[05:48] Good morning [06:21] micahg: btw ff seems to work ok for me [06:24] lifeless: cool, thanks :) [06:36] I am kind of a noob. So I have what is probably a noobish question. I have bunch of changes after running pdebuild on a package should I shelve these changes before proposing a merge? [06:43] dpolehn: you shouldn't be running pdebuild in the same dir as the bzr checkout [06:44] oh alright how should I be running it then? [06:45] dpolehn: use bzr-builddeb [06:45] I had a hard time finding good usage documentation [06:45] ok I am testing to see if it will build on oneiric [06:45] still use bzr-builddeb? [06:46] dpolehn: yeah, you can pass it an argument to use whatever build env you want [06:46] great thank you [06:48] dpolehn: you're welcome [07:49] good morning [08:11] good morning === smb` is now known as smb === doko_ is now known as doko [09:33] hello all [09:33] I cannot bridge iwl3945 module [09:34] Which I can do in debian [09:34] any solution [09:34] ? [09:34] anyone ? [09:34] here ? [09:34] Have you tried #ubuntu? That's a better support channel. If you're trying precisely the same steps in Debian and Ubuntu and it's not working, you might want to file a bug. [09:35] Yes I tried [09:35] but seems like its a bug in ubuntu [09:37] since its release [10:18] cjwatson: if you're around, we have a small q in launchpad about Contents.gz and non-utf8 paths [10:18] cjwatson: seeking an off-the-cuff-opinion about LP dropping such paths from Contents[.gz] at some future date (probably a few months out) [10:18] lifeless: I've already asked [10:19] StevenK: ah! [10:19] StevenK: i didn't see it [10:19] Not in this channel :-) [10:22] StevenK: and did you get an indicative response? [10:23] Nope [10:35] lifeless: my answer was "I would not be excessively sad if you felt like dropping non-UTF-8 names from Contents, although it would be better to deal with non-UTF-8 data" [10:45] cjwatson: Hmm I have a weirdo with ssh/sshd; I can ssh locally quickly, but if I pass -X or -XY then it takes a long time before it completes login; it's stuck at "debug2: x11_get_proto: /usr/bin/xauth list :0.0 2>/dev/null", after it unstucks it says "Warning: No xauth data; using fake authentication data for X11 forwarding." [10:46] cjwatson: oddly, if I 'ssh host "/usr/bin/xauth list :0.0"' this completes immediately [10:46] now it gets uglier: if I open a regular ssh (no -X/-XY) in another term while the first one is trying to connect, the other ssh gets stuck too [10:47] it seems sshd is stuck on the server side in some long-running call; maybe a reverse DNS check of some form? [10:48] I don't know, sorry - you'd want to run with -vvv, and possibly run a temporary server on a different port with -ddd in order to see what it's doing [10:48] lool: do you use connection-sharing for the 2nd ssh? [10:48] geser: no [10:48] cjwatson: I ran the client with -v, but not the server; ok, will try running a debug server, thanks [10:48] (s/-v/-vvv/) [10:49] if you're very lucky there's already something useful in server:/var/log/auth.log, but probably not [10:49] Didn't find anything relevantin the logs :-/ [10:49] it's just a stream of sessin opened, disconnect, session closed [10:50] but dumping network traffic, I see no additional DNS request over a regular SSH [10:50] (that is, same requests for ssh -XY/-X or for plain ssh) [10:53] cjwatson: with a sshd -ddd, the last line before the hang is "debug1: server_input_global_request: rtype no-more-sessions@openssh.com want_reply 0" and the first one after the hang is "debug1: server_input_channel_req: channel 0 request x11-req reply 0", which confirms it's X related, albeit I don't understand how [10:53] -4 doesn't change anything [10:55] http://paste.ubuntu.com/629742/ [10:55] perhaps gethostname() is slow [10:56] I'd be inclined to strace the server [10:57] actually, your pause is *before* processing x11-req, why would that be ... [10:57] strace might have private key, right? [10:58] yes [10:58] 9563 11:56:55 select(8, [3 7], [], NULL, NULL) = 1 (in [3]) [10:58] 9563 11:57:15 rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0 [10:58] Now what's fd 8 [10:58] I doubt that approach is useful [10:59] wtmp [10:59] oh, sorry, misread [10:59] well, that's just "waiting for responses from anything at all for 20 seconds" [11:00] it sounds to me as though it's the client pausing, not the server [11:00] lool, thats not fd8, its any of the first 8 fds, though 3 and 7 i think are marked as intersting .. 