99.4-r3.ebuild in the overlay manually and at the moment i emerge media-video/xine-ui-0.99.4-r3. as far as I know, this is the correct solution; the other point of an overlay is that you keep an eye on it, and if an overlay ebuild does in fact make it into the regular Portage tree, you should remove the overlay build from the overlay and let Portage take over (since, if I recall something that Neil said some time ago, overlay ebuilds will always supercede Portage ebuilds if they duplicate each other, version-wise). That's what works for me, anyway. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user]
Steven S. schreef: On Mon, 6 Feb 2006, Hemmann, Volker Armin wrote: well, he has a point - he just missed to type it: -signatures longer than 4 lines are considered offensive -double signatures are considered offensive -triple signatures like yours are even worse. Erm... offensive? Maybe annoying but I can't think of anyone being offended. Well, insofar as it suggests a certain arrogance to consider one's own wisdom so deathless that you'd attach it to every mail, even if it's longer than the actual content of your response, thereby forcing the recipient to read it (or at least accept a larger mail in their mailbox despite any possible bandwith or message size hit that might entail for the recipient), and insofar as a suggestion of arrogance does tend to offend people (who do you think you are??!!), then Volker has a valid point, though it might be considered more in the rather sensitive group of possible responses, rather than the average Joe, or indifferent. But some people are, after all, rather sensitive, and they can be that. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Xine-ui masked? [Solved]
Mariusz Pękala schreef: Portage overlays are our 'private' portage trees, and they are in no thing 'worse' than 'official' ebuilds. You clearly haven't seen, or even imagined, some of my initial attempts at ebuild writing, which naturally resided in my overlay tree. :-) Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Firefox
Daniel D Jones schreef: I'd like to run Firefox 1.5. The latest stable build appears to be 1.07. I added ww-client/mozilla-firefox ~x86 to my package.keywords file, which resulted in Deer Park being installed. I don't want Deer Park, just the 1.5 release version. The version labelled as Deer Park under Gentoo *is* the 1.5 release version, to the best of my knowledge. As I understand it, there is a licensing requirement from the Mozilla group that specifies that compiled versions provided by anyone other than Mozilla.org cannot use the 'official' branding. Therefore, the release displays Deer Park but is otherwise identical (built from the same source release tarball as the Mozilla.org compiled release, just not built by Mozila.org, but by you or the Gentoo devs). Apparently you can 'fix' this (if it must be fixed) by activating the ./configure switch --enable-official-branding. You can of course do this by copying the current ebuild to your PORTDIR_OVERLAY, making the adjustment, and compiling that instead, or by --as you said-- installing or compiling the official Mozilla binary or source from the Mozilla.org site. Check the forums, I believe that that's where I discovered what information I know about this issue. I've considered downloading the source and manually installing but I'd prefer not to do something which may confuse portage at some point. Thanks for any suggestions. That's what Portage overlay directories are for; to do something like this without confusing Portage. Self-regulating, of course-- if the issue is not important enough to the user to make it worth their while to learn how to set up an overlay, and how to populate it with the relevant ebuilds (especially if the relevant ebuild must be created from scratch rather than just adjusted), then the issue is not necessarily important enough to warrant solving, offering the user an opportunity to re-evaluate the severity of the issue. As with all things Gentoo, it's your choice as to how you want to handle it. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] fglrx works. Yeah! Umm, now nothing exits. Uh oh!
Bruce Burden schreef: Okay, I have fglrx as the driver in the xorg.conf file. And, at 24 bit resolution, xdm starts X. Bliss! Then I realized that nothing was completing if it created an xterm of output or so. Drat! OS is Gentoo 2.6.15-gentoo-r1. Xorg is 6.8.2. Machine is a HP ZD8000 with an ATI X600 Mobile Radeon. Any ideas what is going on? Clearly it is related to X and fglrx, as the xorg.conf ATI driver doesn't do this. Modules: agpgart ati_agp fglrx Hi, Bruce :-). Number one, please don't hijack threads. This post came as a reply to the update-eix has memory problems thread; apparently you hit reply and changed the subject, rather than sending a new mail to the list (which would have started a new thread). This is bad mailing-list etiquette. Number two, as to your actual problem: a) I don't understand what nothing was completing if it created an xterm of output or so means. What is nothing? (not) completing how? I suppose what you're saying is that applications that produce terminal output are not completing in some fashion, but even if that's a correct reading, it still doesn't tell me much about what the problem actually is. Some error messages would be much more useful in this regard. b) I am wondering about that ati-agp module. Does that mean you're using the internal fglrx agp module (UseInternalAGPGart set to yes in xorg.conf), or that you're using the kernel module for ATI IGP motherboards? Do you actually have an ATI IGP mobo in your laptop, is it? c) does the fglrx driver actually now support the X600 series (mobile)? I know that people with X800s (desktop) don't have support (and, man, are they P.O'd), and mobile chip support tends to be somewhat choppy, so it's possible that there's only partial support for the chipset, resulting in this issue (I have a 9800SE, so while I don't find much to be grateful for in the ATI drivers, I am at least grateful for an older model card that is supported as well as anything is supported by the proprietary Linux drivers). Anyway, hope this helps, sorry for posting to a hijacked thread, but hopefully Bruce will repost in a new thread and we can go further. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] fglrx works. Yeah! Umm, now nothing exits. Uh oh!
Benoit Joseph schreef:
Hello All,
I've read that the ati-drivers does not support 2.6.15 kernel yet...
Not true; I've been using them together for the last two versions of the
ATI drivers:
motub - uname -r
2.6.15-gentoo
motub - fglrxinfo
display: :0.0 screen: 0
OpenGL vendor string: ATI Technologies Inc.
OpenGL renderer string: RADEON 9800 SE Generic
OpenGL version string: 2.0.5582 (8.21.7)
(upgraded from 8.20.8 yesterday).
I guess when you say nothing finish it means that your command hangs and
desn't give you the prompt. I have the same problem.
(ati Rf250 M9)
Here are a site that point the problem and a way to patch 2.6.15 kernel...
The gentoo ebuild(s) already contains this patch:
# $Header:
/var/cvsroot/gentoo-x86/x11-drivers/ati-drivers/ati-drivers-8.21.7.ebuild,v
1.2 2006/01/21 21:15:52 anarchy Exp $
snip
# fix kernel oops for acpi
if kernel_is 2 6 15; then
epatch ${FILESDIR}/${P}-linux-2.6.15.patch
fi
}
Holly
(sorry again, but I couldn't let that misinformation stand)
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] battstat requires apmd?
