[00:01] hmm, ok [00:22] when you select reboot the system in rescue mode, it just dumps you back into the top level menu [00:23] known bug? if not, which package should I be looking at, to file a bug on? [00:23] known bug, fixed in Debian, we forgot to do the sync after RC, will fix for 8.04.1 [00:23] k, thx [00:23] bug 218549 [00:23] Launchpad bug 218549 in rescue "Choosing "Reboot" in rescue menu jumps to main menu -- sync rescue from Debian" [Medium,Fix committed] https://launchpad.net/bugs/218549 [00:35] hi === heno is now known as heno_away [00:57] bryce: Would you have a moment to help me with a release note about multi-monitor support for Kubuntu? As I understand it with displayconfig all we have is xinerama and that's not so much supported. If you can give me a few hints, I can make it sound nice. [00:57] ScottK: sure [00:57] Thanks. [00:57] ScottK: I'm not totally up on where Kubuntu is with xrandr support, but I recall seeing some KDE screenshots that seemed to provide xrandr 1.2 options - is that correct? [00:58] For KDE3 we have what displayconfig provides (I assume kde/gtk would be the same). [00:58] displayconfig uses guidance for its backend, which only supports Xinerama. I'm not sure if xrandr got added in at a higher level [00:59] KDE4 has some kind of xrandr tool, but not AFAIK kde3. [00:59] ScottK: ah, ok. for G/Ubuntu we're deprecating our displayconfig-gtk in favor of gnome's Screen Resolution tool, which is xrandr-based. [01:00] Right, so if you can help me describe the displayconfig situation, that's what I need. [01:00] I know little about X. I mostly made displayconfig and the back end crash less. [01:01] Is there a subset of users that Xinerama will continue to work for? [01:01] ScottK: ok, so the current situation for you is that for most video drivers, dual screen can be set up only via the 'xrandr' command line tool and/or manual xorg.conf settings [01:01] OK. [01:02] ScottK: the Screen Resolution tool we're adding for Ubuntu relies on gnome-settings-daemon to do the heavy work, which kubuntu would not be using, so that tool would not be of use to kubuntu users [01:02] Right. [01:03] (something I'm fairly bummed about, but since this was RedHat's work for the most part, I didn't really have much influence on that) [01:03] I was hoping to be able to say something like, If you have blank types of video then Xinerama will still work and you can use Display and Monitor (displayconfig) otherwise you have to do ... [01:04] KDE4 has at least a basic xrandr tool, but I haven't used it. [01:04] the binary drivers (-nvidia and -fglrx) may still support Xinerama. [01:05] but they have their own multi-screen setup tools, so it would be better to encourage users to use those, than displayconfig anyway [01:05] OK [01:06] the -nv driver for older model cards still support Xinerama. [01:06] OK. === RadiantFire_ is now known as RadiantFire [01:08] ScottK: ATI users can go back to Xinerama by installing the old 6.6.3 version of -ati [01:08] OK. [01:08] * ScottK starts to write something up. [01:10] bryce: BTW, I'd have been totally screwed during this development cycle by fixing displayconfig if not for the xfix option. Thanks very much for that. [01:11] ScottK: for Intel, the legacy -i810 driver supports Xinerama. [01:11] sure === kitterma is now known as ScottK2 [01:32] bryce: What's the official end user type name for the xfix option? [01:32] ogasawara: bug #216927 also sounds like it may be a kernel bug (iwl3945) [01:33] Launchpad bug 216927 in ubuntu "mouse & keyboard became suddenly completely unresponsive" [Undecided,New] https://launchpad.net/bugs/216927 [01:33] ScottK, what was the xfix option again? [01:33] That's the one where you reboot to the 'restore' option and you get the chance to have the system try to fix your broken X automagically [01:34] oh, bulletproof-x? [01:34] Sounds reasonable. [01:34] heh, hadn't heard it described