3 was the one which did something [11:01] apw: yes sorry [11:02] I can see the PAM session started already [11:02] do you have -vvv output where you've noted the position of the pause? [11:03] (client) [11:03] yep, that was the first message when I pinged you; I'll get a full log [11:04] 'ssh host "/usr/bin/xauth list :0.0"' is the result of confusion [11:04] ssh is trying to get xauth data on the client side, not the server side [11:04] lool that shouldn't be remote, but local xorg list [11:04] heh what cjwatson said [11:04] snap [11:05] * apw bets the xauth lock is 'lost' [11:05] oh right [11:05] indeed, this command is hanging here [11:05] gah, sorry for the confusion folks! === jjohansen is now known as jj-afk [11:36] I filed LP #799674 on the xauth issue mentioned earlier [11:36] Launchpad bug 799674 in lightdm (Ubuntu) "/var/run/lightdm permissions break xauth list" [Undecided,New] https://launchpad.net/bugs/799674 === dpm_ is now known as dpm === Ursinha is now known as Ursinha-brb === MacSlow is now known as MacSlow|lunch [13:44] Hey guys quick query I'm running a vm of oneiric and on my netbook too. I noticed in the vm that you get a warning about not being able to write to /run/udev I thought it was just a vm thing however it is also displayed on the netbook is this an issue, also I get a string of modem types when shutting down instead of a splash screen. [13:48] davmor2: it's known but there's no particular need to worry about it for the momenet [13:48] cjwatson: okay cool just thought I'd check :) === bregma_ is now known as bregma === MacSlow|lunch is now known as MacSlow [14:26] anyone core-dev or motu willing/able to sponsor https://code.launchpad.net/~dpolehn-gmail/ubuntu/oneiric/pygame/fix-755980/+merge/65114 ? [14:37] smoser: sure, I'll upload it, hold on a sec [14:40] mvo: hi dude :) [14:40] are you there ? [14:45] hey RoozbehOnline, I'm good, thanks [14:46] mvo: daniel holbach intoduce you to me :) [14:46] mvo: you are developer of ubuntu software center yes ? [14:46] introduce * [14:47] RoozbehOnline: yes, one of the developers indeed [14:47] :) [14:47] good [14:47] i have an idea and im working on since last night [14:48] what is this idea about :) ? [14:48] mvo: my idea (project) is an online ubuntu app center [14:49] with web2.0 features [14:50] RoozbehOnline: nice! there is some django code with some basic work on this already, let me try to find the branch [14:50] :) [14:50] RoozbehOnline: sounds like you and that django project should join forces! [14:50] is it active ? [14:51] yes [14:52] mvo: http://roozbehonline.net/filez/app-get-soon.jpg [14:52] mvo: can you link me ? [14:53] RoozbehOnline: check https://launchpad.net/ubuntu-webcatalog it has a bunch of django code already, I haven't checked yet how complete it is at this point though [14:54] mvo: the problem is the base of django is python [14:54] RoozbehOnline: I'm not really part of that effort, but the ISD people behind it will certainly welcome your input/code/ideas [14:54] mvo: im working with php [14:54] RoozbehOnline: oh, i see [14:56] mvo: if i do this from scratch will you support this project ? [14:56] smoser: the fix for pygame is wrong [14:56] mdeslaur, oh? [14:56] tseliot: ah, another missing module for py3ification of jockey is python3-curl [14:56] it does fix the ftbfs [14:57] smoser: the patch fixes the prompt, but the actual fix is to get the dependencies right so the prompt doesn't get asked [14:57] smoser: Warning, some of the pygame dependencies were not found. Pygame can still [14:57] compile and install, but games that depend on those missing dependencies [14:57] will not run. Would you like to continue the configuration? [Y/n] [14:57] pitti: oh, so I guess it's something we'll have to postpone to Oneiric+1? [14:57] well, thank you for not trusting me, mdeslaur [14:57] * smoser embarrased to not have looked deeper. [14:58] smoser: lol, I just looked as the fix didn't work when I build it locally with sbuild [14:58] smoser: it stops until I hit enter [14:58] tseliot: not necessarily; perhaps python3 now has a non-broken proxy handling for https [14:58] tseliot: I just stumbled over it when trying to port it some more [14:58] mdeslaur, it *did* work for me locally with sbuild [14:58] smoser: huh, curious [14:59] smoser: I'm using a wrapper and stuff around sbuild, maybe I have a terminal set or something [14:59] pitti: ah, good. BTW I'm going to have some fixes for Jockey for multi-arch. We can discuss this in PM [14:59] smoser: anyway, sorry for giving you the impression I don't