May 2011 - Arch-general - lists.archlinux.org

Re: [arch-general] [aur-general] Orphans [extra] repository cleanup
by Heiko Baums 12 May '11

12 May '11

Am Thu, 12 May 2011 13:53:10 +0200 schrieb Andrea Scarpino <andrea(a)archlinux.org>: > # These will be moved to AUR > krusader > > If there are no objections I'll move all them tomorrow. I have one objection. Krusader should stay in the binary repos. It's simply the best file manager I know. Heiko

1 0

[arch-general] conntrack does not display bytes and packets
by Vieri 12 May '11

12 May '11

Hi, Since I updated my kernel to 2.6.36 and 2.6.37, /proc/net/nf_conntrack and conntrack -L do NOT show "packets" and "bytes" even for ASSURED / REPLIED connections. It seems that both of these data (bytes and packets) have disappeared from conntrack/netfilter. Am I missing something? Is there an option somewhere to restore earlier behavior (previous kernels/netfilter would always display both bytes and packets)? Example when opening google: ipv4 2 tcp 6 431999 ESTABLISHED src=10.215.144.48 dst=209.85.147.104 sport=52531 dport=80 src=209.85.147.104 dst=10.215.144.48 sport=80 dport=52531 [ASSURED] mark=0 secmark=0 use=2 ipv4 2 tcp 6 431999 ESTABLISHED src=10.215.144.48 dst=209.85.147.104 sport=52533 dport=80 src=209.85.147.104 dst=10.215.144.48 sport=80 dport=52533 [ASSURED] mark=0 secmark=0 use=2 ipv4 2 tcp 6 431999 ESTABLISHED src=10.215.144.48 dst=209.85.147.104 sport=52530 dport=80 src=209.85.147.104 dst=10.215.144.48 sport=80 dport=52530 [ASSURED] mark=0 secmark=0 use=2 ipv4 2 tcp 6 431998 ESTABLISHED src=10.215.144.48 dst=209.85.147.104 sport=52532 dport=80 src=209.85.147.104 dst=10.215.144.48 sport=80 dport=52532 [ASSURED] mark=0 secmark=0 use=2 Previous kernel/netfilter would show something like this: tcp 6 73 TIME_WAIT src=10.215.144.7 dst=209.85.147.104 sport=35407 dport=80 packets=15 bytes=1386 src=209.85.147.104 dst=192.168.92.2 sport=80 dport=35407 packets=11 bytes=10180 [ASSURED] mark=1 use=1 (note "bytes" and "packets") Thanks, Vieri

1 1

[arch-general] Cannot Find Desktop in GNOME3
by Steve Holmes 11 May '11

11 May '11

After upgrading my box to GNOME 3 last week, I can't seem to find the desktop anywhere on my gnome setup. I use the keyboard exclusively for navigating around so I would normally use Ctrl+Alt+d to open the desktop for keyboard focus. Alas I cannot do that any more. Also, with prior versions of Gnome, pressing Ctrl+Alt+Tab would toggle you through the Bottom Panel, Top Panel, and the Desktop. Now it only toggles between the top and bottom panels. Has the desktop, as we know it, gone away? At the moment, I am using fallback mode because gnome-shell is completely broken as far as screen reader accessibility is concerned. Any ideas would be greatly appreciated; thanks.

13 23

[arch-general] [signoff] kernel26-lts 2.6.32.40-1
by Tobias Powalowski 10 May '11

10 May '11

Latest LTS kernel is in testing, please signoff for both arches greetings tpowa -- Tobias Powalowski Archlinux Developer & Package Maintainer (tpowa) http://www.archlinux.org tpowa(a)archlinux.org

1 0

[arch-general] Need for debug - can do i do?
by rafael ff1 10 May '11

10 May '11

HI there! I'm trying to update PCSX2 with some help of its dev team [1], but it is crashing all the time. According to 'gdb' output, it is somehow related to lib32-glibc, but it is omitting some information. I was hoping to be able to activate more verbosity, which AFAIK I can get by compiling it with some debug flags. [2] Is it correct or there is another better way to do it ? Thanks, Rafael. [1] http://code.google.com/p/pcsx2/issues/detail?id=1019 [2] https://wiki.archlinux.org/index.php/Debug_-_Getting_Traces

