February 2013 - Arch-general - lists.archlinux.org

Re: [arch-general] [arch-dev-public] [RFC] Add Wayland/Weston
by Oon-Ee Ng 08 Feb '13

08 Feb '13

On Fri, Feb 8, 2013 at 9:05 AM, Gaetan Bisson <bisson(a)archlinux.org> wrote: > [2013-02-08 01:23:30 +0100] Sébastien Luttringer: >> [4] I'm already existed by arch-general be closed again > > I cannot make sense of that sentence... > s/existed/excited then it makes much more sense =)

2 1

[arch-general] Troubleshooting random crash
by Andre Goree 07 Feb '13

07 Feb '13

Hello all, I'm having an issue on my work machine where for the past week, each day when I come in the computer freezes. I'm using KDE 4.9.5 along with 3.7.5-1-ARCH. I believe this started happening after a recent update but I can't know for sure and I can't really reproduce it...it's happening at seemingly random times, however the pattern I see developing is after my PC has been running all day, I'll come in the next morning open Thunderbird and the system just freezes. I can't ctrl+alt+F[1-12] or do anything. Using another system, I'm able to telnet to port 22 on the "frozen" box (I run ssh on this box) but cannot get connected via ssh. My question is, how do I go about troubleshooting this issue? I know there are so many factors involved that it can really be anything, but I'd think if it were an X11/KDE/Thunderbird issue, I'd at least be able to switch to a console and/or SSH to the box. I also don't know if it's in fact a kernel crash because I can telnet to ports on the box. I AM running root BtrFS which I think may be the culprit. I know I know, not production ready, etc. but I'm not actually using this box as a production server, just as my main work desktop so I figured I'd be fine. Also, again, I didn't start having issues until maybe 2 weeks ago -- and I've had this box running since September '12. Luckily I haven't had any filesystem corruption AFAIK, but I'm thinking it's only a matter of time. Any tips on things I could set up to try to capture some sort of output or perhaps a kernel dump (if it's the kernel crashing)? I'm just at a loss here, and I'm not accustom to needing to reboot my Linux PC so ofter (though I have to troubleshoot a certain other OS daily, and that's usually the resolution :p ). -- Andre Goree andre(a)drenet.info

4 7

[arch-general] Strange X11 problem with fresh install
by Fons Adriaensen 05 Feb '13

05 Feb '13

Hello all, I'm experiencing an odd problem with a new system installed yesterday. Video HW is NV44 [GeForce 6200 LE], driver nouveau, login xdm-archlinux, windom manager is Windowmaker (but the same problem occurs with e.g. FVWM). Every _second_ login the following happens: * Input typed into xterm is displayed as spaces. All the rest (including completion and history) gets printed normally. * The top pixel rows of some lines in xterm are not displayed. * Apps displaying pixmaps show random garbage instead. When this happens, /var/log/Xorg.0.log has a line [ 42.200] (EE) NOUVEAU(0): Error allocating scanout buffer: 0 while in the other 50% of cases there are no errors. Any hints, options I could use to avoid this, ... ? TIA, -- FA A world of exhaustive, reliable metadata would be an utopia. It's also a pipe-dream, founded on self-delusion, nerd hubris and hysterically inflated market opportunities. (Cory Doctorow)

1 0

[arch-general] Status of BlueZ update
by Karol Babioch 05 Feb '13

05 Feb '13

Hi, by now BlueZ 5.2 has already been released. I'm wondering what the current status for the package on Archlinux is. I've found this (see [1]) status report over on "arch-dev-public". However this is over a month old now already, and I would really like to see BlueZ updated in the near future. Thanks in advance! Best regards, Karol Babioch [1]: https://mailman.archlinux.org/pipermail/arch-dev-public/2012-December/02418…

3 2

[arch-general] Syslinux update from 4.06 to 5.01
by Myra Nelson 03 Feb '13

03 Feb '13

My setup is vanilla with the boot partition on a separate partition, and this update went smooth as silk. Great job. Myra -- Life's fun when your sick and psychotic!

1 0

[arch-general] Winter Cleanup of [community]
by Alexander Rødseth 01 Feb '13

01 Feb '13

Hi, It's time again for the yearly cleanup of the [community] repository. Somehow, time passed, and it's now too late for a "Christmas Cleanup" like last year. Instead I'm announcing a Winter Cleanup, which I think is a better name as well. The wonderful developers of Arch Web have added a report of "Uneeded Orphans" since last time, which should make this cleanup considerably easier. (Thanks Dan McGee: https://projects.archlinux.org/archweb.git/commit/?id=5379348c9337a4abe27e8…) Another positive development is that all the ibus-packages that nobody seemed to want to adopt, are no longer orphans. Thanks goes out to Felix Yan, one of our latest TUs, for that. The list of unneeded orphans can be viewed at this page: https://www.archlinux.org/devel/reports/unneeded-orphans/ If you know of other packages that deserves to be cleaned up somehow, in [community], AUR or any of the other repositories, please reply to this e-mail and let us know. (If there should be too many requests, we'll start a wiki page this year as well). If there aren't any protests, I'll wait a couple of days and then move all the unneeded orphans in [community] to AUR. Developers and maintainers of [core], [extra] and [multilib] should feel free to join in on the cleanup, of course. I'll round off this invitation with a somewhat pretentious and slightly akward slogan :] "Happy Winter Cleanup. Let our repositories shine!" Sincerely, Alexander Rødseth xyproto / TU