trust you :) [15:00] my sbuild is running on natty, with my sbuild of 0.62.2-1ubuntu1~natty1 (backport of natty for a dependencies resolution fix). [15:00] http://paste.ubuntu.com/629826/ [15:01] smoser: ok, maybe it's something on my end. Want me to upload it anyway? [15:02] well, i do see that error in my build logwhic his strange [15:04] ah, the one in my log though happens , then patches are a pplied, and it does not happen. [15:06] mdeslaur, i think you're correct though, that something else is amuck. [15:06] smoser: your call, just let me know if you want me to upload it as-is. [15:33] tseliot: hm, no hurry with jockey -- it also needs python3-gobject (which should be doable) and python3-dbus (which won't happen most probably) [15:33] tseliot: and gdbus servers from python are still buggy :( [15:33] pitti: right, you mentioned the lack of -dbus last time [15:33] too bad [15:48] RoozbehOnline: ups, missed your last line. it depends on what you mean with support, but I will welcome it and I don't have anything against it. I would prefer having a single project though :/ [15:49] hey guys :) [15:49] has the glibconfig.h problem been solved yet? [15:53] This is where you come to ask about policies and decisions that have been made for Ubuntu, correct? [15:55] i'm guessing this is multi arch related.. [15:55] but is this right: http://paste.ubuntu.com/629852/ [15:55] libpng only puts stuff in /lib/x86_64-linux-gnu/libpng12.so.0 [15:55] smoser: looks right to me [15:55] the linker knows to look there [15:56] cjwatson, so.. i'm loking at pygame build, it is just asusming that it can look in /usr/lib for dependency libraries [15:56] what would you recommend for that simplistic view of the world ? [15:56] it's not uncommon for stupid build systems to make incorrect assumptions about where libraries live [15:56] they should not make incorrect assumptions [15:57] :) all developers should be as knowledgeable as you, cjwatson, i dont think you'll find argument with that. [15:57] the correct way to link against libpng is to ask pkg-config [15:58] stat: that depends ... [15:58] I'm curious about a choise to disable the "big memory" flag fopr the PAE kernel. [15:58] It seems that the "big memory" flag is disabled by default now. That means you only get a maximum of 3.5GB. If you only get an extra 512MB, why even bother with PAE? [15:59] for the kernel, I'd ask in #ubuntu-kernel instead [15:59] Ah, ok. Thanks. [16:01] cjwatson, ok. so i'll try to move that to use pkg-config. thank you. [16:05] rsalveti, do you know why clutter needed to have the same glx soname for the EGL backend? Hardcoded in some packages using Clutter? [16:08] janimo: problem is the way clutter is handling the backends [16:08] currently you still don't have one common so that then can load the backend one, and have different functionalities selected at run time [16:08] smoser: I think you'd want to do something a bit like DependencyProg in config_unix.py, although the pkg-config calling convention is a bit different (--modversion instead of --version, and ask for modversion separately from cflags/libs) [16:08] like support for gl and gles [16:08] cjwatson, yeah, thats what i was doing. [16:08] janimo: they were still producing different libraries for those different backends [16:09] smoser: *nod* [16:09] janimo: even if the gles abi was almost the same one as the gl so [16:09] rsalveti, and what woudl be lost if we had an egl.so? [16:10] janimo: we would need to build the clutter applications against it if we want gles support [16:10] there was a bug upstream about this, let me try to find [16:10] rsalveti, if an app (gnome shell) wants to use clutter, should it know which backend it uses? Now it looks for glx which can be fixed by renaming our .pc file as well from egl to glx [16:10] but then als includes things from clutter/glx [16:11] janimo: it shouldn't, but currently it's related directly with the backend, as the glx one is the default provider for clutter [16:13] rsalveti, bug 791310 is where I hit the issue. The configure step can be fixed but I am not sure it won't be even more confusing later when GS thinks it uses the GLX backend, but does not have inlcude paths [16:13] Launchpad bug 791310 in gnome-shell (Ubuntu) "gnome-shell version 3.0.1-1 failed to build on armel" [High,Triaged] https://launchpad.net/bugs/791310 [16:13] JanC: http://bugzilla.clutter-project.org/show_bug.cgi?id=2276 [16:13] bugzilla.clutter-project.org bug 2276 in General "Use a single soname for all the clutter