4 5

[arch-general] Gnome 3, a bug?
by Magnus Therning 09 May '11

09 May '11

This is such a basic issue that I refuse to believe I'm the first to bump into this behaviour. Hence I ask here first for some help. I recently upgraded to Gnome3 and switching between windows with alt+tab is rather broken for me. Here's why I say that: 1. I make sure I have two non-overlapping windows on the screen at the same time. 2. I position the mouse pointer in one of the window. 3. At this point I can't switch focus to the other window with alt+tab. Basically I have two options: 1. Move the mouse pointer so that it's not in any window, then I can use alt+tab. 2. Move the mouse to switch focus. Neither of these is very convenient. Does anyone recognise this? /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus(a)therning.org jabber: magnus(a)therning.org twitter: magthe http://therning.org/magnus I invented the term Object-Oriented, and I can tell you I did not have C++ in mind. -- Alan Kay

8 16

[arch-general] makepkg openjd & ca-certificates-java
by Javier Vasquez 09 May '11

09 May '11

Hi, I've tried building openjdk, but it has a make dependency upon ca-certificates-java, and ca-certificates-jave also has a make dependency upon openjdk: http://projects.archlinux.org/svntogit/packages.git/plain/openjdk6/repos/ex… http://projects.archlinux.org/svntogit/packages.git/plain/ca-certificates-j… How is that resolved when building? Reson for asking is building on non x86 architecture... Thanks, -- Javier.

2 2

Re: [arch-general] [arch-dev-public] Anyone want to maintain acpid?
by Damjan 08 May '11

08 May '11

>>> For some reason, I am assigned as a maintainer for acpid, but I never >>> actually touched it. It is out of date and has a number of bugs. Any takers? >> >> I though this had been deprecated for years? Are there still some >> relevant use cases? If so, there is a new version in AUR: >> <https://aur.archlinux.org/packages.php?ID=33871>. > > People still use it. What has been deprecated is the /proc/acpi/event > interface. > > Though, most acpi events are now translated to input events. Even when they are, acpid still has its use to handle these events on a system level, no matter if you have a user session (*DE) running or if you even have X running. My use case is to handle the suspend and hibernate buttons. -- дамјан

2 1

Re: [arch-general] [pacman-dev] pyqt packages confusion
by Andrea Scarpino 08 May '11

08 May '11

On Saturday 07 May 2011 15:35:24 kachelaqa wrote: > firstly: > > why does python2-pyqt now depend on pyqt (for python3)? why should it be > necessary to drag in python3 and pyqt(3) for a python2 package? In the previous packages the python3 version had the python2 version as its dependence; I changed this because we are going to remove python2 a day. > secondly: > > looking at the file lists, it appears that several critical files are > now missing from python2-pyqt, namely: > > /usr/bin/pyuic4 > /usr/bin/pylupdate4 > /usr/bin/pyrcc4 > /usr/lib/qt/plugins/designer/libpythonplugin.so > (and possibly a few others) > presumably, the intention is that those files will be provided by the > pyqt(3) package instead. which makes no sense to me, because they are > not guaranteed to be compatible. > > IMHO these two packages *must* be kept separate - it should be possible > to install each package completely independently, without any danger of > either one clobbering the files of the other. I, with the help of an user from the community, did a search and asked PyQt developers about this, and they say we can use those files for both versions of PyQt built with different python version. > (nb: it might also be worth noting here that the new python2-pyqt > package also breaks the python2-qscintilla package. bug report: > https://bugs.archlinux.org/task/24144) Fixed. Thanks. -- Andrea

4 5

[arch-general] Pruning the bugtracker
by JM 08 May '11

08 May '11

I have browsed through all High and Medium severity bugreports and categorized some of them here: https://wiki.archlinux.org/index.php/User:Fijam . 'Candidates for closing' are divided into two categories: strong and weak. Strong candidates have not been replied to in over 4 months (with some bugs seeing no activity for over a year) with the last comment asking for more information or confirmation whether the bug still persists. I have not yet started issuing closure requests but will do so in two weeks if noone replies to those reports. Weak candidates have not been replied to in less than four months, the resolution of the bug was unclear or the original submitter found another solution and failed to provide any more information. I will wait for another month before issuing closure requests. Note: jelle van der waa (jelly) asked for confirmation on many of those bugs and deserves all the praise. I have also identified some bugs where more input or a confirmation of a fix is needed and asked for it. Will try to do initial triaging on those bugs or mark them as candidates for closing if the submitter fails to respond. There was also a couple of bug reports that seemed to be going nowhere. There was either a failure in communication, unresolved argument, a patch with no feedback from the developers or a request to split a bug into two or more specific reports. These should probably be reviewed again. There are still Low and Very Low severity bugs to go through, so perhaps some other user wants to pick up where I left :) Cheers, JM

6 7