as xfix before [01:34] That's what it calls itself in the boot menu. [01:35] * calc wants kernel modesetting X, so we can get prettier bootup :) [01:36] plus BSOD feature, whee ;-) [01:36] calc: hehe [01:37] well it will be nice to be able to display the error on the screen instead of just freezing [01:37] but it just sounds very funny :) [01:37] bryce: Comments on http://paste.ubuntu-nl.org/64162/ please [01:37] * calc bbl, dinner [01:37] yeah what fedora put out is really more of a proof of concept - it's not even turned on by default. So I think it has a long way to go before it's useful. Still cool to see out in the wild though. [01:40] ScottK, why don't you leave out the "Bullet Proof X" name from that. That was always just a code name. If it's called "xfix" in the boot menu, that's a better name for it [01:40] OK. It does sound cool though. [01:40] also I'm suspicious that xfix may be something completely different from BPX [01:40] Up to you. [01:40] OK. [01:40] yeah leave it out... [01:40] I'll reboot my other laptop to make sure I have it right. [01:44] Three cheers for the forced fsck. It'll be a while. [01:44] bryce: Any other comments? [01:45] ScottK: just one... I don't know your feelings on linking to 3rd party websites, but thinkwiki has the best Xrandr 1.2 config guide I've found so far [01:45] http://www.thinkwiki.org/wiki/Xorg_RandR_1.2 [01:45] I have on my todo list to set up a Ubuntu guide like that... but haven't done it [01:45] I'm all for good information. [01:46] Debian has http://wiki.debian.org/XStrikeForce/HowToRandR12 which is good too. [01:46] but the thinkwiki site addresses Ubuntu more specifically [01:46] It doesn't seem updated for Hardy though. [01:47] its kinda thinkpad specific too [01:53] hrm [01:53] bryce: What would you think about making a stub wiki page on w.u.c that just links to those two for now. I'll link to that and you can fill it out over time. [01:53] alright, well I guess this is as good an opportunity as any to get off my duff and get that guide together. [01:53] Even better. [01:53] ScottK2, I think that is a stellar idea, I'll get on it now. [01:53] well, after coffee. brb [01:53] OK. I need to run out for about an hour. I'll be back later. [01:57] Slight update http://paste.ubuntu-nl.org/64169/ [02:11] https://wiki.ubuntu.com/X/Config [02:11] (WIP obviously) [02:45] Great. [02:45] * ScottK2 is back. [03:02] kudos to the people who refreshed the new http://start.ubuntu.com/8.04/ it looks really nice [03:10] asac: still up? [03:11] crimsun: or you :) [03:11] get a room you guys! [03:12] alex-weej: yes [03:12] crimsun: did you know flash seems to be working fine with the pulse ALSA plug /without/ libflashsupport? [03:13] that is one less reason not to use PA as the default ALSA output [03:13] alex-weej: it works on my config, yes. Then again, FF didn't 'splode with libflashsupport and older Flash, either, here. [03:13] it did for me [03:13] npviewer.bin would frequently segfault [03:14] the only thing we can really do at this stage is grow a larger sampling pool. [03:14] i don't think many of our users are using pulse as the default output without libflashsupport [03:15] you either need the PCM defs in /etc/asound.conf or your own user conf via asoundconf set-pulse or whatever [03:16] yeah, that's why I added it. [03:16] AFAIK, Ubuntu Education Edition has been using it. [03:18] how hard do you think it would be to add a hook in the PA package that added a line to /etc/asound.conf that sourced its default PCM config jazz? [03:18] messing with configs is pretty hairy i guess [03:19] not hard, but it is not recommended and would make several people's neckhair crawl [03:27] crimsun: are you a core dev? [03:27] yes. [03:28] ok, that's good to know [03:28] i commented on 192888 (the flash crash bug) [03:28] basically suggesting we test and then mark libflashsupport as conflicting with the newer flashplugin [03:40] alex-weej: breaks for at least one significant test case as outlined in the bug report. [03:40] oh? which? [03:41] alex-weej: essentially, current hardy PA's default.pa (exclusive hw:X), using pcm pulse plug in an asoundrc, hotplugging another audio device, testing Flash in FF3.0b5 [03:42] crimsun: that's not the Fx bug though [03:42] wait [03:42] maybe it is, sorry [03:42] hotplug another audio device and... Fx crashes? [03:42] no, inaudible audio [03:43] well that's a different bug right? [03:43] yes [03:43] FF doesn't crash, but Flash audio isn't audible [03:43] which is, coincidentially, precisely what the current situation is. [03:43] so let me get this straight [03:43] Flash is using PA [03:44] as in, you can see the audio levels in the volume meter and control it with pavucontrol [03:44] you hotplug an ALSA device [03:44] and suddenly the audio goes silent? [03:44] no, it fails for all audio devices controlled by PA [03:44] fails->inaudible [03:44] only when flash is running? [03:44] only Flash is inaudible [03:45] so that's a pretty small bug in comparison, right? [03:45] it is irrevelant whether other PA clients are running - they continue to be audible [03:45] well, hardy will ship with Flash nondeterministically inaudible [03:46] when you hotplug a DEVICE? [03:46] if libflashsupport is not installed but the pulse pcm plug is used, Flash will be inaudible period. [03:46] yes, regardless whether a device is hot(un)plugged [03:47] crimsun: i don't think that's true -- at least i have pulse PCM here and no libflashsupport and it works fine with no npviewer crashes [03:47] alex-weej: I'm on ia32. No nspluginwrapper. [03:48] i'm still not really sure what you mean here -- is Adobe Flash actually picking up on ALSA device hotplugging and doing something it shouldn't? [03:48] and i'd say the chances of someone having music paused but wanting to listen to youtube are more likely than people hotplugging devices [03:48] no, the pcm pulse alsa-lib plugin fails on ia32 for Flash. [03:49] why did you talk about hotplugging? or was that something else? [03:50] no, it's all related. [03:50] the `pulseaudio -vv' output is there in my comment. [03:50] i guess it's late, sorry i just don't understand [03:50] forgetting hotplugging [03:51] you're saying that setting up the default PCM as pulse [03:51] and trying to use flash without libflashsupport [03:51] fails? [03:51] yes. [03:51] flash 124 doesnt need to depend on libflashsupport as i understand it [03:51] gnomefreak: it doesn't anymore [03:51] the problem is it doesn't CONFLICT [03:51] gnomefreak: it fails horribly on ia32. [03:51] so many people still have it installed [03:51] crimsun: what is your ALSA config? [03:51] crimsun: did you try moving the libflashsupport.so? [03:51] see if it helps [03:51] gnomefreak: I have no libflashsupport.so. [03:52] thats not good [03:52] gnomefreak: what arch you on? [03:52] 686 [03:52] guys, I tested this on ia32 before I started all the dmix & dsnoop fun [03:52] and yours works without libflashsupport? [03:52] gnomefreak: and you're using pulse as your default ALSA output? [03:52] alex-weej: should i just got home from 3 months of being away [03:53] i see [03:53] alex-weej: im working on a few things atm and will test [03:53] believe me, I really don't like the idea of using dmix instead of hw:X, but it is the only audible solution I've found on ia32. [03:53] gnomefreak: yeah, try removing libflashsupport and running asoundconf set-pulseaudio [03:53] then restart your browser and go youtubing [03:54] 'testing with beta5? [03:54] alex-weej: (http://pastebin.com/d17a12c34 is my test config) [03:54] make sure flashplugin-nonfree is uptodaye [03:55] crimsun: does that pulseaudio log repeat over-and-over? [03:55] alex-weej: yes. [03:55] that USED to be what happened for me [03:55] i get sound but its really low [03:55] hm what am i doing, i have an i386 PC sat next