flavours" [Normal,New] [16:13] JanC: sorry [16:13] janimo: the upstream bug [16:14] rsalveti, thanks, no progess since last time apparently [16:14] nops [16:16] janimo: so yes, if we want to maintain the same so with both gl and gles support (based on the arch), then we should make it consistent regarding so name, include paths and so on [16:16] otherwise we'll end up having gl support by default on all archs [16:16] and if you want to use gles you'll install the clutter es package and rebuild your application against it [16:17] :q [16:17] argh === wendar is now known as allison [16:19] janimo: currently this is the abi difference between glx and egl: http://paste.ubuntu.com/629870/ === allison is now known as wendar [16:20] egl is just adding more functions, so it should be at least 100% compatible with the glx one [16:30] janimo: problem is, upstream is not going to accept this change for the 1.x series, and will probably also hard to convince debian to apply them [16:31] janimo: upstream would probably just want to fix 2276, and probably against 2.x series === Ursinha-brb is now known as Ursinha === wendar is now known as allison [16:52] mvo: because i cant work with django and python now , i prefer do this with php :) [16:54] mvo: i hope this project become an official project for ubuntu as firefox or chromium home page :) [16:54] in next releases === jj-afk is now known as jjohansen [17:34] cjwatson: The netcfg BOOTIF fix, is that a mac address, or IP address field? [17:34] mac [17:34] super, thanks [17:34] the syntax is defined in pxelinux's docs [17:35] okay, dandy. [17:35] it's basically BOOTIF=01-$(echo "$mac" | tr : -) [17:39] cjwatson: when you next take a look at new packages to sync from Debian please could you pull in libpam4j [17:42] jamespage: done [17:42] cjwatson: thanks! === allison is now known as wendar === beuno-honeymoon is now known as beuno [20:04] geser: ping (DMB if you happen to be around) [20:24] mdeslaur, if you'd like to revisit the pygame fix, i'd appreciate it at https://code.launchpad.net/~smoser/ubuntu/oneiric/pygame/fix-755980/+merge/65255 . [20:26] smoser: ah! nice... [20:28] smoser: that turned into a can of worms apparently :) [20:28] smoser: sorry about that :) [20:32] mdeslaur, well, i think it wuld have regressed without your suggestions. [20:32] its much cleaner now i think. [20:33] i really dont know if its better to run "wrap-and-sort" or not. I had originally suggested it to the original poster. the ubuntu version is newer than the debian. [20:35] smoser: we should only run wrap-and-sort on stuff that's Ubuntu only, it adds an unnecessary diff when merging from Debian [20:39] smoser: As micahg said, I would tend not to run it if the package is in debian. In this case, it looks like the package diverted from debian anyway quite a bit, so...meh... [20:40] smoser: uploaded. thanks! [20:42] actually, there's a new version of pygame on mentors.d.o apparently [21:03] smoser, are you still patch piloting? tedg could use some piloting help on his lp:ubuntu/json-glib branch [21:06] @pilot out === udevbot changed the topic of #ubuntu-devel to: Oneiric Alpha 1 released | Development of Ubuntu (not support, not app development) | #ubuntu for support and general discussion for hardy -> oneiric | #ubuntu-app-devel for application development on Ubuntu | http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: [21:07] bryceh, so , no. :-( [21:12] hah. a couple coworkers just walked past my desk: "do you use ubuntu?" "oh yeah, definitely" [21:14] were you part of the conversation, or just the listener? [21:14] listener === tremolux_ is now known as tremolux [22:30] maco doesn't (AFAIK) use Ubuntu in any case. [22:30] heh [22:30] * micahg thought she liked Aubergine... [22:31] micahg: nope, i'm K user [22:31] maco: I know :) [22:31] When there was Ubuntu (the project) and Ubuntu Desktop (the product) I think one could reasonably infer that any Ubuntu sibling that has official ISOs was a user of Ubuntu, but since Ubuntu is now the desktop, not so much. [22:44] yeah, I don't use Ubuntu either ... [22:44] it's ironically more confusing === AlanChicken is now known as AlanBell === Ursinha` is now known as Ursinha === Ursinha is now known as Guest75959 === jtaylor_ is now known as jtaylor === phantomcircuit is now known as Donald_ === Donald_ is now known as phantomcircuit === Guest75959 is now known as Ursinha === Ursinha is now known as Guest67245 === Guest67245 is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha