- Arch-general - lists.archlinux.org

[arch-general] arch with sata drives
by Jude DaShiell 03 Mar '12

03 Mar '12

So far as I can tell, it works fine and arch has no problem with amd K8 machines either. I have Tim Brennan's talking arch installed on a 1.5TB sata disk and that was something debian couldn't do. Slackware also had no problem with the sata drive either. ---------------------------------------------------------------- Jude <jdashiel-at-shellworld-dot-net> <http://www.shellworld.net/~jdashiel/nj.html>

1 0

[arch-general] Gringotts, alternatives?
by Magnus Therning 02 Mar '12

02 Mar '12

Gringotts has stopped working with a not completely uncommon error: ***MEMORY-ERROR***: [10817]: GSlice: failed to allocate 1008 bytes (alignment: 1024): Cannot allocate memory Given the time since the last release I'd be surprised if it's fixed any time soon. Also, I've never been extremely impressed with it so maybe this is a good time to look for an alternative. This means I now have two problems to solve, and any help on either would be appreciated: 1. How do I get the data out of the gringotts vault? Is there any other application that can read it and help me pull everything out? (I suppose I can always move the vault over to a Ubuntu box and extract it there, but I'd rather avoid that if possible ;-) 2. What alternatives to gringotts exist? What do other ArchLinux users keep their secret notes save? /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus(a)therning.org jabber: magnus(a)therning.org twitter: magthe http://therning.org/magnus

7 8

[arch-general] mkinitcpio
by Myra Nelson 02 Mar '12

02 Mar '12

I don't think this is significant enough to warrant a bug. Since the linux-3.2.6-1 kernel updates ( I don't remeber which version of mkinitcpio, sorry), I've been getting the following warning when mkinitcpio runs during the kernel install: cp: warning: source file `/lib/modules/3.2.9-1-ARCH/kernel/fs/ext4/ext4.ko.gz' specified more than once I originally thought it was due to my inclusion of the mkinitcpio-numlock hook from the AUR, but I have a fresh install on a new machine, without the numlock hook and get the same warning. I searched the bug tracker and found no reference to this. /usr is not mounted on a separte partition. my box HOOKS="base udev autodetect pata scsi sata filesystems usb usbinput fsck numlock" new box HOOKS="base udev autodetect pata scsi sata filesystems usb usbinput fsck" Since it's a warning not an error and doesn't appear to have any deliterious effects, it's not a problem. I just wondered if it's something I'm doing or if it's something in the hooks in mkinitcpio. Ciao for now Myra -- Life's fun when your sick and psychotic!

5 22

[arch-general] renaming files
by pete 02 Mar '12

02 Mar '12

Hi folks Bit sort of off topic but hopefully ok I have some 350 picture files with names along the lines of "IMG_7127 EOS-1D Mark III copy.jpg" i would like to rename them all to more like "IMG_7127.jpg" i have tried a few times tonight and cant get my head around it anyone got a script that can do it Thanks Pete . -- Linux 7-of-9 3.2.8-1-ARCH #1 SMP PREEMPT Mon Feb 27 21:51:46 CET 2012 x86_64 AMD Phenom(tm) 9600B Quad-Core Processor AuthenticAMD GNU/Linux

11 15

[arch-general] permission problem /var/lib/ntp after update (should it be ntp:ntp ?)
by David C. Rankin 01 Mar '12

01 Mar '12

Guys, After update of ntp to 4.2.6.p5-4, I get the following log entries: Mar 1 09:55:50 archangel ntpd[1180]: frequency file /var/lib/ntp/ntp.drift.TEMP: Permission denied Checking password and group, there is a ntp for each. Should the UIG:GID for /var/lib/ntp be ntp:ntp? -- David C. Rankin, J.D.,P.E.

3 3

[arch-general] Howto look outside chroot for local git tree?
by David C. Rankin 01 Mar '12