to me... [03:55] low or slow? [03:56] low [03:56] not loud [03:56] check your PA volumes [03:56] pavucontrol :) [03:56] can you see Adobe Flash listed as a client? [03:56] might help if it was installed :) [03:56] crimsun: you have the same config as me, FWIW. do you have one in /etc/asound.conf ? [03:57] gnomefreak: Adobe Flash is what we're trying to test here... [03:57] http://www.youtube.com/watch?v=NVky7hwuebU plays [03:57] alex-weej: no, and ~/.asoundrc takes precedence anyhow. [03:57] crimsun: right, but i was wondering if any other PCM's were defined that Flash may be screwing with. [03:57] alex-weej: pavucontrol isnt installed by default [03:57] alex-weej: no, there aren't any additional custom ones. [03:57] gnomefreak: oh right :P [03:58] gnomefreak: do what you have to do then :) [03:58] i am [03:58] sudo :( [03:58] ok no thats not it [03:59] whats wrong with pavucontrol [03:59] gnomefreak: what's the issue? [03:59] gnomefreak@Hardy:~$ pavucontrol [03:59] E: authkey.c: Failed to open cookie file '/home/gnomefreak/.pulse-cookie': Permission denied [03:59] sudo doesnt let it stay open either [03:59] blah, that's a separate PA bug. [03:59] it gives error [03:59] don't run it as sudo [03:59] cant look in it [03:59] are you running pulseaudio as root? [03:59] pkill pulseaudio, rm ~/.pulse* [03:59] no [04:00] sorry, rm -fr ~/.pulse* [04:00] no good together [04:00] wait [04:01] stat ~/.pulse-cookie [04:01] who owns? [04:02] looks like root [04:02] so something is running PA as root... [04:02] just do a sudo killall pulseaudio [04:02] wait [04:02] sudo pulseaudio -k [04:03] not likely but i think i have it fixed atm [04:03] he'll have to nuke ~/.pulse*, too. [04:03] crimsun: pulseaudio shutdown should nuke it no? [04:03] no [04:03] well not after fixing it :( [04:03] hm, wonder how that happened [04:03] yeah just sudo rm .pulse* :P [04:04] did [04:05] ok lets try again [04:07] asac changed dirs for profiles it seems [04:10] i dont have libflashsupport.so at all [04:10] and it seems to work fine here [04:11] PA is a bit on the cranky side atm but flash plays [04:11] Flash "plays" fine here. It's just inaudible due to the pulse pcm plugin in libasound2-plugins choking [04:12] plays as in plays sounds and vid [04:12] crimsun: you are not the only one that i have heard this about but i wrote it off as he didnt update flash [04:12] gnomefreak: with the default pcm set to pulse, correct? [04:13] gnomefreak: is it *definitely* playing through pulse -- you can see the Flash channel in pavucontrol right? [04:13] crimsun: cant find where to set it outside of the command asoundconf set-pulseaudio [04:13] gnomefreak: that's the precise command you need to use [04:13] alex-weej: pavucontrol gets no connection [04:13] it sounds like you're running pulse as root again... [04:14] alex-weej: nope [04:14] something is wrong here [04:14] in my case, it attempts the connection, which connects and abruptly dies [04:14] connection failed: connection refused [04:14] crimsun: pavucontrol!? [04:14] or flash->alsa->pulse? [04:15] pavucontrol gives me the above error [04:15] crimsun: if you run pavucontrol while flash is playing [04:15] do you see the volume control for Adobe Flash appear and disappear all the time? [04:15] here doesnt matter if flash is playing or not [04:15] alex-weej: the stream registers as a pulse client and dies, as shown in the `pulseaudio -vv' log. In pavucontrol, the client "flashes". [04:15] yes, it "appears and disappears" [04:15] crimsun: yeah that's exactly what i used to get on my i386 [04:16] i will check it out in the morning [04:16] why is it looking for root when it wasnt ran as root [04:16] so anyway, aside gnomefreak's pa oddness over authz, that's at least one positive... i say we sit it out and see how many other responses we get [04:16] including my own :) [04:16] gnomefreak@Hardy:~$ pulseaudio -vv [04:16] I: main.c: Called SUID root