Re: [arch-general] How to print to parallel printer with cups support removed?
On di, 2012-08-07 at 09:06 -0500, David C. Rankin wrote: > Guys, > > How do we print to parallel printers now with cups 1.6.x? By installing cups-filters. The backends have moved over there.
Re: [arch-general] is scsi hot swap broken?
On di, 2012-05-22 at 16:05 -0600, Sergey Manucharian wrote: > The controller is : > 00:1f.2 IDE interface: Intel Corporation N10/ICH7 Family SATA Controller > [IDE mode] (rev 01) IDE mode isn't supposed to support hotswap. You'll need AHCI or RAID mode for that. What driver is loaded for your SATA controller?
Re: [arch-general] ../libtool: line 2083: ark_part.cpp: command not found?
On ma, 2012-03-19 at 10:42 -0500, David C. Rankin wrote: > Anybody familiar with libtool know if this is a libtool problem or if this > is > a problem with the tdeutil code? If it is libtool, any work-arounds? > Either you're missing a command, or more likely, the fork of libtool included in the source differs from the libtool version on your system. Running "libtoolize --force", followed by all the autofoo afterwards, would solve your problem I guess.
Re: [arch-general] tp_smapi - LENOVO g470
On do, 2012-01-19 at 14:59 -0300, Paulo Roberto P. Evangelista wrote: > Yes, i did it > Atts, > > Paulo Roberto > > > On Wed, Jan 18, 2012 at 12:52 PM, Jan de Groot wrote: > > On Wed, 2012-01-18 at 12:43 -0300, Paulo Roberto P. Evangelista wrote: > >> Yes, > >> I'm sure this pacman -U > >> > >> Atts, > >> > >> Paulo Roberto > >> > >> AVISO LEGAL > > > >> > >> LEGAL ADVICE > > > > > > Please, when you mail to this list, why do you insert this stupid legal > > statement? You're mailing to a public mailinglist that is put into a > > public archive which is indexed by google. > > Also, please don't top-post, bottom-posting is standard in mailinglist > > conversations. > > > > Anyways, to get back on your question: when you modprobe the module, it > > says "no such device". That means that the driver is either not suitable > > for your hardware, or the driver doesn't support your hardware (yet). > > Try to find out by reading "dmesg | tail" output instead of looking for > > missing files. Some modules require extra options for this to work. > > > So, you did what? You didn't read the message at all, because you're still top-posting and you're still including 100 lines of bullshit in every mail you send to this list. Anyways, dmesg | tail should contain enough pointers to your solution, but as you didn't share the output of that with us, we can't help you.
Re: [arch-general] tp_smapi - LENOVO g470
On Wed, 2012-01-18 at 12:43 -0300, Paulo Roberto P. Evangelista wrote: > Yes, > I'm sure this pacman -U > > Atts, > > Paulo Roberto > > AVISO LEGAL > > LEGAL ADVICE Please, when you mail to this list, why do you insert this stupid legal statement? You're mailing to a public mailinglist that is put into a public archive which is indexed by google. Also, please don't top-post, bottom-posting is standard in mailinglist conversations. Anyways, to get back on your question: when you modprobe the module, it says "no such device". That means that the driver is either not suitable for your hardware, or the driver doesn't support your hardware (yet). Try to find out by reading "dmesg | tail" output instead of looking for missing files. Some modules require extra options for this to work.
Re: [arch-general] Can't build libwebkit
On Fri, 2011-09-09 at 16:45 +0200, Andrea Scarpino wrote: > On Friday, September 09, 2011 11:33:10 AM Bastien Dejean wrote: > > makepkg stops again: > > > > In file included from > > ../Source/WebCore/bindings/js/JSCanvasRenderingContextCustom.cpp:32:0: > > ../Source/WebCore/html/canvas/WebGLRenderingContext.h:32:31: fatal error: > > GraphicsContext3D.h: No such file or directory > I just built both extra and gnome-unstable versions without any issue. If you > are enabling some extra feature, check the requirements. This is caused by --enable-webgl, the release tarballs don't have all the required files in them.
Re: [arch-general] Unable to use tape drive
On Wed, 2011-08-24 at 14:42 +0200, Vitor Garcia wrote: > Any ideas? Yes, provide us with the full outputs of dmesg and lspci. We can't do much with some random greps.
Re: [arch-general] Unable to use tape drive
On Wed, 2011-08-24 at 08:49 +0200, Vitor Garcia wrote: > Good morning. > > I have been trying to my Dell LTO-120 SCSI Tape Drive on Arch without > success. > > I have tried to run $ modprobe st, but still, I have no /dev/st0 and > neither /dev/nst0. dmesg | grep -i tape shows nothing too. > > Is there anything I'm missing here? Dell's user manual says that the > drive should work out of the box on linux. Please paste your dmesg output somewhere. The SCSI controller that connects the drive is probably not even recognized.
Re: [arch-general] Unable to use tape drive
On Wed, 2011-08-24 at 20:11 +0530, Madhurya Kakati wrote: > On Wed, Aug 24, 2011 at 12:19 PM, Vitor Garcia > wrote: > > Good morning. > > > > I have been trying to my Dell LTO-120 SCSI Tape Drive on Arch without > > success. > > > > I have tried to run $ modprobe st, but still, I have no /dev/st0 and > > neither /dev/nst0. dmesg | grep -i tape shows nothing too. > > > > Is there anything I'm missing here? Dell's user manual says that the > > drive should work out of the box on linux. > > > > > > Best regards, > > Vitor > > > > Offtopic: How much does one tape store? How many MBs? http://en.wikipedia.org/wiki/Linear_Tape-Open The LTO-120 is an LTO4 drive, storing 800GB native, sold as 1.6TB compressed.
Re: [arch-general] Enforcing CFLAGS in PKGBUILDs
On Fri, 2011-08-05 at 15:23 +0800, Ray Rashif wrote: > For eg. some developers like to enforce -O3, so they should first get > the system CFLAGS and override it's -O*, if any. > > But in general, I agree. We shouldn't enforce anything either unless > we're trying to fix something. The ardour PKGBUILD does this [1], > maybe it shouldn't, but I assume the -O3 becomes redundant when we > pass system CFLAGS to the build as a configuration flag. > > [1] > http://projects.archlinux.org/svntogit/packages.git/tree/ardour/trunk/PKGBUILD Looking at ardour, I see it's also patched to support gcc 4.6. Now let's say something about -O3: -O3 optimization can be very bad and the results are extremely dependent on the compiler version. Something that works optimal with gcc 4.4 and -O3 can break completely with gcc 4.6 and -O2. Just adding -O3 to CFLAGS without knowing which compiler is used is plain stupid. Check this blog for example: http://glandium.org/blog/?p=1975 There someone from Mozilla claims they're finally able to build with O3 +PGO with GCC 4.5. This was at the time we were already doing O2+PGO from our PKGBUILD with 4.5 and now 4.6.
Re: [arch-general] EveryDNS compulsory migration to Dyn
On Sat, 2011-06-11 at 10:16 +0800, Abdul Halim wrote: > I can't help but notice that ArchLinux is using EveryDNS for its DNS server. > And since I also use EveryDNS DNS server, what is the plan for > ArchLinux domain administrator since EveryDNS users need to migrate to > Dyn by 31-Aug-2011? > Will the administrator use the paid service or something else? We're already working on that issue. Don't know what will be decided, but we'll migrate to a working solution before that date. Could be Dyn, but could be some other host too. Money isn't such a big issue, we've donated to EveryDNS before to get additional features also.
Re: [arch-general] [arch-dev-public] Status of kernel 2.6.39?
On Tue, 2011-06-07 at 14:02 +0200, Tobias Powalowski wrote: > Am Dienstag 07 Juni 2011 schrieb Mark Foxwell: > > # sending to arch-general as I can't post to arch-dev-public > > > > Tobias Powalowski wrote: > > > - lirc does not compile, anyone who uses lirc might take a look at it > > > > > > We have no maintainer listed, this time no commits on their git > > > happened lately to fix compile errors. > > > > I created lirc-utils-git [1] in AUR to work around the compile errors. I > > don't need the main lirc package as my remote is supported in the > > kernel. > > > > To compile I had to remove the lirc_atiusb, lirc_i2c and lirc_wpc8769l > > drivers. > > > > Hope this helps, > > > > Mark > > > > [1] https://aur.archlinux.org/packages.php?ID=49301 > But those are the most needed ones. Well, in fact they're the only drivers in the lirc kernel driver package. All the other drivers are merged with the kernel, either as normal driver or as staging driver.
Re: [arch-general] hwclock and openntpd
On Sat, 2011-06-04 at 03:06 -0400, Yclept Nemo wrote: > Perhaps openntpd does not set the hwclock. Therefore, should openntpd > be used in conjuction with the hwclock daemon? That's true, and that's also the reason why Openntpd doesn't play well with Xen where all guest VMs will take over the clock from the host VM. With the normal ntp distribution that works, but with Openntpd you need something to push the changes to the hwclock.
Re: [arch-general] How to install a group when a package with same name exists?
On Fri, 22 Apr 2011 14:32:07 +0200, Cédric Girard wrote: Hi, Today I wanted to install the group (xorg-apps). But there is a package with the same name. I did not find a way to tell pacman I wanted the group and not the package. I ended doing this: sudo pacman -S `pacman -Sg xorg-apps | sed 's!xorg-apps \(.*\)!\1!'` But there must be an easier way. Any thought about this? Regards, Hm, crap, didn't think of that when I changed the groups and replaced that dummy package... Anyways, I think the only way around this is by making the package disappear from extra.db.tar.gz on your system and then install the xorg-apps group. I'll get around removing xorg-apps soon, there's still a package in the repositories that depends on it.