Iain Buchanan schreef: Hi, I've just rebuilt my system, and emerged gnome-light this time instead of gnome. I noticed battstat isn't part of gnome-applets, but when I try and emerge it, I get all these wierd deps. For a start, why does it need apmd? I thought I had it working with acpi only... (I certainly didn't have apmd on my last install). There are no useful use flags that I can see... Is there another battstat that uses acpi? $ emerge -pvt battstat These are the packages that I would merge, in reverse order: Calculating dependencies ...done! [blocks B ] x11-terms/gnome-terminal (is blocking gnome-base/gnome-core-1.4.2-r1) [blocks B ] gnome-base/gnome-desktop (is blocking gnome-base/gnome-core-1.4.2-r1) [blocks B ] gnome-base/gnome-session (is blocking gnome-base/gnome-core-1.4.2-r1) From this output, it looks like the battstat applet you're trying to install is for GNOME 1.x, not GNOME 2.x, which is what you're most likely running. GNOME is not modular in the way that KDE is; many individual GNOME widgets are not installable separately from the meta-package, most notably with applets. And IIrc (I don't have a laptop, so I generally ignore the battery applet), the applet you're looking for is included with the gnome-applets package, and will use acpid support if that USE flag is enabled: emerge -pv gnome-applets cfg-update-1.8.0-r3 : No new packages have been emerged, checksum index OK... These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild R ] gnome-base/gnome-applets-2.12.2 USE=acpi hal -apm -debug -ipv6 0 kB Yes, in fact it is included: equery files gnome-base/gnome-applets [ Searching for packages matching gnome-base/gnome-applets... ] * Contents of gnome-base/gnome-applets-2.12.2: snip /usr/libexec/battstat-applet-2 snip So, since gnome-applets is not a dependency of gnome-light (!?? never noticed that), just install it and you should then be able to add the battstat-2 applet to the panel via the normal means of adding an applet to the gnome-panel (right-click on the panel, Add to Panel, etc). HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: KDE starts and runs slowly
Korondi Márk schreef: Richard Fish schreef: On 1/29/06, Korondi Márk [EMAIL PROTECTED] wrote: Yesterday I reinstalled gentoo, because I'd made the filesystem dirty (always used sudo, and installed programs not in portage randomly...) and I want a new, clean system. With Gentoo there is almost always an easier way than re-installing...but too late for that now. Is there? And what is that way? (because in the future I'll use sudo,too :-)) Well, the answer to this question kinda depends on how you feel that using sudo somehow made (your) filesystem 'dirty'. After all, sudo exists to be used, so that can't be the problem... but I don't know what you think the problem was, much less why you think sudo was related to it. Now, as to the question of installing programs not in Portage randomly, well, you certainly don't need to reinstall to solve that. Uninstall the programs and install them not randomly (i.e., in /usr/local/src or /usr/local/bin, since that's what /usr/local is for, or even /opt, since that's what /opt is also for). Or uninstall them and check b.g.o to see if there are ebuilds for them and create an overlay directory. Or write an ebuild for the program and put it in your overlay. But the point we're trying to make is that if 95% of the hundreds of applications you have installed are working fine, and you have them configured to your liking (especially DE's like KDE, or applications like Wine or Firefox), then deleting everything and having to set up those applications all over again just because two or three (or even ten or twenty, when you have a couple hundred apps installed) is a waste of your time and effort. Next time, maybe you'd like to tell us what precisely your issues are *before* you go reinstalling-- at least then you would be able to have the use of the rest of the computer while we helped you solve your localized issues. do you set KDE_IS_PRELINKED? No, I didn't. So, now I'm under twm, with my being fresh system - I actually reinstall gentoo -, and I hope that KDE will work properly. Why do you think that the problem was KDE, or that KDE will suddenly work differently if you didn't change any settings? But, good luck anyway. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Installing KDE...Again.... (Now at howto install kde)
Abhay Kedia schreef: On Sunday 29 January 2006 06:41, Ian wrote: unmask all of the packages im building in package.keywords. You seem to have added them to package.unmask as well. Don't do that. Just as a note, there doesn't seem to be a choice in this regard (i.e., you have to add kdewhatever-3.5.1 to package.unmask). If I don't add it to package.unmask, I get a halt from Portage because some of the dependent packages that [ nomerge ] packages want to emerge are masked. The only way I've found around it in the past three days is to just go ahead and unmask them, let Portage fail to download then tarballs, and then do a --resume --skipfirst to proceed. I could of course not --deep the update, but that covers everything, which I don't want. I also upgraded Portage to see if it was a Portage problem, but that didn't help either. It's odd, imo... Portage shouldn't really consider masked packages for updates, but in this case I can't seem to get it to stop. I was also wondering if it was a bug, and in what (KDE, Portage... something). It seems to me that the devs have done this before (put a package in the tree just before it was available, masked), and I can't figure out why its not working this time the way it has in the past. If its something I've done, I can't figure out what But I've been busy, so I've only had time to be annoyed by it, not investigate it. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Installing KDE...Again.... (Now at howto install kde)
Neil Bothwick schreef: On Sun, 29 Jan 2006 12:56:17 +0100, Holly Bostick wrote: Just as a note, there doesn't seem to be a choice in this regard (i.e., you have to add kdewhatever-3.5.1 to package.unmask). If I don't add it to package.unmask, I get a halt from Portage because some of the dependent packages that [ nomerge ] packages want to emerge are masked. 3.5.0 packages should satisfy any such dependencies. Of course, if both 3.5.0 and 3.5.1 are masked in one way or another, portage's message will probably refer to the latest version. Putting 3.5.0* package versions in package.keywords should be sufficient. i've certainly not had anything demand a 3.5.1 package on three very different (but all ~arch) boxes. I'm also running ~x86, and kdelibs and kdebase-kioslaves 3.5.0 (the packages I'm having problems with) are currently installed. But if I do an emerge -uaD(N)tv world, krusader (because it is compiled with konqueror support, so it depends on Konq), and something else I forget (oh, it was knowit, but I uninstalled that, in an effort to solve this problem), demand to upgrade kdelibs and kioslaves to the masked versions (which should not be available). And of course, if Portage finds that something necessary is masked, it stops dead. Adding the two packages to packages.unmask at least allows the emerge to proceed, though of course I have to hang about in order to skipfirst, due to the two unavailable tarballs. Not that I have any particular desire to upgrade these packages to 3.5.1, but Portage wants to make me, for some reason. In any case, I don't so much want to hijack the thread (actually, I don't want to hijack the thread at all), but if this is a 'general' problem of some sort (which seems amazing, but I have no other explanation for the bizarre behaviour that both I and the OP seem to be seeing), I wanted to mention. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Installing KDE...Again.... (Now at howto install kde)
Neil Bothwick schreef: On Sun, 29 Jan 2006 19:08:11 +0100, Holly Bostick wrote: I'm also running ~x86, and kdelibs and kdebase-kioslaves 3.5.0 (the packages I'm having problems with) are currently installed. But if I do an emerge -uaD(N)tv world, krusader (because it is compiled with konqueror support, so it depends on Konq), and something else I forget (oh, it was knowit, but I uninstalled that, in an effort to solve this problem), demand to upgrade kdelibs and kioslaves to the masked versions (which should not be available). That's odd. Krusader doesn't want anything masked here. # emerge krusader -pv These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild N] kde-misc/krusader-1.70.0_beta2 USE=-arts -debug -javascript kde -xinerama 0 kB Are you sure you don't have any KDE packages in package.unmask? Oh, I'm sorry-- I should have mentioned that I'm using the live CVS ebuild from the Krusader site, which has different USE flags: emerge -pv krusader-cvs Calculating dependencies ...done! [ebuild R ] kde-misc/krusader-cvs-1-r4 USE=javascript konqueror -arts -debug -xinerama 0 kB [1] Total size of downloads: 0 kB Portage overlays: [1] /usr/local/portage ... but I found the problem. There was an old setting in /etc/portage/package.mask that needed to be cleared out (I had masked = some 3.4 version of kioslaves, for reasons that I assume were good at the time, but I totally forgot that I had ever done it, so I certainly don't remember why I had done it). Removing that enabled the profile package.mask to work normally once I removed the bogus and unnecessary /etc/portage/package.unmask settings. So that's fine then, sorry for the noise and thanks for the encouragement to get to the bottom of what I had done to screw things up (since I knew it had to be me, given that Portage really is unlikely to have gone so schizo without anyone noticing *but* me :-) ). We now return you to your regularly scheduled thread (with a possible advisory to Ian to check his little-used /etc/portage files; maybe there's a setting that you forgot you put there, as well :-) ). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: php blocking
James schreef: Richard Fish bigfish at asmallpond.org writes: Try PEAR-PEAR (packages names are case-sensitive). # emerge --unmerge PEAR-PEAR --- Couldn't find PEAR-PEAR to unmerge. unmerge: No packages selected for removal. I even 'emerge --sync' again, today and still I have: Add the --tree option to see what is actually trying to bring in dev-php/mod_php and dev-php/php. # emerge --unmerge --tree PEAR-PEAR --tree implies --pretend... adding --pretend to options. These are the packages that I would unmerge: --- Couldn't find PEAR-PEAR to unmerge. unmerge: No packages selected for removal. I'm not so sure I tried what you are suggesting. Can if I missed your point, can you be more explicit in your syntax suggestions? What Neil was saying was to add --tree to your emerge world command (not your unmerge command), so that you could see what requires PEAR-PEAR to be installed. --tree shows the dependency/reverse dependency tree of the items to be emerged, so you can see why something being emerged/upgraded as a dependency of something that is currently up-to-date is being emerged (and what the main application is). Example: emerge -uaDNtv world These are the packages that I would merge, in reverse order: Calculating world dependencies - ...done! [nomerge ] media-gfx/gimp-2.2.10 USE=X aalib gimpprint gnome gtkhtml jpeg lcms mmx mng png python scanner sse svg tiff wmf -debug -doc -hardened -pdf -postscript -smp [ebuild U ] media-gfx/gimp-print-5.0.0_rc2 [5.0.0_rc1] USE=cups foomaticdb gtk nls readline -ppds 0 kB This output shows that gimp-print is being updated as a dependency of gimp (which is no(t being) merge(d) because it is itself up to date). This is of course no great surprise, but with things like PHP and Apache and whatnot, this information can be invaluable. In any case, when you see what is requiring PEAR-PEAR to be installed/upgraded, you may well find that you have more options as to how to handle the block. HTH, Holly Remember, I ran eix php and eix pear and now 'eix PEAR' Absolutely nothing is installed with any of these keywords in the name.on the system with this trouble. James -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Gnome calendar - Start on Sunday instead of Saturday?
Mark Knecht schreef: Hi, Title says it. My wife wants to configure this item and I cannot find the control of it in the preferences page. Anyone know how to make it start on Sunday instead? Thanks, Mark I'm sorry not to be more precise, but I suspect that I may not have the calendar application you are using. But here is what I found: 1. emerge (or run if installed, but it's now a separate package, and I didn't have it, amazingly) gconf-editor. 2. in gconf-editor, go to / = apps = evolution = calendar (this of course assumes that evo controls the calendar, which it well might, since I don't have evo or eds installed and I still have this settings tree) = display 3. In the display tree (items shown now in the right pane), change week_start_day from 1 (Monday, which is what must be the default, as that's what mine is set to and I've never touched it) to 0 (which represents Sunday, as the text will tell you). I suppose there's an easier way to do this (perhaps if you have Evolution installed, you can change it in their prefs setting), and perhaps this is in fact not the setting that controls the calendar, but I hope it is, and I hope it helps. If not, trawl through gconf-editor and see if anything else looks relevant (for example, if you know the name of the calendar application, it probably has a settings tree in gconf-editor). HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Mysterious dev-lang/php 'n' junk 'n' stuff....
gentuxx schreef: Hi all, snip I went to do an `emerge -Duatv world` tonight, and I get dev-php/php and dev-php/mod_php blocking. So I uninstalled them, and thought that I would re-install later (if necessary). When I run it again, I get dev-lang/php blocking dev-php/php-4.4.0-r4 (which is the one I just unemerged). So, same logical progression, I go to unemerge dev-lang/php, and, lo and behold, it's not installed. snip So, my questions to the group are these: 1) Why does `emerge -Duatv world` find a package (which blocks another that it wants to update) that `emerge -CDatv world` does NOT find? As far as I know, emerge -CDatv world is not even a valid command (at least I hope it isn't). This translates to emerge --unmerge --deep --ask --tree --verbose world, and it seems to me would unmerge everything on your system if it was allowed. The usual use for --unmerge is to tell Portage what you specifically want to unmerge; this should not be an automatic function (except in the case of repair utilities such as depclean, which even those give you a HUGE warning that unmerging random --as in not pre-specified by the user, but determined by a script-- packages automatically can break the system). The actual answer to your question, though, is that the blocking package is not a problem unless you want to emerge something that it blocks. So of course it's not going to be noticed until you attempt to perform such an action (emerging a package that is blocked by the installed package). 2) How do I rectify this little version discrepancy fixed so that I can get things updating normally again? Iirc from several threads recently, there are one or more packages on your system that rely on php which must be updated to ~arch, or else they keep trying to bring in PHP4 (only the ~arch version can depend on PHP5; the stable versions have a fixed dependency on PHP4). I don't remember which programs they are, but if you don't feel like searching the archives, you can look at your tree view to see what is actually pulling in PHP4, upgrading that (or keywording it to unstable, rather) and see if you still get the block. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Module names???
Andrew Lowe schreef: Hi all, A simple question, where do I find out modules names? I've currently got menconfig loaded and trying to decide as to whether to go static or module. My problem is that I can't find the names of the modules for most of the options hence how do I know what to load if it is a module. For exmaple, if I want to use the XFS file system, in the menuconfig help it says: ... choose M here; the module will be called xfs... Conversly if I want to use AES encryption there is nothing in the help that tells me what the module name will be. To some degree, they are: The title of the help for the option usually includes something closely related to the module name. For CONFIG_CRYPTO_AES_586 I would guess that the module name might well be aes-586. Apparently there's a 64-bit module as well, but it may be that that doesn't appear in my kernel config because I have already specified that I don't have a 64-bit CPU. In any case, the format for the help title is CONFIG_GROUPNAME_SOMETHING_LIKE_THE_MODULE_NAME (but module names usually use hyphens where the kernel config uses underscores). However, you most likely will not need to load the modules manually anyway; assuming you have automatic kernel module loading enabled, the kernel will load necessary modules when it notices that they are necessary (when a device using the module is detected and initialized). Also, I would not myself think that encryption is the kind of thing you'd want to load only some of the time (so I'd compile it statically anyway). Thirdly, are you sure that all of the options that you're looking for the module names for are capable of being compiled as modules? Some aren't. In the case that you don't know (it could happen :-) ), options appearing with brackets ([ ]) can only be compiled statically; options appearing with greater-than/less-than signs ( ) can be compiled either statically or as modules. So it's also possible that some of the help doesn't contain module names simply because the option cannot be compiled as a loadable module. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] How many people use KDE?
b.n. schreef: I'm just writing it for the sake of curiosity, so no flaming is here. Just because some answer sound quite sarcastic, but that's just a style thing to get it short. :) Yes, but you then have bloat (because Konqueror contains web browsing features that you are not using, therefore the code is unnecessary for you, but nonetheless present). [...] Code, code, code. Bloat (for me). [...] Fine, I can turn them off, but again, there is then a whole lot of backend code for a feature that I do not want in the first place and know I don't want. Ehm. Perhaps it's me being dense but: who cares about unused code? Ok, you have unnecessary, unused code sitting on your HD: where's the problem? You never see it. If I wanted unused and unneccessary code sitting on my PC, I'd use a binary distribution. Why do I bother with disabling USE flags to not compile code that is unnecessary for me, if I didn't care about such things? On the rare occasions that I compile Mozilla (becoming less and less necessary, thank goodness), I compile it +moznomail, +mozcompose, +moznoirc, and -mozcalendar because I don't use those features in Mozilla, so why should I wait for them to be compiled? But I can't trim Konq down to be just a file manager (or browser, depending on which function I hypothetically like Konq to perform and which I prefer to use another application to perform). And of course, maybe I don't have so much HDD space that I want some portion of it to be used by applications that have extra functions that are unused, when that space could be used by applications that do things I *do* want. I have to then spend time finding out how to disable it or avoid installing it. It's quite odd you obviously had spent the (worthwile but not instantaneous) time to learn Linux, install Gentoo etc. but then you can't type emerge --unmerge kmix. Well I would if I liked KDE (and on the occasions that I have installed KDE, I've done that), but I've already objected to this idea that it's OK to install a whole complex and then have to go through it again with a fine-toothed comb to edit it down to something manageable. Fine if you don't mind working that way, but I do. I can't even understand what do you mean here. If you don't want icons, don't put them on the desktop. It's that simple. You have to do *nothing* to avoid icons on your desktop! The (presumably) default setting (since I've never touched it, and it is checked in kcontrol) is Show icons on desktop. There are then two additional tabs for kinds of icons that you can enable or disable (for file types and drives). But if you don't actively link things on the desktop, *nothing* appears on your desktop!! That's not the point, which is where we have a failure to communicate. Openbox and FVWM-crystal (and ICEwm, for that matter) are lighter, faster desktops than KDE partially because they do not contain the code to put icons on the desktop (whether I enable it in KDE or not). If I suddenly change my mind and want icons on my desktop, I have to install idesk or something. That's the way (unh-huh, unh-huh) I *like* it. If I want an application to perform a function that I want or need, then I install it. If I don't want or need the functionality, it *is not present*. I have no interest in going through 6 tabs to specify Window Behaviour (I'm looking at the KDE Control Center right now). Ok, that's a good point. However that 6 tabs are more probably than not a wrapper to a plain text config file, that you can configure with your favourite editor all at once. Code for a gui function that I'm not using if I'm just editing the base text file anyway. ? Same as before. If I'm going to be editing the text file, all I need is nano (which I already have). So the code to create, manipulate and draw those tabs and their functions in the KDE control center is unnecessary. But developers have spent time to write it, and debug it, testers have spent time testing it, and I've spent time compiling it... and they've all wasted that time with respect to me, because I'm not using it, I'm just editing the text file in nano (which I already had). So for all of me, they could have done something else with that time (like make the code modular, so if I didn't want it, I could disable it with a USE flag or something, or of course, not include it at all, since I find Devilspie works just fine to control my windows if I need to do that for some reason, or just tell me where the text file is and how to edit it, like the FVWM man pages do, so I use the internal settings to control my windows if I want, but don't have a whole GUI that I find unnecessary to do so). Yes they work fine, but they look like poop unless you jump through some hoops to integrate them with the look of your KDE desktop. This may involve installing additional applications (gtk-chtheme or gtk-engine-qt), or editing a text file (if you need to
Re: [gentoo-user] Running hddtemp with plain user rights
Alexander Skwar schreef: Hello! I'd like to be able to run hddtemp http://www.guzu.net/linux/hddtemp.php with plain user rights - ie. not with root rights. What's to be done, so that this is possible? snip As you can see, I get the error message Permission denied when I run hddtemp /dev/hda. As you can further see, the group disk has read-/write-access on the device file. And lastly, you can see that my user is member of the group disk. Why do I get the Permission denied error message? And what's to be done? Isn't hddtemp a daemon (therfore executable)? Does the group have the right to execute hddtemp? Just an idea, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] How many people use KDE?
Neil Bothwick schreef: On Fri, 20 Jan 2006 17:10:13 +0800, Linux Java wrote: I wanna to know KDE and Gnome which is more popular. Why? Use whatever suits you. I hope that you all appreciate my extreme restraint in not posting to this thread until now, given how very much I dislike KDE. But for the record, just so that all you KDE-heads don't skew the results com*plete*ly: I always (from my first attempts at Linux some 3 years ago) preferred GNOME to KDE. Never liked Nautilus, though (it's tied for second on my list of most hated file managers), and since I've never been fond of desktop icons and all that cr... junk... I still found it too heavy. So I switched to Openbox 3 (with a GTK backend), and now I use fvwm-crystal (with a GTK backend). Gnome-light is (always) installed, but I don't use it as a desktop. I have only two KDE-specific applications that I would not do without (both compiled -kde and -arts to the greatest extent possible): Krusader (though this needs Konq and some other KDE utils for best usage, as it recognizes KDE apps much much better than GTK apps for viewing files and the like), and K3b. These apps require kdebase, so I've got that, but the day you see me logging into KDE, you can rest assured that either: 1) my system is so seriously broke that it's the only DE/WM I can get into (which is pretty unlikely. I mean, I've got iceWM and *afterstep* on the system, for Pete's sake; the day that doesn't work but KDE does will be... The day); or 2) I have been replaced by an alien clone (shoot first, ask questions later). I prefer to use GTK-based applications wherever possible because I find them more attractive in general, and I'm more used to them (as a GNOME user originally), unless they're junk, like Totem, in which case I use non-affiliated programs like Xine or mPlayer. Yes, I know Totem can be configured to use a Xine backend. Imo, there's no point; if that's the only way Totem works, I might as well just use Xine. Plus I want to see when gStreamer gets its act together. However I have no objection to QT-based apps (as opposed to KDE apps) when necessary. It does need to be necessary, though (meaning, if I need it, I'll install qdvdauthor, because there's no GTK alternative that I know of, but I can just as well use the CLI original, unless the GUI version has some additional feature or makes it easier to understand than the CLI version's man page). So anyway, Neil is of course right: use what you want; it's *your* desktop (finally!). I don't need a whole lot of GUI features (in fact I dislike a whole lot of GUI features), so KDE is not for me, the one who never liked Windows(-like) desktops, even when I was using Windows; I used an alternative shell from my Win98 days on. But for those who feel more comfortable with a more Windows-like environment, and Windows-like assumptions about what a user wants/needs from their desktop, KDE may be just the thing; that is, after all, what it's designed to do to a great extent. You can have it, though. I'll be elsewhere. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] vim USE flag: vim-with-x
Tom Smith schreef:
Rumen Yotov wrote:
Hi, Two ways (i know) to see the USE-flag descriptions. 1.Run
:#euse -i opengl ('euse' is part of app-portage/gentoolkit).
2.Use#grep USE-flag-name /usr/portage/profiles/use.desc (or
use.local.desc) use.desc - global USE flags, use.local.desc - for
flags loxcal for a package. PS: IIRC there was some third
alternative but can't remember right now. HTH.Rumen
The first place I always look for USE flag descriptions is in
/usr/portage/profiles/use.desc--this particular flag isn't listed.
That's because it isn't a global USE flag, like gnome; it's a local
USE flag, which applies to vim only.
Local USE flags are described in /usr/portage/profiles/use.local.desc.
Myself, I use an alias to search useflags:
(in ~/.bashrc)
alias useflag='grep /usr/portage/profiles/use.*desc -e'
This searches both use.desc and use.local.desc for whatever I type after
the command useflag
so:
motub - useflag vim-with-x
/usr/portage/profiles/use.local.desc:app-editors/vim:vim-with-x - Link
console vim against X11 libraries to enable title and clipboard features
in xterm
Great for when I've done an emerge -uaDtNv world, and don't know what
thus and so new USE flag does.
Holly
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] How many people use KDE?
Abhay Kedia schreef: On Friday 20 January 2006 19:59, Holly Bostick wrote: more comfortable with a more Windows-like environment, and Windows-like assumptions about what a user wants/needs from their desktop, KDE may be just the thing; that is, after all, what it's designed to do to a great extent. That statement is extremely unfair to KDE. Every DE needs to make at least some assumptions to present a working environment for the user That may be true, but it assumes that I want a Desktop Environment in the first place, which I don't, particularly. As I said, after finding even GNOME too heavy, I switched to Openbox 3, which basically presents *no* working environment, and now use fvwm-crystal, which presents a relatively minimal one, certainly by comparison to KDE. You may see it as unfair (like I care; KDE has no feelings, and I was actually being positive about it, given that I don't much like DEs in general and KDE in particular), but the very concept of a Desktop Environment is very similar to the Windows design of the OS being indistinguishable from the GUI (or rather the OS functions being as concealable within the GUI as possible), so I'm not quite sure why that's so terrible to say. and KDE is not alone in doing that. The quality comes in when the DE not only makes some choices for you but also gives you an easy access to editing/changing those features. As far as I can see, you can edit every choice that KDE has made for you with minimal fuss. I myself don't see it as minimal fuss, not least because KDE makes so many choices for me in its feature richness that I have to spend two hours (I'm being kind) finding all the bloody options that I don't want and change them or turn them off or whatever. And of course this has its limits; suppose I don't *want* my file manager integrated with my browser (didn't want it in Windows, don't want it now). I don't *want* to figure out how to tell KMix not to override my Alsa mixer settings, and I don't want to have to decide whether I want drive icons but not application icons (or no icons at all) on my desktop, and then tell KDE my decision. I don't want to name my desktops, or put a separate wallpaper on each one. I have no interest in going through 6 tabs to specify Window Behaviour (I'm looking at the KDE Control Center right now). And I certainly don't care to be bothered with the problem of how to make KDE play nice with my GTK apps (I do have some GTK 1 apps, which are much more problematic than GTK 2 apps in this respect) simply because I might happen to want to use some program whose name doesn't bloody start with K. You can say that I don't have to do any of this, or change any of the settings, but if I don't, then I'm stuck with KDE's sane defaults, and in that case then all these settings that I *could* change (but don't bother to) become bloat -- features I don't need, since I'm not going to use them. If I don't like KDE's default choices, then I have to fix them, within the parameters that KDE allows me, which is no longer minimal fuss. Certainly not minimal fuss when I want to use non k apps for certain functions that KDE would prefer that I use their provided applications for... I have several K-apps installed that I actually don't want, because the K-app I do use (Krusader) won't open files from within an archive using GTK apps like eye of gnome or Open Office. karc can't pass the file to these apps but it works fine with KView or KWord. Because K apps like other K apps. That makes perfect sense, since it's all supposed to be an integrated environment, but to me it feels like a prison. If that is being Windows-like, then you can not be more wrong. Insofar as Windows doesn't allow you to customize as much as KDE does, you're right. But in the idea that you give the user everything up to and including the kitchen sink by default unless the user themselves whittles the list down (within limits), then no, I'm not all that wrong. But I simply don't like DEs. If I'm going to spend time fine-tuning my desktop, I want exactly what I want, exactly the way I like it, not as close to how I like it as the DE supports. That's why I use build-it-yourself WMs like OB3 and FVWM. Each to his own taste, and if your taste is KDE, then more power to you. I accept that it's very good for what it is. I just don't happen to like what it is. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] How many people use KDE?