01 Mar '12

Guys, Is it possible to look outside of the chroot while building in an archroot from the PKGBUILD to check for a local copy of a git tree? Logically, what I want to do from inside the archroot is: (1) check for an existing git source in the ${srcdir} (if so, then use it) (if not, then...) (2) check for a local copy of the git tree outside chroot (if found) (3) change to that directory and update and copy to ${srcdir} (else) (4) clone new copy into ${srcdir} Is this even possible? I have made an attempt at implementing this, but I think I'm prevented from looking outside the chroot. Everything works except for (2) above. The check for the local copy always fails forcing a new clone into ${srcdir} instead of simply being able to update the local tree and copy to ${srcdir}. ## GIT repository variables and location _gitlocal="/home/david/tde/tde/main" _gitname="${pkgname#*-}" _gitdir="${_gitlocal}/dependencies/${_gitname}" _giturl="$url" _gitsfx="-git" _prefix="/opt/tde" build() { cd ${srcdir} <snip> ## update or clone from GIT msg "Checking for existing GIT source: $_gitname" if [ -d $_gitname ] ; then msg "Using existing source in: ${srcdir}/${_gitname}" else msg "Looking for local GIT tree: $_gitdir" if [ -d $_gitdir ] ; then msg "Updating local tree: $_gitdir" cd $_gitdir && { git pull git submodule init git submodule update --recursive git submodule foreach --recursive "git checkout master" git submodule foreach --recursive "git pull" } cd ${srcdir} && cp -a "$_gitdir" . msg "The local files are updated." else msg "Connecting to GIT server...." msg "Cloning new source from: $url" git clone --depth 1 $_giturl fi msg "GIT checkout done or server timeout" fi msg "Starting configure..." # ./configure \ msg "Building - ${pkgname}..." # make } The output from makechrootpkg is always the same even though the local copy of the git tree is present: ==> Making package: tde-libart-lgpl 1.3.2.4-1 (Wed Feb 29 23:55:20 UTC 2012) ==> Checking runtime dependencies... ==> Checking buildtime dependencies... ==> Retrieving Sources... ==> Extracting Sources... ==> Starting build()... ==> Checking for existing GIT source: libart-lgpl ==> Looking for local GIT tree: /home/david/tde/tde/main/dependencies/libart-lgpl ==> Connecting to GIT server.... ==> Cloning new source from: http://scm.trinitydesktop.org/scm/git/libart-lgpl Cloning into 'libart-lgpl'... ==> GIT checkout done or server timeout ==> Starting configure... ==> Building - tde-libart-lgpl... ==> Entering fakeroot environment... ==> Starting package()... ==> Packaging - tde-libart-lgpl-1.3.2.4 Possible? Or would the solution be to set the local copy of the git tree up inside a apache (or whatever it takes) and pretend to run a local copy? Even then, there would be no way to check for its existence from inside the chroot? What say the experts? -- David C. Rankin, J.D.,P.E.

4 4

[arch-general] How to update all the VCS packages at the same time?
by 郑文辉(Techlive Zheng) 01 Mar '12

01 Mar '12

I have installed several VCS packages from AUR which the PKGBUILD has not been updated for a while, is there a way to force update them all together? By the way, I use yaourt, but I don't know if there is a option can do that. Thanks

1 1

[arch-general] Howto prevent LDFLAGS --sort-common, --as-needed, <snip>, --hash-style=gnu
by David C. Rankin 29 Feb '12

29 Feb '12

Guys, I'm working on building kwin-style-crystal and I have patched acinclude.m4, aclocal.m4 and configure to remove references to --as-needed, but when I run configure, I still get the three unwanted LDFLAGS put back in the makefile somehow? Example: Makefile:LDFLAGS = -Wl,-O1,--sort-common,--as-needed,-z,relro,--hash-style=gnu I just want it to read: Makefile:LDFLAGS = -Wl,-O1,-z,relro What magic autotool does this and how do I fix it? The flags show up in the Makefiles and the config.status file. I have completely searched the source code with grep -r and after patching none of the unwanted flags are present before starting the build. When the build craters, I check the Makefiles and config.status and the unwanted flags are back. This results in the build failing with: /bin/sh ../libtool --silent --mode=link --tag=CXX g++ -march=x86-64 -mtune=generic -O2 -pipe -fstack-protector --param=ssp-buffer-size=4 -D_FORTIFY_SOURCE=2 -Wl,-O1,--sort-common,--as-needed,-z,relro,--hash-style=gnu -o embedtool -L/opt/trinity/lib -L/opt/qt3/lib -L/opt/trinity/lib/trinity embedtool.o -lqt-mt -lz -lpng -lz -lm -lXext -lX11 -lSM -lICE -lpthread g++: error: unrecognized option '--sort-common' g++: error: unrecognized option '--as-needed' g++: error: unrecognized option '--hash-style=gnu' What's the trick? Thanks. Extra info: For those interested, the source is: http://www.kde-look.org/content/show.php?content=13969 and the patch I'm applying to try and get rid of the flags is: http://www.3111skyline.com/dl/dt/trinity/arch/src/crystal-1.patch -- David C. Rankin, J.D.,P.E.

3 3

[arch-general] /var/run should not be contained in the package file list.
by 郑文辉(Techlive Zheng) 28 Feb '12

28 Feb '12

Due to /var/run now is a tmp filesystem, so the path /var/run/* should not be packaged anymore. Instead, rc.script should take care aware create needed /var/run stuff. Otherwise, there are a missing wraning while running "pacman -Qk".

7 12

Re: [arch-general] Ways to power on broadcom bcm4311 wireless adaptor without booting into windows
by mercator 28 Feb '12

28 Feb '12

On Sat, 25 Feb 2012 14:41:05 +0100, Tom Gundersen wrote: > Not quite sure what you mean by "touch panel", I guess mine is not > though. It is a Fn+<something> combination. > > Cheers, > > Tom > My wifi switch is not a button, and there's no function key for it. I think it is activated through infrared light sensor or something similar to that. And thank you guys for replying. I think I'd better consider sell this HP and buy a new laptop instead. :P Regards, Tony

1 0