Re: [arch-general] seahorse-tool: problem with GTK+ 2.x and GTK+ 3 symbols
On Thu, 2011-04-14 at 12:29 +0200, F. Gr. wrote: > Hi, > there's an error when I've been trying to decrypt a file. > > --- > % seahorse-tool -d cont.ods.pgp > > Gtk-ERROR **: GTK+ 2.x symbols detected. Using GTK+ 2.x and GTK+ 3 in > the same process is not supported aborting... > [1]5404 abort (core dumped) seahorse-tool -d cont.ods.pgp > --- > > It has been built a file named 'core' (size: 1.1 MB). > > --- > % file core > core: ELF 64-bit LSB core file x86-64, version 1 (SYSV), SVR4-style, > from 'seahorse-tool -d cont.ods.pgp' > --- Seahorse-plugins needs an update. The version in the repositories is built against seahorse 2.32.x which links to gtk2, the seahorse version in testing links to gtk3.
Re: [arch-general] Gnome 3 + KDE 4 are both large disappointments.
On Sun, 2011-04-10 at 21:50 +0200, Dennis Beekman wrote: > 1. You cannot change the panels anymore you stuck with the 2 given by > gnome 3. > 2. Changing themes also is inpossible.. or so it seems. > 3. Why do we need a system settings menu with all the options in one > menu ? where are my seperate icons i love so much ? why can we choose > wich icons or options we want ? > 4. What about the people ho don't have or don't wich to use they're > video hardware to run the these stupid graphics ... are we stuck with > "fallback mode" wich is even more stupid and backward ? > 5 Where did all the nice applets go ? and why can i not add them to > my > taskbar anymore 1. Try pressing alt. It should be documented somewhere, but it's in the NEWS file for gnome-panel at least 2. The UI was removed, but you can still set things in dconf and gconf. When Adwaita looked like crap in fallback mode, I've used Mist for a long while. 3. Design choice, Mac OS X has this too and in the view of gnome-shell, this is a logic choice. The control-center with grouped applet view already existed for a long time in 2.x versions also, but only SuSE used it I think. 4. Fallback mode is just like classic GNOME in looks, though you have the new theming and new apps. I used fallback mode for a long while and I'm still forced to do so on this desktop, the step from 2.x to 3.x in fallback mode is very small 5. Applets are not nice, but for the icons and things that were ported to the new panel-applet API, you can just press alt while right-clicking your panel to add it like you used to do.
Re: [arch-general] Using "python3 setup.py" in PKGBUILDs
On Mon, 2011-04-11 at 12:06 +0800, Auguste Pop wrote: > relinking python is a bad idea, but imho, explicitly envoking python3 > when packaging it not. > using python to install python3 packages is like linking to foo.so > instead of foo.so.3. Which is what is done by default when you tell gcc to link -lfoo ;)
Re: [arch-general] gtk3-engines doesn't work
On Mon, 2011-04-04 at 23:35 +0800, Heiher wrote: > Hello, > > Could you tell me why gtk3-engines doesn't work? Seems the gtk3 not > load it by 'strace'. Gtk3 uses a different interface for theme engines. Gtk3-engines is dead, it hasn't been ported and still uses the old interface. So far I only see translations getting updated. http://library.gnome.org/devel/gtk/unstable/ch25s02.html#gtk-migrating-GtkStyleContext
Re: [arch-general] [signoff] kernel26-2.6.38.2-1
On Mon, 2011-04-04 at 10:08 +0200, Adrian C. wrote: > plain text document attachment (message.txt) > On Thu, 31 Mar 2011, Tobias Powalowski wrote: > > > Hi guys, > > please signoff 2.6.38 series for both arches. > > Can't signoff i686. LILO can not boot it. > > # /sbin/lilo -v > ... > Fatal: Setup length exceeds 31 maximum; kernel setup will overwrite boot loaer > > Machines would be left in a constant state of reboot if it is missed by > a user. Is this related to the fact we compress kernels with xz now?
Re: [arch-general] Can a ISP modify incoming http traffic to display ads using transparent proxy server ?
On Sat, 2011-04-02 at 12:15 +0530, Partha Chowdhury wrote: > My ISP is running a transparent proxy server on port 80 as I found out > from this thread > > http://mailman.archlinux.org/pipermail/arch-general/2011-March/019172.html > > Now from last two or three days, I keep getting pop-up ads from a > particular domain on every website i go to - even on text only websites > ! If i choose another proxy server in firefox , then that does not > appear on any site. But i normally connect to internet, that specific > pop-up to a very particular website appears on every single page. If I > install adblock plus in firefox, then of course no ads are displayed. > > So is it possible that my ISP is serving that ad by intercepting > incoming http traffic ? Yes, proxyservers can intercept pages and modify them. I used to have a transparent squid proxy that would filter ads and replace them with a 1x1px transparent pixel. If you're paying for this internet line, I would suggest to switch to a different provider. Inserting ads in HTTP pages for a connection that is being payed for is not-done.
Re: [arch-general] Port 80 is shown open in port scan without any web server running
On Wed, 2011-03-30 at 17:27 +0200, Philipp Überbacher wrote: > If you live in a civilized country in Europe data retention either is > already in place or will be rather soon. The US might have a different > approach but I doubt the net result is much different. Those regulations are about email only. Providers have to store recipient and sender for every mail that passes their network and they have to store it for a long time, depending on what country implemented the rules.
Re: [arch-general] Port 80 is shown open in port scan without any web server running
On Wed, 2011-03-30 at 15:45 +0530, Partha Chowdhury wrote: > So it seems my ISP is running squid version 3.2.0.4-20110203 in > transparent mode , just like you said. > > Interestingly when connecting to random ip addresses on port 80, the > error page returned is quite different from normal ones. > > http://www.freeimagehosting.net/image.php?280f0ef980.png > > Does this transparent proxy pose any threat and what can I do to > stop > that ? The threat here is that your ISP will log every page visit you do and also has the ability to block certain websites. The only thing you can do is setting up a tunnel or using a different proxyserver that you trust.
Re: [arch-general] Port 80 is shown open in port scan without any web server running
On Wed, 2011-03-30 at 14:06 +0530, Partha Chowdhury wrote: > Now with current provider port 80 is shown open in every > port scan test This is usually caused by a transparent proxy. When nmap hits port 80, it will get redirected to the proxy server. Try doing an nmap -sV and you'll see what software is running on the proxyserver.
Re: [arch-general] Web sites down?
On Fri, 2011-03-18 at 10:34 +0530, Madhurya Kakati wrote: > 2011/3/17 Cédric Girard : > > 2011/3/17 Cédric Girard > > > >> I cannot access (timeout) any archlinux web site except AUR > > > > > > Well, seems back. Sorry for the noise. > > > > -- > > Cédric Girard > > > > Next time please wait for sometime before reporting on mailing list. :) The mailinglist was down also, so reporting it here doesn't help that much.
Re: [arch-general] Can't connect to WEP-protected network
On Thu, 2011-03-17 at 11:41 -0400, John Albright wrote: > The channel is 11, and I just realized that I can only specify 1-4 in > my > connection settings. So maybe that's the issue. However, I'm just > going to > skirt around the issue and change the network to WPA instead. I didn't > at > first because it's not my network, but I've been given the go-ahead. Check your wireless regulatory domain in /etc/conf.d/wireless-regdom. BTW: top posting is annoying.
Re: [arch-general] xorg-font-utils package reorganization
On Wed, 2011-03-16 at 21:09 +0200, Grigorios Bouzakis wrote: > Hi, > the previous xorg-font-utils package included 2 utilities that are not > available anymore. > fonttosfnt and ttmkfdir2 can not be found as split packages but AFAIK > they are not part > of any other joint packages either. > What happens to them? Will they be available in the future or they're > dropped completely? ttmkfdir is a redhat-specific tool, I haven't seen other distributions including that. I decided to drop that one. The fonttosfnt tool is deprecated so I decided to drop that unless someone might really feel they need it.
Re: [arch-general] xorg group name messed up after latest xorg package update?? PKGBUILD depends=('xorg') triggering complete reinstall..
On Tue, 2011-03-15 at 15:32 -0500, David C. Rankin wrote:
> Problem with depends=('xorg') is NOT limited to this one box. I have
> added the
> info to bug:
Sorry, but depending on a group is plain stupid. I didn't even know
pacman could handle that.
Re: [arch-general] xarchiver patch
On Wed, 2011-01-26 at 21:04 +0100, Andreas Radke wrote: > Am Wed, 26 Jan 2011 20:35:51 +0100 > schrieb Sven-Hendrik Haase : > > > You should bug upstream against, perhaps using direct mail to > > maintainer. > > > > In general, you can post up feature requests but they might be closed > > with "upstream" as a reason as we can't do much about it. We will not > > have feature patches in 99% of the cases if it can be helped. > > > > You seem to be able to read my mind? Yes, I'd close it as won't > implement. Though it would be really useful it's simply against Arch > philosophy as it would add unneeded code to the upstream release. Depends on what you call unneeded. If I would add support for additional compression formats, I would just rewrite a big piece of the code to use either bsdtar instead of parsing output from tar and unzip. File-roller is very famous for its error messages about .tar.gz files that have been un-gzipped by your browser, but which are stored with the same extension. As tar will complain that it's not a gzip archive, file-roller can't open it... The same will be true for xarchiver.
Re: [arch-general] [arch-dev-public] [signoff] kernel26-lts 2.6.32.27-1
On Mon, 2010-12-13 at 07:42 -0600, David C. Rankin wrote: > [31.342] drmOpenByBusid: Interface 1.4 failed, trying 1.1 The radeon driver segfaults when KMS is disabled, so yes, that's normal.
Re: [arch-general] Patent protection
On Wed, 2010-12-08 at 19:47 -0200, Armando M. Baratti wrote: > It works because the patents are granted perpetually to OIN members > that > have jointed before the closing date, even in the case of a sell > (like > occurred in this case). If Microsoft wants to get you, they will do so anyways, OIN member or not. OIN does not have a patent license for ALL patents that are being sold, and even if they had them, Microsoft has several other trivial patents like the ones on VFAT that can be used to sue everybody. Newsflash: they don't. Remember Google getting sued by Oracle? Both are OIN members, it's just a fact that not all patents owned by Oracle are shared with OIN, so Oracle is free to sue any OIN-member that uses those patents without a license.
Re: [arch-general] Pulseaudio
On Mon, 2010-11-29 at 17:25 +0800, Ng Oon-Ee wrote: > Good luck getting pulse-haters to understand that a 1.14 MB package > isn't going to kill their system though. > > Thanks JGC for your work on this. I didn't think it was possible to > have > such choice, and it looks like you've jumped through quite a few > packaging hoops to get here. I haven't done anything in this, as I have always avoided pulse like the plague. It was Jan Steffens who did all the work. I brought him on the team to take over GNOME together with Ionut, so I can focus on Xorg with my limited time.
Re: [arch-general] Pulseaudio
On Mon, 2010-11-29 at 10:09 +0100, Philipp Überbacher wrote: > I guess I am one of those pulse-haters. I don't care whether it's in > [extra], some other official repo or not since I simply don't need it. > But now mplayer pulls in libpulse, and I have no idea which consequences > this could have. I don't see why I need to have libs for a > soundserver that I have no use for floating around on my machine. It is > at best unnecessary and does nothing, at worst.. I don't know. I hope I > don't need to install GNOME to turn it off or something.. (gconf and > stuff). Sorry, but this is plain. When GNOME switched to pulse, we made the choice to boycot it and patch our applications to use GStreamer instead. We still do that with our GNOME packages, as we still don't want to force pulseaudio on systems. The reason for libpulse is that without pulseaudio installed, it will not have any function besides offering optional support for pulseaudio. No pulseaudio with libpulse just means no pulseaudio. We are the first binary distribution that offers you the complete choice of pulse or not. There's not any binary distribution out there that dares to implement pulseaudio this way. Either they force it up your ass completely, which is the way upstream wants, or the don't support it at all, which we used to do before. If you don't want that tiny lib on your system, be my guest, recompile all your mediaplayers to get rid of all those optional codecs for media you don't use. Those are useless libs too, but somehow nobody complains about that.
Re: [arch-general] PulseAudio in [testing]
On Fri, 2010-11-26 at 21:34 +0100, Jan Steffens wrote: > I've put PulseAudio into [testing] some time ago, but received no > feedback on this. > > This either means it works well, or nobody uses it. > > So I'm asking for it now, especially from GNOME and KDE users running > PulseAudio. Tested yesterday: I was running a pulse-free installation on my laptop. The changes you made in testing don't affect pulse-free systems running GNOME. I tried switching to pulseaudio by installing the pulseaudio-gnome group. That causes a conflict between the gstreamer0.10-pulse package in community and the gstreamer package containing the plugin now, but besides that, no problems have been found and everything works as expected.
Re: [arch-general] [signoff] kernel 2.6.36.1-3
On Thu, 2010-11-25 at 18:36 -0600, David C. Rankin wrote: > Tobias, > > You'da Man! kernel26 2.6.36.1-3 boot FINE on that whacky MSI based box. > I don't > know if you know what changed, but if so, I'm dying to find out. After running > it by the Redhat dmraid (dm-devel) list, the kernel.org list, all I got back > were the ideas I passed on here, but no real answers > > I updated my primary box (Tyan based with nv dmraid) that had shown the > the > weird drive initialization that caused the random read/write head excursions > that happened if the box didn't boot just right -- and it booted perfectly, > first try with kernel26 2.6.36.1-3. So I decided to give the MSI based box a > try. It too -- booted just fine! For that boot problem, you really should try grub2 instead of grub. Last week we had a bug on the bugtracker that looked like your bug, but with a regular installation with / on ext4. Grub couldn't boot it with the same error message as you're facing now and then. After switching to grub2, things worked fine. I think grub is bugged somehow with its ext3/ext4 drivers, causing issues like yours. With your current grub installation, every kernel update could result in a surprise.
Re: [arch-general] [PATCH] add upgpkg
On Mon, 2010-11-01 at 16:25 +0100, Andrea Scarpino wrote: > On Monday 01 November 2010 14:38:23 Florian Pritz wrote: > > +if [ -x "rebuild" ]; then > > + ./rebuild > > +else > > + makepkg > > +fi > Which official script is named 'rebuild'? > And, this script does not use a chroot to build the package! And official packages should never ever just "makepkg -g" without actually checking the checksums generated by that. I always copy&paste the sha1sums or sha256sums into the PKGBUILD from announcement mails, so I know that a downloaded tarball is the same one that is announced in my mailbox.
Re: [arch-general] Benchmarks (GUI)
On Fri, 2010-10-08 at 22:34 +0800, Ian-Xue Li wrote: > On Fri, Oct 08, 2010 at 04:32:26PM +0200, Manne Merak wrote: > > Anyone know of some reliable benchmark utilities I can use to > > evaluate display cards? and different drivers (mostly Nvidia on KDE > > 4.5). > You can try glxgears for a start. Or Phoronix Test Suite. > glxgears is not a benchmark and shouldn't be used for that purpose at all.
Re: [arch-general] umount usb stick via udev
On Thu, 2010-10-07 at 12:00 -0600, Sergey Manucharian wrote: > Thunar depends on HAL, so probably it is not correct to use it without > HAL, and the following is from http://thunar.xfce.org "Use the correct > HAL property to determine > whether a given volume or device requires eject" It can be used without hal, but not in the way that you can unplug a device without unmounting it. There's some technological difference between unplugging with and without hal. With hal: - device mounted - thunar open - device ejected without unmount - hal receives uevents and knows the device is dead - hal tells thunar that device is dead - thunar (and other hal or udisks-enabled programs) tries to close all references to the mountpoint - hal unmounts the dead device Now if you do that without hal, the point where hal detects that the device is dead and thunar getting to know that has to be done by yourself. If you forget that, you're screwed.
Re: [arch-general] Dovecot: bug or user error?
On Thu, 2010-10-07 at 13:16 -0300, Norbert Zeh wrote: > After some long debugging sessions, I realized that > my failure to connect to my freshly setup dovecot server was because > dovecot-auth failed. It was looking for the mysql and postgreql > shared > libraries that were not installed. If dovecot-auth links to the mysql and/or postgresql libraries, those libraries should not be in optdepends, but in depends. The problem with adding link-level dependencies for optional components to optdepends is that namcap doesn't know why you added the optdepend, and doesn't report any warning about link-level dependencies for optdepends. We've seen these bugs in several applications, so maintainers should really pay attention to this.
Re: [arch-general] Question about pacman and installing Gnome
On Mon, 2010-10-04 at 17:15 +0200, Christian wrote: > I wait for about 30 seconds before I login as my user but nothing > happens when I press enter after I have entered my password. gdm has this annoying change since 2.32 where it requires a mouse click on the username to select it for login. With previous versions you could just hit enter to select the one and only user in the list and enter your password after that.
Re: [arch-general] xorg 1.9 breaks
On Fri, 2010-10-01 at 04:43 +0300, kurrata wrote: > Or we could just add conflict in pacman output if person has nvidia > driver and stoping the upgrade. Or just write warning message that > the > new xorg doesn't support nvidia 173xx drivers and will result with > break. > That way others could upgrade to the new xorg and nvidia users > wouldn't > break their machine I removed that from xorg-server 1.9, simply because it's unmaintainable that way. I can't remember each and every driver that doesn't work with a new xorg-server version, and I don't have all the hardware to try that either. When we dropped catalyst, we added a versioned conflict on the released driver at that time. That appeared to work fine... until AMD released a newer version that STILL doesn't support the new Xorg server. The same thing will happen with Nvidia if we start doing that.
Re: [arch-general] Cairo and XCB
On Thu, 2010-09-30 at 14:46 +0200, Damjan Georgievski wrote: > Does anyone have reproducile tests that show the problems in > Firefox/Thunderbird with cairo when cairo is built with --enable-xcb? > > I just built cairo with: > --enable-xcb --enable-xlib-xcb=no > > This should leave the normal cairo xlib backend intact (i.e. the same > as when built with --disable-xcb). > So far everything is working fine for me, but even before, with Cairo > 1.8 everything was working fine, > so I'd like to doublecheck about those issues. Upstream does not support it and still marks the XCB backend as experimental. Though some work has been done on the backend for 1.10, there's no dedicated maintainer for it. I simply don't want that in official packages.