and real-time/high-priority scheduling was requested in the configuration. However, we lack the necessary priviliges: [04:17] gnomefreak: that's a red herring. [04:17] ah ok [04:17] i have to sleep guys [04:17] thanks, byee! [04:17] bye [04:18] all of this convinces me more that we shouldn't have gone pulse for an LTS. [04:19] well, it has been pretty clear to me that removing pulse is not viable, but we shouldn't neuter it, either. [04:19] Yeah. [04:20] * gnomefreak confused with the output after running the asoundconf set-pulseaudio command pulseaudio -vv is showing alot of alsa lines [04:20] gnomefreak: it should be, given you're using the default hardy pulseaudio [04:20] -vv is quite verbose [04:20] noticed [04:22] crimsun: are you pnly playing the flash or are you running say totem in background? [04:22] gnomefreak: I've tried both cases. [04:22] * gnomefreak goes to find bug i wrote off [04:22] same result in both, since other pulse clients are irrelevant [05:10] ugh, so I think I've got an alsa-plugins fix, but I'm getting some wack skew eerily reminiscent of the old plugins issue. [05:55] good morning [05:57] hi [05:59] * slangasek waves [06:02] hi jsgotangco, hey slangasek [06:03] dholbach: gutten tag! [06:03] hi nxvl: Guten Tag! :) [06:04] heh, here it is gutten nacht [06:04] it's 00:04 [06:04] hehe :) [06:04] Gute Nacht :) [06:14] interesting stat trend at w3schools, their visitors will be majority firefox users by end of next year [06:14] if the percentage users of IE continue to drop at roughly the same rate as the past few years [06:19] hmm curiously [06:19] where is that stat located at w3c? [06:20] http://w3schools.com/browsers/browsers_stats.asp [06:21] should be ~ 42% next jan and ~ 47% by jan 2010 if the trend continues [06:21] which would put IE below 47% [06:22] er all the versions of IE combined that is [06:22] no wonder microsoft is playing so nice with standards for IE8 [06:24] lol [06:25] yeah, [06:25] IE8 is definately better [06:25] at standards then IE7 [06:25] yea [06:25] they are going to have to start catching up with standards as firefox is standards compliant and will be the defacto standard browser soon [06:26] well mostly compliant [06:26] yeah, FF3 is definately a real threat [06:26] to IE [06:26] we'll see if M$ actually continues following the standards tho [06:27] TheMuso: I've cleaned up and pushed the correct fix into alsa-plugins's bzr, but I'm unable to test ATM (lack of sleep) [06:28] ->Z [06:28] crimsun: No problem, I'll have a look a bit later. === asac_ is now known as asac === happinessturtle is now known as nenolod === Whoopie_ is now known as Whoopie [07:04] Spare a thought for cjwatson. [07:04] woops wrong channel [07:13] how look for the stable release [07:14] if i want install a new machine in the next 5h, should i wait for the final? [07:19] use the release candidate and upgrade === Sebast1an is now known as Sebastian [07:44] crimsun: are you still awake? [07:46] crimsun: when you get up it seems this PA+flash bug started in Gutsy from what i gathering from someone suffering from it [07:47] is there a reason that ff can't sort bookmarks? [07:48] calc: no not yet but a bug or 3 on it atm [07:48] i havent seen it happen yet [07:49] gnomefreak: oh ok [07:49] yea it won't let me sort either :\ [07:49] calc: if i get done and sleep ill talk to asa_c or someone about it since ive seen that bug in my email everyday [07:49] there are lots of nice organize bookmark sorting methods but it doesn't keep it that way when you quit the organize screen [07:50] gnomefreak: ok [07:50] xalff3 i take it? [07:50] ack [07:50] and the right click sort is greyed out [07:50] calc: ff3 i take it? [07:50] yea, 8.04 current [07:51] yeah its new to beta5 afair [07:52] ok [07:52] this cant be good [08:17] Good morning [08:19] whee akademy and guadec combined :) [08:29] gnomefreak: welcome back!!! [08:32] Does anybody here have