b.n. schreef: I myself don't see it as minimal fuss, not least because KDE makes so many choices for me in its feature richness that I have to spend two hours (I'm being kind) finding all the bloody options that I don't want and change them or turn them off or whatever. Sorry, I simply can't understand what are you talking about: suppose I don't *want* my file manager integrated with my browser (didn't want it in Windows, don't want it now). Where's the problem? I use Firefox as a browser and Konqueror as a file manager. Yes, but you then have bloat (because Konqueror contains web browsing features that you are not using, therefore the code is unnecessary for you, but nonetheless present). I don't *want* to figure out how to tell KMix not to override my Alsa mixer settings, Disable it. Do not install it. I have to then spend time finding out how to disable it or avoid installing it. and I don't want to have to decide whether I want drive icons but not application icons (or no icons at all) on my desktop, and then tell KDE my decision. I can't even understand what do you mean here. If you don't want icons, don't put them on the desktop. It's that simple. You have to do *nothing* to avoid icons on your desktop! The (presumably) default setting (since I've never touched it, and it is checked in kcontrol) is Show icons on desktop. There are then two additional tabs for kinds of icons that you can enable or disable (for file types and drives). Fine, I can turn them off, but again, there is then a whole lot of backend code for a feature that I do not want in the first place and know I don't want. I don't want to name my desktops, or put a separate wallpaper on each one. Nor I want, nor KDE ever forced me to do it. Code, code, code. Bloat (for me). I have no interest in going through 6 tabs to specify Window Behaviour (I'm looking at the KDE Control Center right now). Ok, that's a good point. However that 6 tabs are more probably than not a wrapper to a plain text config file, that you can configure with your favourite editor all at once. Code for a gui function that I'm not using if I'm just editing the base text file anyway. And I certainly don't care to be bothered with the problem of how to make KDE play nice with my GTK apps (I do have some GTK 1 apps, which are much more problematic than GTK 2 apps in this respect) simply because I might happen to want to use some program whose name doesn't bloody start with K. What do you mean by play nice? I use a lot of GTK apps (both 1 and 2) and they work perfectly fine here without any fuss. Yes they work fine, but they look like poop unless you jump through some hoops to integrate them with the look of your KDE desktop. This may involve installing additional applications (gtk-chtheme or gtk-engine-qt), or editing a text file (if you need to fix GTK 1 programs, which are generally not affected by the theme consolidation programs, which generally assume you're working with GTK2). Since one of KDE's big selling points is an integrated look-n-feel, outside apps that break the loveliness of the KDE desktop are very noticeable. I have several K-apps installed that I actually don't want, because the K-app I do use (Krusader) won't open files from within an archive using GTK apps like eye of gnome or Open Office. karc can't pass the file to these apps but it works fine with KView or KWord. Because K apps like other K apps. That makes perfect sense, since it's all supposed to be an integrated environment, but to me it feels like a prison. I understand what you mean here, that's something that I hate too. It's because other apps don't understand kioslaves. I think kioslaves are the best thing after sliced bread. Typing simply smb:/ to access a Samba share, or ftp:/ for accessing transparently FTP filesystems, or zip:/ for what's in a compressed archive is wonderful. I think other apps should work on compatibility about this. It's simply a good idea. I don't even type things like that, I bookmark locations in my file manager (admittedly, Krusader, if installed with konqueror support-- which means I have to install Konq, though I don't use Konq-- does recognize kioslaves, so I can bookmark folders in media:/ or smb:/ ) and just go where I intend to go. without further ado. But I can bookmark locations (even Samba shares and HAL mounts) in most file managers I have available (Nautilus, Krusader, TuxCommander, emelFM2) But I simply don't like DEs. If I'm going to spend time fine-tuning my desktop, I want exactly what I want, exactly the way I like it, not as close to how I like it as the DE supports. That's why I use build-it-yourself WMs like OB3 and FVWM. I can't see why a WM can be more exactly what you want than a DE. I want to use the programs that I like, without worrying about whether they are compatible with my DE. I don't want to be
Re: [gentoo-user] Multiple Firefox v1.5 Instances
Richard Ruth schreef: How do I start a second instance of FireFox V1.5? (I want to do this because I have different proxy settings, etc. with each different profile.) There is no need to start a second instance of Firefox just to change your proxy settings; there are several extensions that allow you to manage your proxy settings on the fly: SwitchProxyTool: https://addons.mozilla.org/extensions/moreinfo.php?id=125application=firefox QuickProxy: https://addons.mozilla.org/extensions/moreinfo.php?id=1557application=firefox xyzproxy: https://addons.mozilla.org/extensions/moreinfo.php?id=1273application=firefox You might also find Quick Preference Button: https://addons.mozilla.org/extensions/moreinfo.php?id=1851application=firefox useful. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] New install, I still can't send email. Same error too.
Dale schreef:
On Tuesday 17 January 2006 23:00, Richard Fish wrote:
find ~/.mozilla -name prefs.js -exec grep mail.smtpserver {} \;
This is on my old install.
[EMAIL PROTECTED] / # find home/dale/.mozilla -name prefs.js -exec grep
mail.smtpserver {} \;
user_pref(mail.smtpserver.smtp1.auth_method, 1);
user_pref(mail.smtpserver.smtp1.hostname, mail.exceedtech.net);
user_pref(mail.smtpserver.smtp1.port, 25);
user_pref(mail.smtpserver.smtp1.try_ssl, 0);
user_pref(mail.smtpserver.smtp1.username, dalek);
user_pref(mail.smtpservers, smtp1); [EMAIL PROTECTED] / #
This is from the new install that I mounted on /mnt/gentoo, you know,
like in the install guide. LOL
[EMAIL PROTECTED] / # find /mnt/gentoo/home/dale/.mozilla -name prefs.js
-exec grep mail.smtpserver {} \;
user_pref(mail.smtpserver.smtp1.auth_method, 1);
user_pref(mail.smtpserver.smtp1.hostname, mail.exceedtech.net);
user_pref(mail.smtpserver.smtp1.port, 25);
user_pref(mail.smtpserver.smtp1.try_ssl, 0);
user_pref(mail.smtpserver.smtp1.username, dalek);
user_pref(mail.smtpservers, smtp1); [EMAIL PROTECTED] / #
Well, this may explain why neither one works. They are the same.
Does that look right??
Looks right to me; these are the same settings I have in Thunderbird
(with different hostname and username, of course), and I don't have any
problems.
What I don't get is why moz is trying to relay directly. I'm no
mail/network guru, but I do know that if I send a mail to someone
outside my network (like, for example, the list), *I* don't relay (send
the mail via another ssmtp server) but rather I send it to my ISP and
*they* relay it (just like snail mail; I drop the letter in the post box
near me and the Postal Service takes it to the next node in the town
where the recipient lives, and *their* local mail service delivers it.
It's not like if I want to write somebody in NYC, I have to mail the
letter *in* NYC for it to be delivered...!)
About all I can ask at this point is:
1) what version of Mozilla are you using? What are your USE flags for
your installation? Have you checked b.g.o or mozillazine
(http://www.mozillazine.org/ ) to see if this is perhaps a known bug of
your version of Mozilla? Have you tried upgrading (maybe it's a bug
that's fixed in later versions)?
2) If you go to Edit=Preferences=Account Preferences=Outgoing server
(SMTP), is there more than one server listed, and is your ISP's server
selected as standard? Like Iain, I had similar problems (a long time
ago) when attempting to send mail via my home ISP from work, where we
had a different ISP, which wouldn't work because I was logged into my
work's mail server (of course), and not my home ISP's SMTP server, so my
home ISP wasn't going to send my mail, since I wasn't an authorized user
(not logged in). It's hard to believe that there isn't *some* problem
with your SMTP settings that we're missing, though there is no error to
be seen, and of course KMail is presumably set the same way, and *that*
works
3) I'm wondering if this isn't a Mozilla problem; very curious as to
whether Thunderbird would exhibit it as well. If you want to try it, you
can use the same mail folders and settings as Mozilla, which I can
quickly tell you how to set up (it's like 5 steps, if T-bird doesn't
just offer to import the Mozilla settings for you) if you decide to try
it. Also curious as to whether there's *any difference whatsoever*
between your KMail settings and your Mozilla settings (yes, it's
fine-toothed-comb time).
4) I'm also wondering if this is a local problem (to a specific
outside address), or a general problem (to any outside address).
Have you tested mail to multiple/different outside addresses, or just
one? If just one, maybe that one is the problem. You can send me a test
mail; I'm certainly outside your local network :-) .
5) It's a long shot, but is there any possibility that you're sending to
an outside address to which your ISP will *not* relay, i.e. someplace
that *cough* Homeland Security doesn't want you writing to (maybe you
have a relative in the service, I don't know)? If there are such domains
(and I'm sure there are), it's within the (distant, we hope) realm of
possiblility that
your ISP is being excessively cautious in order to avoid any spotlight
of investigation falling on them.
6) I don't know much (all right, anything) about sending command-line
mail, but there must be a way to get a traceroute of a mail. Maybe
somebody here knows how we could get some useful output about what
happens when this mail pretends to relay itself for some reason.
I just thought of something. My new install has two accounts, the
dalek one and my new one rdalek. Where's rdalek?? I'm going to
change later and get rid of some spam, I hope.
I had to fill in the user directories since I was logged in as root.
Root don't have a .mozilla directory.
Regular users don't have any rights to read root's home directory.
Probably Mozilla
Re: [gentoo-user] Java and java.library.path
darren kirby schreef: quoth the Trenton Adams: on the java command line put -Djava.library.path=/usr/lib I don't know if gentoo has a config for this or not, I didn't know ADDLDPATH existed. Who knows, perhaps editing the JRE version of that file would help? I am really sorry, but I don't know what you mean by 'java command line'. The app itself, is in /usr/bin like everything else, it just gives me that error when I start it from an xterm, ie: $ iriverter insert error message Well, I don't know much about java either, but 4 out of 5 java programs I use are not started by program_name, but by java (-jar) program_name. This would be the java command-line, I imagine. Basically, the idea is that you have to invoke java so that java runs the program. Because the first argument in the command is java, you are then able to use various command-line switches for Java (such as the aforementioned -Djava.library.path=/usr/lib) before telling Java what program you want it to run (in this case iriverter). java -help Usage: java [-options] class [args...] (to execute a class) or java [-options] -jar jarfile [args...] (to execute a jar file) where options include: -client to select the client VM -server to select the server VM -hotspot is a synonym for the client VM [deprecated] The default VM is client. -cp class search path of directories and zip/jar files -classpath class search path of directories and zip/jar files A : separated list of directories, JAR archives, and ZIP archives to search for class files. -Dname=value set a system property -verbose[:class|gc|jni] enable verbose output -version print product version and exit -version:value require the specified version to run -showversion print product version and continue -jre-restrict-search | -jre-no-restrict-search include/exclude user private JREs in the version search -? -help print this help message -Xprint help on non-standard options -ea[:packagename...|:classname] -enableassertions[:packagename...|:classname] enable assertions -da[:packagename...|:classname] -disableassertions[:packagename...|:classname] disable assertions -esa | -enablesystemassertions enable system assertions -dsa | -disablesystemassertions disable system assertions So I think what Trenton probably meant was that you should try starting the program with java -Djava.library.path=/usr/lib iriverter (assuming that iriverter is not a *.jar file in which case you'd need -jar before iriverter). You know, there's a high likelihood that the iriverter file in /usr/bin is a script that is readable in a text editor. You might want to see if you can have a look at it and see just what it's doing. A readme that tells you how to run the program (man iriverter? iriverter --help?) might be of some use, too. Hope this helps, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] New install, I still can't send email. Same error too.
Dale schreef: On Wednesday 18 January 2006 07:43, Dale wrote: Well, Mozilla won't even open now. Well, you can add the binary version of Mozilla to the Do Not Open list. It didn't work either. Does anybody know if using the ipv6 flag when you don't actually have ipv6 available (from your ISP) might do something like this? Istr that I had a lot of difficulties with Internet applications when I first installed (because ipv6 is enabled by default) until I disabled it. I'm really out of other ideas, since I can only imagine that this is a Mozilla issue, and I can't imagine any issue that might have been caused by a bad install (and which could be fixed by a reinstall) that would not affect other such applications (so I have no idea what could possibly be wrong). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] mozilla firefox 1.5 and gnome
Trenton Adams schreef: On 1/15/06, Holly Bostick [EMAIL PROTECTED] wrote: Trenton Adams schreef: On 1/14/06, Richard Fish [EMAIL PROTECTED] wrote: gnome carcharias rjf # emerge -pv mozilla-firefox These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild R ] www-client/mozilla-firefox-1.5-r9 USE=java -debug -gnome -ipv6 -mozdevelop -xinerama -xprint 0 kB The appearance of USE flags (which is what we're talking about) is unrelated to whether or not you have the package installed already snip Oh, I get it. the thing that confused me was that he put the USE=, so I thought he was showing me what use flags he had set, not the output of the emerge command. Now I see he was showing the output of the emerge command and adding USE=. Either that, or he's using a newer version of emerge that outputs USE= for him. Yes, he probably is using the unstable version of Portage; mine does this as well if I use the -v (--verbose) switch (I just went fully ~x86 a couple of days ago): emerge -pv mozilla-firefox portage These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild R ] www-client/mozilla-firefox-1.5-r9 USE=gnome java xprint -debug -ipv6 -mozdevelop -xinerama 0 kB [ebuild R ] sys-apps/portage-2.1_pre3-r1 USE=-build -doc 0 kB It's a new Portage feature. Not sure if I like it, but it's not important enough for me to get worked up about, really. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] alsa-utils + gnome
habutre . schreef: Hi guys! My problem it's following: I've a AC97/I82801 sound device building as a built-in on my kernel. My sound work a fun, but I can't store the sound mixer's without a alsa-utils, then all time that i wanna hear a good music, i did should change my volume controls. Then i installed a alsa-utils in my PC, but when i installed this package by emerge, my sound don't work. Any error appear, any exception occour, but the sound don't go out. The equalizer it's work, only the sound don't go out OK, I'm very sorry to say I'm having a difficult time understanding this. So let me first ask if what I've figured out is what you're actually saying: 1. Your sound device is AC97/I82801, and you're using the kernel ALSA drivers to run it, but you did not have (at the time) any of the additional packages installed (such as alsa-utils). 2. Sound worked correctly, but mixer levels were not stored on shutdown or restored on startup (because you didn't have alsa-utils installed, which provides a script to do this). 3. You then installed alsa-utils, but now your sound doesn't work. this is where I get a bit confused there are no errors or exceptions, but you do not hear any sound. The equalizer (*which* equalizer??) works, but sound don't go out (means, you don't actually hear anything?) Anyone already passed for this problem? Remembering : my user it's on audio group: audio:!:18:habutre my sound device has this permissions: crw-rw 1 root audio 14, 3 Jan 16 21:58 /dev/sound/dsp my sound mixing it's: esound my alsa-utils version: 1.0.10 This all looks fine (except for maybe esound, but that's a choice and doesn't necessarily seem to have a bearing on this problem). I think the first thing I'd want to know is: have you actually run alsamixer to set your mixer levels; maybe they're all still muted?; The second thing I'd want to know is: can you hear anything if you turn off esd (esound) or set gnome/esd to use ALSA (never quite figured out how to do that, so I just turned esd off)? Sorry not to be more help, but I'm not sure you've given all the information needed to troubleshoot this (since I'm not sure I understand what the problem actually is). HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] How To set compile time options in emerge
Simon Prosser schreef: On Wednesday 18 January 2006 12:36, Oumar Ndiaye wrote: Hi, I installed php-4 via emerge but I am having problems getting php to work with mysql. After many research I have concluded that php-4 is not compiled with the -with-mysql option. I need to recompile php-4 with the -with-mysql option and reinstall it. How do I do that with emerge or some other ways? USE=mysql emerge dev-php/php Which will work for this emerge, but if php is then upgraded, the USE flag will be disabled again. A better way is to either 1. add the mysql USE flag to /etc/make.conf USE flags (if you want to enable mysql support for all packages that might support it); or 2. add dev-php/php mysql to /etc/portage/package.use (to enable mysql support for this package only). If the folder /etc/portage already exists, you can just do # echo dev-php/php mysql /etc/portage/package.use to enable the USE flag for that specific package. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] *.bin and /dev/loop help
krgn schreef: Symbol: BLK_DEV_LOOP [=y] Prompt: Loopback device support Defined at drivers/block/Kconfig:261 Location: - Device Drivers - Block devices mount -o loop ok, installed it as module and its there... snip but if I try to mount the iso image.. it outputs this - RockHead doom # mount -t iso9660 Doom3cd1.iso01.iso /mnt/iso/cd1/ -o loop=/dev/loop0 ioctl: LOOP_SET_FD: Device or resource busy so I wonder what this means. Karsten Wrong mount format. Try mount -o loop Doom3cd1.iso01.iso /mnt/some_folder_in_mnt (I think, working from memory) HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] more problems emerging world
Antoine schreef: Hi, I had put unicode in my make.conf, well, just cos unicode is a Good Thing. However, I now get audacity telling me that I must emerge wxGTK without unicode. Fine, I'll get rid of it I say to myself. Only I get ncurses telling me I will break my system if I emerge without unicode. Anyone got any suggestions? Cheers Antoine # echo x11-libs/wxGTK -unicode /etc/portage/package.use re-emerge (--oneshot) wxGTK emerge amarok Done. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] UDEV permissions
Tom Smith schreef: Well, I didn't emerge the Gentoo qemu--it is a few versions behind the official so I opted for using the official release (0.8.0, I believe). I don't know why you think this: motub - eix qemu * app-emulation/kqemu Available versions: 0.7.2 Installed: none Homepage:http://fabrice.bellard.free.fr/qemu/ Description: Multi-platform multi-targets cpu emulator and dynamic translator kernel fast execution module * app-emulation/qemu Available versions: 0.6.0 0.6.1 0.6.1-r1 0.7.0 0.7.0-r1 0.7.1 0.7.2 0.8.0 Installed: none Homepage:http://fabrice.bellard.free.fr/qemu/ Description: qemu emulator and abi wrapper meta ebuild I'm ~x86 so it's marked as available for me, but 0.8.0 is unstable for x86, ppc, and amd64. You must be running stable? Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] How can I unsubscribe?
Markus Döbele schreef: What do I have to do to unsubscribe this mailinglist? I tried everthing that the page tells me todo. I sent en empty mail to: [EMAIL PROTECTED] [EMAIL PROTECTED] Why do you not give even one example at the page how it should look like? And why is there not a field where I can enter my mailadress and then simply click Subscribe or Unsubscribe ? Perhaps the page has not been updated; the list address seems to now be gentoo-user@lists.gentoo.org, so *maybe* the unsubscribe address is now [EMAIL PROTECTED] You did try to unsubscribe from the same email address that you subscribed from, yes? Anyway, it's just an idea (since I've never tried to unsubscribe :-) ). HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Error compiling Kde-Graphics 3.4.3-r3
Thiago Lüttig schreef: Hi, i´m trying to compile the kde-graphics-3.4.3-r3 but it can´t find the libungif package. When I emerge that, the portage says it´s deprecated, and recently, simply doesn´t find it. How to fix this ?? Yes, libungif was replaced by giflib recently. Everyone who's advised to re-emerge imlib was on the right track, because imlib2 is what's compiled against libungif, but you don't have giflib at all, so the dependencies are not being recompiled, so of course it's not helping. You might want to have a look at http://bugs.gentoo.org/show_bug.cgi?id=83238 In any case, what you most likely want to do is 1) unmerge libungif 2) emerge --oneshot giflib 3) emerge --oneshot imlib2 4) re-emerge kdegraphics Or follow the various instructions in the bug . HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] mozilla firefox 1.5 and gnome
Trenton Adams schreef: On 1/14/06, Richard Fish [EMAIL PROTECTED] wrote: On 1/14/06, Trenton Adams [EMAIL PROTECTED] wrote: The old version could disable gnome, but 1.5 can't. Perhaps it's a firefox 1.5 ebuild limitation? I hope that's what it is, because I can't stand gnome. False. carcharias rjf # ldd /usr/lib/mozilla-firefox/firefox-bin | grep gnome carcharias rjf # emerge -pv mozilla-firefox These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild R ] www-client/mozilla-firefox-1.5-r9 USE=java -debug -gnome -ipv6 -mozdevelop -xinerama -xprint 0 kB If you already have gnome installed, gnome won't show up unless you go --tree, will it? Either way, you can take a look in the ebuild file for firefox 1.5, and there's no reference to gnome. No checking of the gnome useflag or anything like that. 1.0.7 has it though. The appearance of USE flags (which is what we're talking about) is unrelated to whether or not you have the package installed already-- a USE flag does not appear or disappear based on whether you have the relevant package installed. In fact, the USE flag controls to great degree whether you install the relevant dependency or not (i.e., if K3b optionally depends on KDE, then enabling the kde USE flag will install additional, optional KDE support programs. But since K3b requires kdelibs and kdebase to run at all, these hard dependencies are not controlled by the USE flag). In any case: emerge -pv mozilla-firefox gnome-light These are the packages that I would merge, in order: Calculating dependencies ...done! [ebuild R ] www-client/mozilla-firefox-1.5-r9 USE=gnome java xprint -debug -ipv6 -mozdevelop -xinerama 0 kB [ebuild R ] gnome-base/gnome-light-2.12.0 0 kB As you see, I do have GNOME installed, and the gnome USE flag does still appear in the Firefox prospective emerge output. | Runtime Dependencies | mozilla-firefox-1.5-r9 | | gnome = gnome-base/gnome-vfs - 2.3.5 | Title: gnome-vfs | Description: Gnome Virtual Filesystem So enabling the gnome USE flag adds a runtime dependency on the GNOME virtual file system. Which sounds an awful lot like something that might affect dialogs. However, since there were a lot of complaints about the old dialog, I strongly doubt that disabling this flag would change the dialogs terribly significantly, except that you wouldn't have the GNOME extras that I mentioned before. It's quite likely that even the naked new dialog is a lot like the current GNOME dialog. And again, the likely reason that there is no reference to GNOME in the ebuild is because the references to GNOME integration and compatibility are *in the source code*, because the *Mozilla developers chose* this dialog *as the default* (but not the only) choice of dialog. That is their good right, just as it is your good right to change that if you so desire (since that is apparently quite possible to do). Ebuilds don't do anything but compile the code as given, with some slight modification to work within the Gentoo system (as opposed to the RedHat system or the SuSE system). Ebuilds certainly don't change things as integral to the operation of the program as the format and type of file dialogs. I suppose you could trawl through the source and see where this is set and then write an ebuild to make the choice of file dialog type a configure option-- maybe. If it's already a configure option. Which I doubt. But you could look. Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] netscape-flash doesn't work with firefox 1.5
Iain Buchanan schreef: Hi, (up late I see :) It's only 1:20. Not all that late (but it will be in about a half an hour) On Mon, 2006-01-16 at 00:46 +0100, Holly Bostick wrote: the symlink is already there... Hope this helps somewhat (at least you know it can work) kind of :) it _was_ working once apon a time... I have the same version of flash also... what version of firefox are you running? 1.5-r9 (compiled, not -bin). Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Deer Park instead of FireFox
Sven Köhler schreef: Hi, i just installed mozilla-firefox-1.5-r9. So when i start it, it says hooray, you've got Deer Park Alpha 2 installed! Thank you, for helping us testing. So what is Deer Park? What's the difference to a normal FireFox 1.5? Or the main question is: Why does that gentoo-ebuild install Deer Park, and not just plain FireFox? Thanks Sven Deer Park is the codename for Firefox 1.5 while it was in development. According to this thread http://forums.gentoo.org/viewtopic-p-2934098-highlight-.html#2931694 on the forums, the reason that Firefox you install under Gentoo is not allowed to be called Firefox is that only official builds from Mozilla.org may use the --official_branding configuration flag that names the titlebars and whatnot. Builds you (or the dev team) built from source are not official, so therefore may not use this ./configure flag. However, you are free to copy the ebuild to your overlay, add the flag to the ebuild, and then recompile it. Instructions can be found in the thread. As far as I know, the application built is the final, but it appears that the release was released early since rc3 was so stable. Maybe a few typos slipped through. I certainly don't remember seeing such a message when I started Firefox 1.5 for the first time, but maybe that's because it was an upgrade or something. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] mozilla firefox 1.5 and gnome
Trenton Adams schreef: The old version could disable gnome, but 1.5 can't. Perhaps it's a firefox 1.5 ebuild limitation? I hope that's what it is, because I can't stand gnome. No, it has nothing to do with the ebuild, it's a choice of the Mozilla.org developers. But perhaps you missed Edwin Kapauni's post, linking you to a thread on the forums where they tell you how to change the dialog to either the old one, or even the KDE one. For ease of searching, here it is again: http://forums.gentoo.org/viewtopic-p-2934098-highlight-.html#2931694 Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] I can't send email anymore. O_O
Dale wrote: Well, I can't even get Mozilla to open any more. I even created a new user from scratch and it still will not start. I downloaded a new snapshot and am about to start a new install. I've never seen someone so eager to reinstall for so little. Honestly Mozilla won't start, so you're reinstalling the entire OS? How... Windows-like. I don't even know where to start to fix this one. First thing you might want to do is killall -9 mozilla-bin I have often found that if Moz (or Firefox) crashes, the process is still running, and therefore new instances won't open. Look in kdesysguard (or whatever the KDE System Monitor is called), or top, and you might see some 5 instances of mozilla-bin are already running (invisibly). Kill those, and Mozilla may well start. Whether your problem will be solved is another question, but we'll come to that, if not. HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] I can't send email anymore. O_O
Dale wrote: On Friday 13 January 2006 05:58, Holly Bostick wrote: Kill those, and Mozilla may well start. Whether your problem will be solved is another question, but we'll come to that, if not. HTH, Holly That would make sense. I killed the processes that were running and still get this: Oy, I see it now. The problem is here [EMAIL PROTECTED] / # su dale When you su, the X server is refusing to allow the user (dale) to connect to the running X server (which is owned by root). Now the solution for this is something that I normally do in the other direction (to allow root to connect to the user's X server); and while I'm sure it will work in reverse, I myself wouldn't suggest it since I don't necessarily want to impinge on user privilege separation if the base user in question is root. So what I would say is firstly why are you running as root and su-ing to a user, rather than running as a user and su-ing to root, since there's no reason whatsoever to do it this way (it's certainly not more secure, in fact, much less so afaics). Secondly, does Mozilla open as the user logged in (whether that's root or dale) But anyway, what you need to do, iirc, is to copy ~/.Xauthority from the logged-in user's HOME folder to the folder of the user who is the target of su-- normally root, so this would mean copying from Dale's home to /root... but you're running as root, and I just don't like the idea of copying .Xauthority from the /root folder to dale's HOME. So I would suggest running as a user and copying the file from dale's home to /root, after which root should not receive this error (if in fact su-ing to root and running Mozilla does in fact generate this error, which it may very well not). HTH, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] kde new user login icons missing
Shawn Singh schreef: I note that on my system the range for UIDs that won't show up is UID 1000 and UID 65000. If your users had a UID inside of that range it might show up... I'm not sure b/c on my system they are w/i that range and I don't see an icon for them, but they can key in their credentials to authenticate. They aren't by any chance explicitly hidden? In the KDE Control Center, System Administration, Login Manager, Administrator Mode, Users tab, there is a big list with checkboxes for all the users. What is kinda counter-intuitive is that this is not the list of users *shown* in KDM, but the users who will be hidden if their checkbox is checked. So if you perhaps checked these users in an attempt to unhide them, you have in fact hidden them from the KDM screen, afaik, not being a KDM user under any circumstances, but I do have a minimal KDE installed, and I've looked at the Control Center a number of times (I did try KDM once or twice). I'd also suggest checking whether icons and the like are properly set (on the same tab). Anyway, hth, Holly -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] mozilla firefox 1.5 and gnome
Trenton Adams schreef: Hi everyone, I'm just curious about something. I installed mozilla-firefox-bin, and I got what I *thought* was the gnome open/save dialog. And I absolutely I can't stand that thing. So, I decided to compile mozilla-firefox for myself, as I have -gnome in my use flags. Well for some reason, I still have that open/save dialog. So, is that a gnome open/save dialog, or is it something else? Yes, as far as I know, it is (though I myself do have GNOME installed). Firefox depends on gtk: (from www.gentoo-portage.com ; the compile-it-yourself version used for clarity, since even the bin has to be originally compiled by someone): Runtime Dependencies mozilla-firefox-1.5-r9 |app-arch/unzip |app-arch/zip |dev-libs/expat |= dev-libs/glib - 2.8.2 |= dev-libs/libIDL - 0.8.0 |= media-libs/jpeg - 6b |= media-libs/libmng - 1.0.0 |= media-libs/libpng - 1.2.1 |= sys-libs/zlib - 1.1.4 |= www-client/mozilla-launcher - 1.42 |= www-client/mozilla-launcher - 1.39 |! x11-base/xorg-x11 - 6.7.0-r2 |= x11-libs/cairo - 1.0.0 === x11-libs/gtk+ - 2.8.6 |x11-libs/libXmu |x11-libs/libXmu |x11-libs/libXrender |x11-libs/libXrender |x11-libs/libXt |x11-libs/libXt | = x11-libs/pango - 1.10.1 |gnome = gnome-base/gnome-vfs - 2.3.5 |virtual/x11 |virtual/x11 |java virtual/jre Firefox is a GTK application, so it uses the GTK toolkit (as opposed to the QT toolkit, which KDE uses). GNOME is a GTK-based DE (as opposed to KDE which is QT-based). So the reason why the open/save dialogs are the same is because they use the same resources (GTK settings, many of which can be set in GNOME). I can't say if the extra features that I have because I have GNOME installed are present in your open/save dialog (bookmarked folders, for example). Is there going to be a way of NOT making it the standard open/save dialog when firefox 1.5 gets unmasked? Masking or unmasking is not going to change the source code, which depends on GTK. Only upstream can recode the program with a different toolkit. And that seems unlikely to happen, since it hasn't happened yet (and if the toolset was going to change, it probably would have before 1.0, not suddenly now when the release is finally stable enough to be picking up steam fast in the popularity stakes). Holly -- gentoo-user@gentoo.org mailing list