Re: [arch-general] Removing HAL
On Fri, 2010-09-10 at 11:07 +0600, Галымжан Кожаев wrote: > Hi list. > Few days ago I've succesfully installed arch. It was booting really fast > first time (i.e. without gnome, networkmanager, etc). > After installing GNOME and some daemons, adding them to the rc.conf, the > system now takes 30-40 seconds to load. > I heard that HAL daemon takes a lot of time during boot and it's > functionality can be replaced by udev. > How could I completely remove HAL and still use GNOME without troubles? Is > it possible? > I have: > - GNOME 2.30 > - xorg 1.8 > - hal 0.5 You can remove HAL from DAEMONS, as GNOME doesn't use it. Only gnome-vfs enabled applications can still use it, but GNOME itself doesn't contain such applications anymore.
Re: [arch-general] sudoers file change - not much on dev list - reason for changes?
On Wed, 2010-09-08 at 03:25 +0100, Mario Figueiredo wrote: > I confess I'm baffled as to why upstream did this change, why it was > needed and if it actually was needed. > I can't seem to find anything about it in the online changelog > either. > Seems something completely arbitrary. Or an oversight. Or a mistake. I'm baffled by the fact that people make a huge issue about a comment for a possible configuration in /etc/sudoers. The change doesn't introduce anything, it just adds an example of something you could possibly implement. This is actually the same reason why I'm annoyed by bugreports that request a different order of comments or examples in the kernel26 PKGBUILD for configuration. IMHO we should just remove that completely and not care about such things, I mean... I don't add "#add --enable-xcb to enable XCB" in the cairo PKGBUILD either.
Re: [arch-general] [Packages] [dbus-core] DeveloperWiki:CoreSignoffs wiki needs update
On Sun, 2010-09-05 at 22:34 +1000, Allan McRae wrote: > So you are saying that everyone should be able to signoff? I think > the > main point is everyone can block a signoff by filing a bug report. The funny thing with this specific case is that I didn't even ask for signoff. I asked on the bugtracker where this bug has been discussed to see if the bug has been fixed without side-effects. The only comment I received in days is "Why didn't you update to 1.3.1?". If this package was really so important, why is nobody, including the KDE maintainers that filed it, focused at getting this package to core?
Re: [arch-general] Abiword and wv
On Tue, 2010-08-24 at 02:39 -0500, Myra Nelson wrote: > [m...@gandalf abiword > unable to find libwv-1.2.so.3 > > My troubleshooting went like this: > > [m...@gandalf /usr/lib]:locate libwv > /usr/lib/libwv-1.2.so.4 > /usr/lib/libwv-1.2.so.4.0.3 > /usr/lib/libwv.a > /usr/lib/libwv.so This is because you built your own version of wv. Upstream did a useless soname bump without reason, which is why I reverted that (several other distributions do that also). This change saves me a bunch of rebuilds, but will cause you pain if you didn't stick to official packages.
Re: [arch-general] [PATCH 28/48] Use bash-style conditionals when setting up the hardware clock.
On Thu, 2010-08-19 at 00:56 -0400, Dave Reisner wrote: > Couldn't we avoid all this by just flipping a switch in the kernel? > > CONFIG_RTC_DRV_CMOS=y > > If it's compiled into the kernel, udev picks it up and creates > the /dev > nodes for us. Which still locks out the people who use a custom kernel with this driver compiled as module. IMHO the init scripts should work with both module and built-in.
Re: [arch-general] Referencing $srcdir in PKGBUILD?
On Tue, 2010-08-17 at 10:05 +0100, Magnus Therning wrote: > Lately my PKGBUILDs result in a warning when building: > > ==> WARNING: Package contains reference to $srcdir > > It's a bit confusing, since PKGBUILD(5) has the following to say about > $srcdir: > >srcdir >This points to the directory where makepkg extracts or > copies all source files. > > What should I do to get rid of the warning? Depends on your package. Somehow your $srcdir of $pkgdir ended up in one of the installed files in your package. Use grep to find references to it (the full path, not the variable name) and find a way to get rid of those. What package are we talking about btw?
Re: [arch-general] mayday - new updates killed httpd - make_sock: could not bind to address [::]:443
On Mon, 2010-08-09 at 04:49 -0500, David C. Rankin wrote: > [04:45 nirvana:/home/david] # rccups start > /etc/rc.d/functions: line 303: .: /etc/rc.d/functions.d/functions.d: > is a directory > :: Starting CUPS Daemon > [DONE] > > Now how do I fix this in rc.conf to make sure cups doesn't > screw me over again? > Take http out of the background? > > I've booted this box 15 times or so in the past 1.5 years and > this has never > been a problem until the updates tonight. What changed? How many > others are > going to get caught? First of all, why don't you fix that annoying error with /etc/rc.d/functions.d that pops up on every rc script invocation? As for cups: the default cupsd.conf contains "Listen localhost:631" and "Listen /var/run/cups/cups.sock". I think your cupsd.conf contains an entry for 443 also there. The problem with your setup is that your setup relies on a failure: apache binds to https before cups can, so cups doesn't bind to 443 anymore. Because you're using backgrounding, any update to your system can interfere with timing and cause issues like these if your configuration is not right.
Re: [arch-general] mayday - new updates killed httpd - make_sock: could not bind to address [::]:443
On Mon, 2010-08-09 at 04:19 -0500, David C. Rankin wrote: > (98)Address already in use: make_sock: could not bind to address > [::]:443 > (98)Address already in use: make_sock: could not bind to address > 0.0.0.0:443 Maybe you should read the error messages and check out with tools like lsof what process is using that port.
Re: [arch-general] Shouldn't pacman restart dovecot after update?
On Mon, 2010-08-02 at 22:39 -0600, Tavian Barnes wrote: > Imagine the story with a different daemon: SSH. You ssh into your > box, su, and pacman -Syu. Halfway through the upgrade, openssh gets > updated, which automatically restarts the server, which SIGHUPs > pacman, which is left in an inconsistent state. OpenSSH is designed to keep its connection alive during restarts. You can even stop sshd without losing connection, as long as the rc.d script is not buggy (we had a buggy script that would kill all child processes too a while ago...)
Re: [arch-general] Shouldn't pacman restart dovecot after update?
On Tue, 2010-08-03 at 01:00 -0500, David C. Rankin wrote: > Well sometimes the stupid ones among us don't always catch > that dovecot is > being updated when it is one of 73+ updates that take place. But we > 'do' always > catch the failure to copy to sent -- later :p > > Seriously, one KISS thing that really works for Arch is having > the little > 1-line notes that are printed during update like: The point is that people don't read, so adding an update notice is not a solution either, as people will miss that also.
Re: [arch-general] What broke ctrl+c ??
On Sat, 2010-07-17 at 00:55 -0500, David C. Rankin wrote: > Can anyone else try in gnome-terminal and see if ctrl+c is broken for > you? Just > type 'ping whatever' and then try to kill ping with crtl+c. I can't > and that's a > problem. > > I'm using using gnome-terminal 2.30.2-1 Check your hotkey assignments (Edit -> Keyboard shortcuts). I guess you set CTRL+C somewhere.
Re: [arch-general] Boost
On Fri, 2010-06-18 at 23:56 +0200, Alexander Duscheleit wrote: > But there seems to be a trend starting in this direction which worries > me a little. IIRC there is nothing holding back Xorg 1.8 except for > legacy nvidia drivers, and that for quite a while. If that is true, > it's even more depressing for me, since nobody waited for us poor ATI > R500 users when catalyst dropped support. :) Which is not true... The reason for dropping catalyst is that they only release drivers when Ubuntu does a release. if Ubuntu doesn't ship a version of xorg-server, no driver will be made. The reason for holding back xorg-server 1.8 in testing was because of bugs and crashes. Until a week or two ago when I did the latest changes to the Intel driver, I couldn't get OpenGL on intel stable for more than 30 seconds. Also, the kernel contais massive changes for AMD and Intel drivers, so we basically needed the new kernel that was also in testing.
Re: [arch-general] [arch-dev-public] dropping flashplugin x86_64
On Thu, 2010-06-17 at 00:53 -0700, Patrick Burroughs wrote: > Personally, if it wasn't for HTML5 I wouldn't be able to use YouTube. > My laptop is ancient and decrepit, and cannot handle Flash on Linux, > but the element works just fine, and loads as fast as I'm used > to Flash video loading. Maybe there's a bit of a slowdown versus Flash > if you have a cutting-edge system, but not everyone is in that > situation. I don't care much about performance, but what is irritating is that whenever some website loads anything flash-related, my CPU gets speedstepped to max frequency and my laptop fan prepares my laptop for a takeoff. When you're on battery, this can mean you'll lose half of battery runtime, just by having a browser window open that includes a flash banner. This isn't only a problem on x86_64, but on every non-windows platform. Maybe it extends to windows also, but I haven't tested that in years. Besides the performance problem, flash also makes browsers unstable. I'm very happy with the out-of-process plugins in Firefox 3.6.4 prereleases, it's just too bad that I don't use firefox for daily browsing.
Re: [arch-general] Xorg-server 1.8 - when?
On Wed, 2010-06-16 at 09:14 +0200, Lukáš Jirkovský wrote: > Hello, > I'd like to ask what are the plans for the xorg-server 1.8, especially > now when RC1 of xorg-server 1.9 is out. Please don't get me wrong, I'm > not demanding anything but I'm wondering what is holding back the move > to [extra]. > > I'm using new xorg-server on two different computers (notebook with > intel and desktop with nvidia's blob) for some time without any > problems. Though xorg-server 1.8 is not completely issue-free, I don't think 1.7 from extra isn't issue-free either. I've been running these packages for months now without much problems. Before we start moving xorg-server 1.8, I would like to know when we can move kernel26. The latest kernel has a lot of fixes for Intel and ATI/AMD DRM. Do we have really important open bugs for 2.6.34, or are we just waiting for the magic .1 release?
Re: [arch-general] udev
On Tue, 2010-06-08 at 15:49 +0900, Juan Diego Tascón wrote: > Actually it is installed, it is called gspca_zc3xx, if I load it the > webcam works, the thing is it is not being loaded automatically > > > All modules have device aliases for the devices it supports. Udev just > > tries to load a module by devicename and gives your error if it can't > > succeed. > > Should I report this to the udev developers? When I run "modinfo gspca_zc3xx", I get an alias for your device (usb:v0AC8p305Bd*dc*dsc*dp*ic*isc*ip* matches). This is with kernel 2.6.34 from testing. What kernel do you use? Older versions might not have the device alias assigned to the module.
Re: [arch-general] udev
On Tue, 2010-06-08 at 15:16 +0900, Juan Diego Tascón wrote: > Good day, > > I'm getting an error every time I plug-in my webcam as a result the > corresponding module is not being loaded automatically so I have to > load it by myself. Any one knows a way to fix this? > > This is the error: > > load-modules.sh: 'usb:v0AC8p305Bd0100dcFFdsc00dp00icFFiscFFipFF' is > not a valid module or alias name There's no kernel module for your webcam installed. You'll have to find out what driver it uses and udev will load the driver when you hotplug your webcam. All modules have device aliases for the devices it supports. Udev just tries to load a module by devicename and gives your error if it can't succeed.
Re: [arch-general] Burning From Command Line
On Wed, 2010-05-26 at 14:29 +0200, Joerg Schilling wrote: > Jan de Groot wrote: > > > > Sun legal department. Do you know of a single Linux distro that > > > dropped libcdio > > > because of the obvious licence violations in libcdio? > > > > Libcdio doesn't violate any license, but it's GPL, while Sun doesn't > > want GPL'ed libraries in Solaris. GPL for libraries is very restrictive. > > In fact, everything you link against libcdio will have all restrictions > > applied by the GPL license, even if that software is LGPL. > > You are obviously not correct, check Solaris. > > libcdio has two legal problems: > > 1)It claims to be under GPL but it is called from LGPL code. > Most people believe that this is not permitted. > > 2)libcdio is based on code that is available under > > - GPLv2 _only_ > > - CDDL > > The related code was never made available under a different > license. The "Autor" of libcdio first claimed that the code > is "GPLv2 or any later" now he claims it is GPLv3. He did > however never ask the real author of the related code for > permission to do this license change and he now as a result > of his violations would definitely not get this permission. > > > Please stop spreading this nonsense. > > It is you who spreads nonsense :-( > > Please stop this! 1) This is permitted, though it turns the complete package into GPL. This is also why libcdio has moved from gst-plugins-good to gst-plugins-ugly. Note that LGPL gives permission to change the license to ordinary GPL in section 3. 2) I found some bugreport on launchpad with that claim from you, but besides that, I can't find any information. The bugreport says you should take it up with the FSF, but somehow I can't find any reference about that. If linking GPL and CDDL code together isn't a problem for you and your lawyers, then I don't know why 1) would be a problem for you either. As for your claims, there's still an open question for you: http://mailman.archlinux.org/pipermail/arch-general/2010-February/011082.html
Re: [arch-general] Burning From Command Line
On Wed, 2010-05-26 at 13:35 +0200, Joerg Schilling wrote: > ust to make clear how picky license violations are handled with > Solaris, check > that "libcdio" has been removed from Solaris distributions in 2007 > because of > the license violations in libcdio that have been found by Sun lawyers > in the > Sun legal department. Do you know of a single Linux distro that > dropped libcdio > because of the obvious licence violations in libcdio? Libcdio doesn't violate any license, but it's GPL, while Sun doesn't want GPL'ed libraries in Solaris. GPL for libraries is very restrictive. In fact, everything you link against libcdio will have all restrictions applied by the GPL license, even if that software is LGPL. Please stop spreading this nonsense.
Re: [arch-general] new pc; keep arch installation?
On Tue, 2010-05-25 at 15:46 +0200, Vincent Schut wrote: > Hi, > > I'm going to get a new pc, replacing my now more than 5 years old pc. I > would like to keep my current arch installation, and hope to prevent the > need to do a full reinstall. I like the software I have, and I like the > configuration I have... > > Could I just put my old harddisk (which is relatively new and still OK, > I have SMART enabled) in the new pc and reboot? > > Both are amd 64 bit cpu's, the old one one of the first athlon 64's, the > new one a Phenom II X6 1035T (and my current arch linux install is 64bit) > > Of course stuff like chipset, network chip, video etc. are all > different. For video that would be no problem, I could install other > (current = nvidia, new pc = ati) drivers from the text console. However, > I'd like basic stuff and network to work :-) > > Could anyone give me an idea about the chance of luck for such an > operation? Tips, hints? Or would you just advise a clean install and > install and reconfigure all software again? Linux is not Windows, so you'll get away with it. You could put the harddisk in your new system and boot the new system with the fallback kernel option which includes an initramfs image with all drivers. You'll probably have to reconfigure X then. After booting the fallback kernel, it's advised to regenerate your initramfs images using mkinitcpio, or by just reinstalling kernel26. My Archlinux installation survived multiple mainboards and harddisks, it's years old by now.
Re: [arch-general] A question about Arch Sixty Four
On Mon, 2010-05-24 at 17:05 +0200, Nicky726 wrote: > I have experimentally found out, that 64 bit Linux distro uses like 40 -- 80 > % > more of RAM than 32 bit. Now it seemed to be both aplication and distro > dependant, with Arch being on the better side. Though I've got to say again, > it > was not a benchmark, just my personal experiment. > > As for me, if I had a machine with plenty RAM (that is from my perspective > ~GB), than I would choose 64 bit. If I had like 1 -- 2 GB, than I would > definitely go 32 bit. > > Hope I did understand your question right. Depending on the software you run, programs can take more memory due to larger integers and pointer sizes. If memory usage is a problem, go for a 64bit kernel with 32bit userland. Memory allocation above 896MB is much more efficient on 64bit kernels because you don't have to use the "highmem" method to access that.
Re: [arch-general] [arch-dev-public] kernel 2.6.34-1
On Wed, 2010-05-19 at 16:39 +0200, Thomas Bächler wrote: > Am 19.05.2010 16:19, schrieb Dan McGee: > > On Wed, May 19, 2010 at 6:10 AM, Thomas Bächler > > wrote: > >> Am 19.05.2010 10:56, schrieb Evangelos Foutras: > >>> Also did a diff [1] between the file lists of kernel26-firmware-2.6.34-1 > >>> and linux-firmware-git-20100519-1. It shows that ralink firmware has > >>> indeed been added to the linux-firmware repository, which should resolve > >>> FS#19519 [2]. > >> > >> I did that too and noticed the same. It would also replace all intel > >> ucode packages, and some more I don't know about. However, it is > >> considerably bigger than kernel26-firmware (2MB vs. 12MB). > > > > How often does it need to be updated? We now (needlessly?) have the > > firmware package track the actual kernel package so it ends up getting > > re-downloaded a lot more often than probably necessary, so the above > > package, while large, would probably be updated less anyway (and would > > be an 'any' package?). > > Everytime someone complains about a firmware file missing. > > Seriously, look at the commit log, it only had 23 commits this year: > http://git.kernel.org/?p=linux/kernel/git/dwmw2/linux-firmware.git;a=summary > > I guess we would upgrade it at most once a month. +1 for replacing all firmware packages included in this one. I don't care about the few extra megabytes on my system. This saves us uploading and downloading a binary firmware package on every kernel update, generates less packages in the repositories and moves kernel26-firmware to an arch=any package. The only disadvantage is the 10MB extra size, but the advantages are bigger than that.
Re: [arch-general] BASH no longer does 'for i in $(ls); do ls $i; done'??
On Wed, 2010-05-19 at 01:41 -0500, David C. Rankin wrote: > So the bottom line is don't alias 'ls'. And, I guess, the next > question would > be, how or where can I safely customize the behavior of ls without > screwing > myself again. Solution is simple: don't use ls for the for loops you're doing. I just tested it on my system, the colored files and directories give "No such file or directory", which makes sense, as the variable contains ascii chars for color information. I would suggest using globbing, why would you use for i in $(ls); while you can use for i in *;
Re: [arch-general] Fw: NDN: Question regarding the ArchLinux iso core image
On Tue, 2010-05-18 at 05:04 -0400, Keith Hinton wrote: > Hey guys, I wanted to forward this message. > Has anyone ever gotten this? I get this consistently when posting messages to > Arch-General, from the following email adddress. > Hope someone can tel me why this happens-or talk to whomever owns > Arch-general, and update it. > Is this a problem with the mailman configuration on Arch Linux's server that > runs Arch-general? Thanks. > Obviously my post was delivered but I don't know why this happens. > Note: this doesn't happen with Arch-Releng. > Sorry for posting this hear, but there you are. > > > From: Post Office > Sent: Tuesday, May 18, 2010 3:17 AM > To: Keith Hinton > Subject: NDN: [arch-general] Question regarding the ArchLinux iso core image > > > Sorry. Your message could not be delivered to: > > > Joey Kingery (Mailbox has been deleted. Try re-entering the address.) That's because you have the reply-to header set. Mailman replaces this field with arch-general@archlinux.org, but if you already have it set in the mails you send, it just appends the address. When the mail bounces, you'll get the bounce too.