6 9

[arch-general] nmbd fails to start
by Andre Goree 01 Feb '13

01 Feb '13

Having an issue with nmbd failing to start, with no real reason shown in the logs as you can see below: Jan 30 11:10:34 sideswipe systemd[1]: nmbd.service: main process exited, code=exited, status=1/FAILURE Jan 30 11:10:34 sideswipe systemd[1]: Unit nmbd.service entered failed state Jan 30 11:12:21 sideswipe systemd[1]: nmbd.service: main process exited, code=exited, status=1/FAILURE Jan 30 11:12:21 sideswipe systemd[1]: Unit nmbd.service entered failed state smbd.service seems to start fine. I've tried enabling NetworkManager-wait-online.service, but to no avail. I'm wondering why nmbd is not producing any relevant error messages in my logs, too. Any ideas? -- Andre Goree andre(a)drenet.info

5 9

Re: [arch-general] [arch-dev-public] syslinux update to 5.01
by Matthew Gyurgyik 01 Feb '13

01 Feb '13

On 01/31/2013 12:49 PM, Tobias Powalowski wrote: > Am 31.01.2013 18:43, schrieb Pierre Schmitz: >> Am 31.01.2013 16:09, schrieb Tobias Powalowski: >>> Hi, >>> ok syslinux 5.0 series should come to testing again. >>> The problem with this release: >>> You need to copy all .c32 modules to your /boot/syslinux path. >>> - Those who used our shipped install script, will end up in a none menu >>> based syslinux shell. >> As long as we ship this install script we should maintain it. So this >> script needs to be altered to copy the needed files. >> > The script has already been modified to at least to syslinux shell. > If it should do more, Pyther needs to change it. > > greetings > tpowa > I'll copy and paste from my previous message (Re: [arch-dev-public] syslinux 5.00 in [testing]). ----- Below you will find the links to the patches for the syslinux-install_update script, PKGBUILD, and syslinux.cfg During an install, the syslinux-install_update script will copy all .c32 modules to /boot/syslinux. This is recommended by upstream [1]. The size cost is minimal, 996K. For updates, I added an array called core_modules. During an update, we only copy modules that already exist in /boot/syslinux. However, if any core_module does not exist in /boot/syslinux it will be copied/symlinked. With these modifications, when a user upgrades from 4.06 -> 5.00, ldlinux.c32 will be copied/symlinked to /boot/syslinux as it is core_module. Other modules such as libutil_com.c32 and libcom32.c32 will not be copied/linked. On boot, if a menu is being used, the menu will fail to load (missing depends: libutil_com.c32, etc...). However, the user will be given a syslinux shell they can boot by entering a label that corresponds to a defined label in syslinux.cfg. A post_install message or a news item suggesting users to copy / symlink all modules to /boot/syslinux would be ideal. Users who miss this message, will still be able to boot, but instead of the menu loading, they will be dropped to a syslinux shell (as explained above). cp /usr/lib/syslinux/*.c32 /boot/syslinux (/ and /boot on seperate fs) or ln -s /usr/lib/syslinux/*.c32 /boot/syslinux (/ and /boot on same fs) In my opinion, we shouldn't add new modules during an update to /boot/syslinux unless, without the module, the system becomes unbootable. The rational here being - the user knows best. Lastly, since com modules are no longer supported and no one has ported poweroff.com, I have removed the poweroff section from the syslinux.cfg [2]. Patches: http://pyther.net/a/syslinux-5.00-patches-v1/PKGBUILD.diff http://pyther.net/a/syslinux-5.00-patches-v1/syslinux-install_update.patch http://pyther.net/a/syslinux-5.00-patches-v1/syslinux.cfg.patch [1] "In general, unless you have a reason *not* to install all the .c32 files, it is probably a good idea." - hpa [2] #syslinux @freenode: pyther : Hello. Is there a poweroff module for syslinux 5? Ady2 : pyther: no. all .com modules are not supported in 5.00. someone needs to create a new poweroff.c32 compatible with 5.00. Regards, Matthew Gyurgyik

2 1