Re: [arch-general] Question regarding the ArchLinux iso core image
On Tue, 2010-05-18 at 03:17 -0400, Keith Hinton wrote: > Hi. > I've recently seen threads on this list pertaining to interesting issues with > the Arch Linux core isos. > I haven't had that problem in my primary tests of those Isos in particular. > Is there any reason why a successful installation may occur in one > environment over another? Or is this a common Linux problem. > Thanks for answering my question. > I hope that the rest of you folks have successful Core installations, however > a net-install is the only real way to isntall Arch Linux. > I do not see the point of the core installer media, personally. Why would one > wish to use that? A snapshot will become outdated in a flash, and all of you > archers know this. As soon as a newer package version is available the entire > snapshot is outdated instantly. Or will be, given a few > weeks/months/years/whatever. > I therefore have always installed from what I believe to be the Arch-Way, > wich is installing via the Internet alone to have an updated system. > Does anyone agree/disagree with my idea on the proper way to install Arch? :D > Regards, --Keith My prefered way of installing any linux distribution is using bootstrap tools. For Debian this is debootstrap, for Archlinux this is pacman -r. The point of having an installer with packages on the ISO is that you don't always have the possibility to do a network installation.
Re: [arch-general] testing xorg packages may not be nvidia friendly
On Thu, 2010-05-13 at 18:37 -0400, Caleb Cushing wrote: > I tried installing testing on a box running nvidia drivers that I have > at school. X didn't come up. I don't know why and didn't really > investigate (as I was really testing to see if my box at home had an > env issue seems it does). using nvidia-173xx-utils For the legacy drivers you need to make an xorg.conf that sets "IgnoreABI" in the serverflags. Nvidia hasn't released official drivers for xorg-server 1.8.
Re: [arch-general] Brasero Wont Copy Disks
On Mon, 2010-05-03 at 16:11 -0400, Carlos Mennens wrote: > I just installed Brasero to burn / copy disks. I first added my user > 'carlos' to the 'optical' group. However when I try and copy a disk on > my machine, I get an immediate error that says it fails. Then when I > save the log, it's blank. It has no details as to why the copy failed. > Anyone know what I can do to resolve this and allow my machine to burn > / copy disks? I know my optical drive is running fine because I booted > from Ubuntu Live CD and was able to copy a disk fine. > > Thanks for any help! You don't need to add yourself to the optical group. You need to have polkit permissions for this. I don't know if brasero wants to elevate to root permissions or if local console login registered with consolekit is enough, but your error looks like you don't have a polkit authorization agent installed. Try to install polkit-gnome and restart brasero, it should come up with an authorization dialog.
Re: [arch-general] Firefox graphics licensed under MPL
On Sun, 2010-05-02 at 13:28 -0400, Denis Kobozev wrote: > Why not do like we're told then, and send them an email? "Arch has > such and such package, with such and such modifications. Firefox logo > is now under MPL and it looks like we can add it to our package. Can > we?". We did that before, and we got permission to ship with branding. The issue here is that whenever we update the package or add/remove a patch, we have to ask for permission again. This doesn't look a big problem for you, but it means that we have to stick to a schedule to get releases to the repository. Right now we package firefox releases as soon as it lands on upstream FTP. With the approval policy, it will take days or even weeks before we can package a new firefox version.
Re: [arch-general] Latest NVIDIA en kernel update
On Fri, 2010-04-23 at 15:35 +0200, Patrick Baumgart wrote: > Another one: I also have in mind that glxgears, started without > parameters on default size, gets me around 45000 fps and now it has > abouot 15000 fps. glxgears is NOT A BENCHMARK. This simple demo just confirms that 3D works, nothing more, nothing less. Operations done in glxgears don't represent day to day usage. Compositing window managers do completely different things than displaying spinning gears.
Re: [arch-general] gstreamer0.10-good-plugins need a whole load of GNOME stuff?
On Thu, 2010-04-22 at 06:22 +0800, Ng Oon-Ee wrote: > > > Would you prefer the developer reimplement security-authorization and a > configuration parser, then? Its not even as if gconf and its editor > aren't separated into different packages. As long as GNOME doesn't start using it, that would be fine. I mean, who wants to use a generic network library that appears to support proxies and authentication storage? I think it's awesome to implement this in each and every application, just because we hate storing passwords in a central location.
Re: [arch-general] HAL dependencies
On Sun, 2010-04-18 at 01:33 +0200, Thomas Bächler wrote: > Am 18.04.2010 01:22, schrieb Rob Bean: > > It looks like there are only 3: gnome-vfs, > gstreamer0.10-good-plugins, and > > vlc. I can use ABS to rebuild these without HAL, but I'm concerned > that > > these packages may actually need it. > > From short discussions with Jan, I determined that the HAL dependency > can probably be safely removed from all of those. gstreamer0.10-good-plugins has been fixed in SVN. I might actually add pulseaudio support to that soon though, so a final rebuild can take time. vlc, no idea about that, but I guess it can do without automatic media detection. gnome-vfs will keep the hal dependency. It's a deprecated library for deprecated applications. Losing removable media support for deprecated applications is not an option to me. The GNOME goal is to get rid of gnome-vfs, so after a while this library will be gone from your system completely. Note that though you still have hal installed on your system, it's not needed to start it if you don't (want to) use it. I stopped hal on all my systems as I don't use it anymore.
Re: [arch-general] gstreamer0.10-good-plugins need a whole load of GNOME stuff?
On Wed, 2010-04-21 at 10:35 +0100, Nathan Wayde wrote: > Do you mind listing these Gnome dependencies, or at least your > proposed > PKGBUILD? > I've looked at this and I can see only 2 (gconf and libsoup-gnome). > Ofcourse that is based on the assumption that the pacma dependencies > are > correct(e.g good-plugins depends on x,y,z but only x is listed > because > it also depends only y and z). The GNOME dependencies involved: libsoup-gnome libgnome-keyring (gnome-keyring) gconf orbit2 libidl2 I don't consider this as a big problem, and I closed bugs about it before. The libsoup-gnome library adds proxy lookup and password lookup functionality to libsoup. GStreamer can work without it, but it will lose quite some functionality without it. The gconf*sink elements in GStreamer need gconf. Without this, there's no way to configure the audio and video element in a central place. Some applications even ask for this specifically. I listed the gnome-keyring dependency as optional, as it's a fake dependency added to libgnome-keyring. libgnome-keyring needs a keyring daemon, AFAIK the next version of KDE will also get one. In the future, GNOME applications using libgnome-keyring can read passwords from KDE password storage. Libidl2 isn't a big issue either, as firefox already depends on it. I haven't seen much Arch Linux desktop installations without firefox. Note that the hal plugin will get removed soon. The only application I know that could have benefit from this plugin is cheese, but that has been ported to libgudev since 2.30.0.
Re: [arch-general] 4k sector drives
On Wed, 2010-04-21 at 10:31 +0200, Uli Armbruster wrote: > I can second that, I don't have Arch installed on a 4k drive but I > have an external drive, which I partitioned with fdisk -c -u /dev/sdX > and there's no speed drop at all, so you can use this command line > without any worries Some of these 4K drives export their sectors to the OS as 512 byte sectors. If you happen to align your partition incorrectly then, every inode read/write will result in 2 sectors being touched. For drives that export their sectors as 4K units to the OS, there's not a single issue. Just be aware of this fact, I don't know how fdisk aligns stuff these days, but for my SSD I chose to use parted.
Re: [arch-general] 4k sector drives
On Tue, 2010-04-20 at 20:52 -0400, Caleb Cushing wrote: > has anyone installed arch on one of these drives? thinking about > buying a pair and running them mirrored... but not sure if everything > will be working optimal. note: since I'll have to migrate off an > existing drive I can use the tools installed on that to partition it > first. Is there anything I should know first? You will have to align partitions on a 4K boundary, performance will be miserable without that. I think parted is the best tool to do partitioning in this case, as parted supports aligning.
Re: [arch-general] [arch-dev-public] xorg-server 1.8 repository
On Sat, 2010-04-17 at 23:41 -0400, Ray Kohler wrote: > On Sun, Apr 4, 2010 at 4:56 PM, Jan de Groot wrote: > > Things todo before this merges to extra: > > - Fix xf86-video-siliconmotion, xf86-video-unichrome, xf86-video-nouveau > > I'm curious as to what this means. In particular, what's broken with > nouveau? Does it not compile, or not work properly? And is this with > the source snapshot in [extra], or some newer version? With nouveau, every update to libdrm usually requires changes in nouveau. I haven't tried the current version of nouveau with libdrm 2.4.20.
Re: [arch-general] xorg-server 1.8 - Wiki?
On Fri, 2010-04-09 at 16:01 -0400, pyther wrote: > Alright I started editing the wiki. The majority of the changes were > made under the configuration section. I pretty much copied verbatim > the > InputClasses section. If someone can look over them and possibly > improve > the wiki I would appreciate it. > > Do we need 'Changes with modular Xorg' anymore? > Do we need Automatic Configuration with hwd? From what I can gather > hwd > won't generate a working configuration with Xorg 1.8 In case there's anything related to hwd in there: kill it. We don't support hwd as X.org configuration tool. The best autoconfiguration tool you can get is Xorg -configure, and these days even that is not needed anymore.
Re: [arch-general] xorg-server 1.8 - Wiki?
On Fri, 2010-04-09 at 09:34 -0400, pyther wrote: > > Has there been any work on a wiki entry for Xorg 1.8? I don't see > anything on the main wiki. I just don't want to duplicate efforts. > > Cheers! Nothing has been done on documentation yet. Feel free to use the documentation from Fedora to add information to the Xorg input hotplugging wiki page.
Re: [arch-general] testing xorg-server 1.8 [xorg18]
On Tue, 2010-04-06 at 01:30 +0200, Marek Otahal wrote: > ...well, here's an update after a deeper tryout :) > *touchpad(synaptics) - issues with tapping which does not > work /buttons do/ For the previous synaptics package, we used to include default configuration that enables tapping. With the release in the xorg18 repo the only configuration done is driver assignment. I haven't noticed this problem as GNOME overrides synaptics settings anyways.
Re: [arch-general] tmpfs
On Fri, 2010-03-26 at 00:53 +0900, Juan Diego wrote: > Good day everyone, > > I want to delete the next two lines from my fstab: > > none /dev/pts devpts defaults 0 0 > none /dev/shm tmpfs defaults 0 0 > > however I would like to know the secondary effects of doing such > thing, all the information I could get on the internet is that shm is > for POSIX shared memory support and that almost any program actually > use it, for devpts I couldn't find much information about it. > > does anyone knows if it is safe to remove those lines from my fstab? > > thank you. /dev/pts is required for virtual terminals like xterm, gnome-terminal and others. Without that filesystem you can't start a virtual terminal. /dev/shm is used for shared memory, it's a requirement for NPTL semaphore functions. Now that /dev is also on tmpfs, I don't know if it's harmful to remove. The main requirement is that this folder has 1777 permissions.
Re: [arch-general] MySQL Starts But Fails to Restart
When you stop MySQL, the kill command completes immediately, while MySQL is still doing shutdown tasks to make sure the database is in a consistent state. As MySQL is still running when you try to start it, it will fail to do so. There's 2 ways to fix this: - use mysqladmin shutdown, needs a user in the database with process privileges, this is the way used by debian - implement a loop that waits for MySQL to shutdown after executing the kill command. This is what the upstream MySQL start/stop script does. On Tue, 2010-03-23 at 22:46 -0400, Carlos Mennens wrote: > 4. I then run the command to restart the daemon for MySQL and it fails... > > Code: > > [r...@ghost /]# /etc/rc.d/mysqld restart > :: Stopping MySQL Server > > [DONE] > :: Starting MySQL Server > > [FAIL] > > I can use the 'start' and 'stop' commands for /etc/rc.d/mysqld but > 'restart' for some reason fails and this concerns me. Can anyone help > me understand what is wrong?
Re: [arch-general] HAL depreciation
On Wed, 2010-03-17 at 03:06 +0100, Damjan Georgievski wrote: > Does anyone know how much of HAL is needed in ArchLinux these days? > I'm asking because I've learned that both udev and HAL configure the > keymap of input devices nowdays and > I wonder what other former HAL features are already implemented in udev. > At this moment several applications, including XFCE and KDE, use hal for removable device handling. xorg-server still uses hal to configure input devices. Starting from xorg-server 1.8, we'll disable the hal backend and switch to udev, devices will get configured through udev rules in that case. This will cause some breakage of existing setups, but there's no way to enable both backends. KDE should support udisks/upower in the next major release. I don't know about XFCE though, but if that one follows, I'm almost sure hal will no longer exist in 2011.
Re: [arch-general] [patch] AIF, partition table compatibility with grub2.
On Mon, 2010-02-22 at 23:21 +0100, Tobias Powalowski wrote: > Just poked around with grub2 and kvm, > you really need the extra space, else it will fail to install. > creating partitions with cfdisk will also leave this space. With nowadays RAID and SSD stuff you want to align your partitions on stripe or blocksizes anyways. Some newer harddisks come with 4Kbyte sectors and have a mapping to 512byte sectors so old OS'es can work with it. You really want to align such devices, as performance is terrible without it.
Re: [arch-general] policykit vs polkit
On Wed, 2010-02-10 at 00:04 +0100, Jan de Groot wrote: > policykit and policykit-gnome will die soon. At this moment only > pulseaudio in community uses it, but I'm sure someone has ported that > to polkit also. I just checked upstream and Fedora: either the pulseaudio package in community is outdated, or it contains a bullshit dependency. Later versions of pulseaudio don't call into policykit for realtime permissions anymore, but rely on being started with realtime permissions, or, if it's available, requesting realtime capabilities through realtimekit.
Re: [arch-general] policykit vs polkit
On Tue, 2010-02-09 at 19:17 +0100, Thomas Bächler wrote: > Am 09.02.2010 19:11, schrieb Lukáš Jirkovský: > > Hi, > > I just found out that policykit on my package is no longer needed. > > Because it looked strange to me a took a look at other packages and > > found polkit package. What is the difference between these two > > (policykit and polkit)? Both seems to provide the same package. > > policykit is dead and being replace by polkit. I think now that KDE 4.4 > uses polkit, policykit might in fact be obsolete. > policykit and policykit-gnome will die soon. At this moment only pulseaudio in community uses it, but I'm sure someone has ported that to polkit also.
Re: [arch-general] KMS again with 2.6.33?
On Tue, 2010-02-09 at 01:20 +0100, Xavier Chantry wrote: > That's true, but the conflict goes both way. And the situation isn't > different with ati vs fglrx, is it ? > > Anyway, there is no reason to enable nouveau in the kernel, the > current out-of-tree build is just fine and allows for more frequent > updates. > As long as users still have to install it explicitly, they should be > able to handle this conflict. > But it would indeed be more problematic if we wanted to have it > directly in the kernel package. The point is that it moved to the staging tree now, where further development will take place. After a while, there won't be an out-of-tree version anymore because all development happens in the official kernel git tree. An example of this is intel: there's several kernel trees around with intel drivers, but no possible way to build intel drm out of tree.
Re: [arch-general] KMS again with 2.6.33?
On Mon, 2010-02-08 at 18:52 +0100, Xavier Chantry wrote: > It seems the question was not only about kms but also about dri2, > which is needed to get 3d support with kms. > > About nouveau, ums was dropped from ddx on 11th January and > extra/xf86-video-nouveau package is from 17th January. So the only > option is to use kms, which should also be enabled by default now. Which means autoloading of nouveau modules by udev, which means breaking the nvidia drivers for everyone not blacklisting it. This is going to be so much fun... especially now nouveau was merged in the staging tree.
Re: [arch-general] A suggestion for the devs regarding rebuilds
On Mon, 2010-02-08 at 13:37 -0700, Brendan Long wrote:
> Couldn't the piecemeal update problem be fixed by just putting
> version
> numbers in the depends() section in each updated package, so for the
> libpng rebuild for example, depends(... libpng>=#.#)? It would fix
> the
> problem in the most obvious/arch-like way.
Some of us do that, so what happens next:
very-old-system# pacman -Sy firefox
pacman downloads new repo db, checks for firefox and downloads all the
versioned dependencies. The user can run firefox fine now, but
everything else on the system is broken because it wasn't updated for
the new libpng.
Now this can be caught by using depends=('libpng>=1.4.0'
'libpng<1.5.0'), but what if upstream decides to fuck up the ABI in a
release and bump soname in the middle of the 1.4 branch? I don't expect
libpng to do that, but there's examples of others around. We can never
be sure that the next incompatible version will be 1.5.0.
Re: [arch-general] Memory Management problems with 2.6.32
On Sun, 2010-02-07 at 02:09 +0200, Ionut Biru wrote: > On 02/07/2010 02:00 AM, Gaurish Sharma wrote: > > Hi, > > I am using ArchlInux 64bit 2.6.32.7-1. but having some memory allocation > > errors, > > > > it might be because of oss. > > http://bugs.archlinux.org/task/18014 I'm thinking more about this one: http://git.kernel.org/?p=linux/kernel/git/stable/stable-queue.git;a=blob;f=queue-2.6.32/drm-i915-selectively-enable-self-reclaim.patch;h=94f3bb2a29415a68d74173319a3937764e135410;hb=c2846447fc922df45ad45a85e57f5519afdeedb3
Re: [arch-general] Libraries should also be built statically
On Mon, 2010-02-01 at 14:59 +0100, Heiko Baums wrote: > Am Mon, 01 Feb 2010 22:58:24 +1000 > schrieb Allan McRae : > > > I disagree. Static libraries generally suck and hide rebuilds needed > > for security issues. So unless something specifically needs the > > static library, I think it should be removed. > > Such rebuilds are only hidden if a program is linked statically against > a library even if it can be linked dynamically, which should be avoided > anyway. If a program is linked dynamically it doesn't touch the static > library. If a program is built static against an insecure library, upgrading the insecure library means the static binary is still vulnerable. That's what Allan means. > So if a dev, TU or AUR maintainer builds a package which depends on a > library. This package should usually be linked dynamically against this > library. In this case the static library isn't needed. But if such a > package like fbsplash needs to be statically linked against a library > it's only possible if the static version is available. When we switch to glibc-based initramfs, there shouldn't be any need for static compiled binaries anymore, ever. > And other people who probably just want to write software for their > own, who want to build a portable app or to learn programming or > whatever and want or need to use statical linking can't do this without > the static libraries. Static binaries are everything but portable. Static linking to glibc can freak out when you're using nss_* libraries for uid and gid resolving. > And do you want everybody who needs the static version of a library for > whatever reason to file a bug report for every single case and to > explain why he needs this static version? No, they shouldn't need those libraries. > That would take too much time while building a static library doesn't > take much more time and disk space. Static libraries are bad. Besides taking up diskspace, they're just bad to use. Ulrich Drepper has a nice PDF about this.
Re: [arch-general] CDROM cannot mount
On Mon, 2010-02-01 at 13:20 +0200, Thanos Zygouris wrote: > Output of lsmod | grep cdrom > cdrom 36354 1 sr_mod > > As for the controller, i am not sure...my mothermoard uses the nVidia 430 > chipset...is that enough info? > Could you just attach the output of dmesg?
Re: [arch-general] CDROM cannot mount
On Mon, 2010-02-01 at 12:51 +0200, Thanos Zygouris wrote: > Thanks for replying... > It's a IDE drive: > 1) wodim --devices > 0 dev='/dev/scd0'rwrw-- : 'TSSTcorp' 'CD/DVDW SH-S182D' > > 2) dmesg | grep sr0 > sr0: scsi3-mmc drive: 48x/48x writer dvd-ram cd/rw xa/form2 cdda tray > > And i did try to disable hal...nothing changes... What IDE controller is it connected to and what driver is that using?
Re: [arch-general] CDROM cannot mount
On Mon, 2010-02-01 at 12:26 +0200, Thanos Zygouris wrote: > Hi gyus, > i got a problem...(do not know how/when did it started though). > I cannot mount CD roms...(i do not speak about automounting) > Here is what "dmesg" tells: > > sr 5:0:0:0: [sr0] Unhandled sense code > sr 5:0:0:0: [sr0] Result: hostbyte=0x00 driverbyte=0x08 > sr 5:0:0:0: [sr0] Sense Key : 0x4 [current] > sr 5:0:0:0: [sr0] ASC=0x3e ASCQ=0x2 > sr 5:0:0:0: [sr0] CDB: cdb[0]=0x28: 28 00 00 00 e6 2a 00 00 02 00 > end_request: I/O error, dev sr0, sector 235688 > Buffer I/O error on device sr0, logical block 29461 > > I got more or less the same problem with almost all my data cds (i do not > use audio ones) > Also, i cannot even mount them manually (via mount /dev/cdrom0 or > /dev/sr0/ ) > But one or two, mounts normally. (I use pcmanfm as a file manager, and yes i > have ck-launch-session in my .xinitrc) > The strange thing is that all of them can be mounted under Windows XP > SP3...so it's not a hardware problem. > Looked through Archlinux forums, but found nothing that i could use... > Any clues? > > P.S.1 : Most of the CDs are bootable ones, writen with "cdrecord -v speed=8 > dev=/dev/scd0 -data -tao -multi". They boot normally though. > > -- Beware of he who would deny you access to information, for in his heart > he dreams himself your master. First, try stopping hal, this could be a polling issue that freaks out your drive. If that doesn't work out, please give more details about your system: - SATA? IDE? - What chipset - What recorder
Re: [arch-general] core/linux-api-headers?
On Sun, 2010-01-31 at 23:45 +0100, f...@kokkinizita.net wrote: > On Sun, Jan 31, 2010 at 11:26:02PM +0100, Giovanni Scafora wrote: > > 2010/1/31, Hussam Al-Tayeb : > > > Or simply tell him the the package kernel-headers was renamed to > > > linux-api-headers? > > > > Nope, pacman already said him that. :-) > > So, if pacman ever asks: > > Replace cdrkit by cdrtools ? [Yn] > > that means that cdrkit has been renamed to cdrtools ? :-) No, that means that either someone uploaded a rogue package to community, or that you're using a repository that contains a cdrtools package that replaces cdrkit. Nothing more, nothing less.
Re: [arch-general] kernel-lts ....
On Sun, 2010-01-31 at 10:34 -0600, William A. Mahaffey III wrote: > > I am new to the list, used Linux since Caldera 2.2. I noticed > references to a kernel-lts, labelled as 'long-time-support' on the Arch > website. I did a bit of googling & noticed references to such from Arch > & Ubuntu forums. I found no references at kernel.org, although I noticed > that they listed an upgrade to kernel 2.6.27.45, same number as the > '-lts' kernel in Arch. Is the 'lts' kernel project Arch/Ubuntu-specific, > or (semi-?) supported from kernel.org ? Also, I noticed that the > bleeding edge kernels (seem to) include firmware packages, but I didn't > see them for the lts kernels, did I just miss them or are they absent > from the 'lts' series of kernels ? TIA The -lts kernels are stable kernels as maintained by kernel.org people. The 2.6.27 kernel has been maintained as stable kernel for 2-3 years now. When we added the kernel, our intentions were to have a maintained stable kernel that doesn't bring surprises after a security update, something that going from 2.6.31 -> 2.6.32 won't offer you. The firmware binaries were included in drivers before, they've been split to standalone files in later versions of the kernel. That's why 2.6.27 doesn't come with a -firmware package.
Re: [arch-general] Firefox 3.5.7 not support eh !??
On Sun, 2010-01-31 at 16:58 +0530, Nilesh Govindarajan wrote: > Something is seriously funny going on here with FF 3.5.7 (Shiretoko). > > I'm using it from the arch repos, but on visiting Google Help or Orkut, > it says Browser not supported ? > > And the supported browser list says FF 1.5+ > > Something is wrong with the arch build ? There's nothing wrong with our firefox package, it's buggy browser detection done by these websites. You might want to change the user agent string through about:config to get Firefox in the name. Websites with browser detection like this are plain stupid, you should complain about it and advise them to check for a gecko engine version.
Re: [arch-general] ca-cf keyboard layout in hal
On Sat, 2010-01-30 at 08:56 -0500, Sébastien Leblanc wrote: > I cannot find the "cf" variant in the ca layout (canadian french) in > /usr/share/X11/xkb/rules/xorg.lst. What happened with it? I am trying > to use a USB keyboard on my laptop, however it gets set as US instead > of ca-cf. I check the file 10-keymap.fdi in /etc/hal/fdi/policy, but > when I put ca as layout and cf as variant, my keyboard does not > respond when I plug it in. Any other layout (that exists) works. > > Did Xorg remove the ca-cf layout? The layouts are listed in evdev.lst, as that's what the evdev driver uses. Looking through that file, the variant for canadian french is fr-legacy or fr-dvorak.
Re: [arch-general] An old, tiresome discussion: cdrtools vs cdrkit
On Fri, 2010-01-29 at 14:35 +0100, Joerg Schilling wrote: > I am not sure whether this is the best solution. I recommend to use > star as star > is the oldest free tar implementation and as it supports ACLs since 10 > years > already. Adding more meta data is relatively simple. Implementing it in star has no use, as our package manager doesn't use star but libarchive, the library that bsdtar is based on.
Re: [arch-general] An old, tiresome discussion: cdrtools vs cdrkit
On Wed, 2010-01-27 at 16:39 +0100, Heiko Baums wrote: > 2. Jörg as the author and seeming copyright holder of both software > cdrecord and mkisofs has already stated that he has no problem with > the release of cdrtools (linking cdrecord against mkisofs) and won't > sue you. If he would be the full copyright holder of mkisofs, he would have re-licensed it to CDDL also, solving the whole problem.
Re: [arch-general] KDE 4.3 no power functions ?
On Wed, 2010-01-27 at 16:20 +0100, Alberto Bonacina wrote: > 2010/1/27 Jan de Groot : > > 1) Power group? Damn, that's old. Does KDE still use that? > > I don't know if it's old but the beginners guide [1] tolds that when > you want to create a new users you should "put" him in the power group > to e.g.: shutdown with power button. > > [1] > http://wiki.archlinux.org/index.php/Beginners_Guide#Step_4:_Add_a_user_and_setup_groups With consolekit and the fixed ACL management in the latest package of udev none of that should be required anymore.
Re: [arch-general] KDE 4.3 no power functions ?
On Wed, 2010-01-27 at 16:10 +0100, Alberto Bonacina wrote: > 2010/1/27 Nilesh Govindarajan : > > I recently installed KDE 4.3.4 > > > > The Leave section in the menu doesn't have power functions like shutdown and > > restart. > > > > How to solve this ? > > There are two answer: > > 1) Are your user in the power group? > 2) You start KDE with a login manager like GDM/KDM/slim or with > .xinitrc and startx. > > > Also GTK apps are appearing very ugly, any suggestions ? > > You could install qtcurve-gtk2 qtcurve-kde4 and manage them with > gtk-theme-switch2 1) Power group? Damn, that's old. Does KDE still use that? 2) Using a display manager that supports ConsoleKit should fix this issue. KDM/GDM are two display managers that have native support for ConsoleKit, others need pam modules and/or ck-launch-session.