experience with python-central + dh_pycentral ? [08:32] I'm trying to create a python package, which works well for $PYSHARED/mypackage _except_ for $PYSHARED/mypackage/subfolders [08:33] When I install my package, only files directly into the mypackage dir in the pycentral-dir will be available in /usr/lib/python2.x/site-packages/mypackage [08:33] I can't figure out why [08:36] Do I really need to manually make my package install files in /usr/lib/python2.4/site-packages/foo/ AND /usr/lib/python2.5/site-packages/foo/ ? [08:37] I hope this is the right channel? :) [08:38] Flawless: #ubuntu-motu is better, and we have a release happening within hours. Maybe wait a few days. [08:38] Oh, my bad :) [08:38] All the best with 8.04! [08:40] We needs mod_evasive :) [08:49] rock on, ubunteros :-) [08:50] Hey jdub. [08:52] hi jdub [08:53] I'm still used to type jd - but there's 3 jd* now :) [08:53] * dholbach hugs jdub === hunger_t is now known as hunger [08:57] jdub: !!!!!!' [08:57] hey jdub [08:57] jdub, how's down-uder? [09:03] MacSlow: quiet, apparently :P [09:07] MacSlow: rockin' [09:07] MacSlow: the future is awesome. :-) [09:08] jdub, it should be... glad to hear :) === heno_away is now known as heno [09:32] bah, f-spot lacks a good maintainer upstream apparently [09:33] seb128_, gratuitous WONTFIXes? [09:33] mpt: no, but the things lacks a lot of polish, there is over 100 strings which are not in the translation template [09:34] it floods stdout when there is no errors (I've distro patched to workaround a bit the issue) [09:34] it was printing things like "open uri = URI" for every photos in your collection [09:34] debug prints ftw. [09:35] I reported a few bugs and they were all marked WONTFIX because I got the version number wrong by mistake [09:35] debug prints should be conditional to some debug flag to turn in the source or using a command line option [09:35] mpt: utch [09:35] release builds with -DDEBUG.. mmm. :-P [09:35] seb128_: Yup, that's the policy normally. [09:36] Mithrandir: no, the code just has lot of random System.Console.WriteLine calls [09:36] mpt: that's a shame because otherwise the application itself is nice [09:36] seb128_: Seems like the author of that code's been debugging something and not using the debugging mechanism offered by the language. [09:37] And, after the problem has been fixed, the stuff is left there [09:37] $ grep Console.WriteLine * -r | wc -l [09:37] 686 [09:37] it has been debugging a lot then [09:37] heh [09:38] Uh. :) [09:38] seb128_: i guess upstream doesn't believe that anybody is still running applications from a terminal ;) [09:40] and they don't think anybody is using an another locale than english [09:41] seb128_, nice in comparison with gthumb, perhaps, but compared with the proprietary equivalents it's pretty low-grade [09:42] mpt: right === seb128_ is now known as seb128 [10:07] mvo: hi. Upgrades from Xubuntu dapper do not work if universe is not enabled, is this known. [10:07] james_w: no, its not known - we should release note it at this point, I can prepare a fix for -proposed [10:07] mvo: also, I can't get the update-manager gui to offer me a hardy upgrade, though do-release-upgrade works [10:08] is that related? [10:11] james_w: that is with dapper? and the lastest update-manager from dapper-updates? [10:11] mvo: yes and yes [10:12] 0.56~dapper5 [10:12] james_w: that is update-manager-core, I beliefe what is the version of update-manager? [10:13] 0.42.2ubuntu22 [10:13] james_w: please update that to 22.2 (from dapper-updates) [10:13] I tried, but it wasn't offered [10:14] let me try again. [10:20] james_w: I tried an xubuntu dapper->hardy upgrade just now and was offered 8.04 by 'sudo update-manager -d' [10:20] heno: ah, ok, thanks. It looks like an error on my part. [10:21] I rebooted after updating xubuntu dapper, FWIW [10:21] it does fail on calculating the upgrade though, citing a missing *-desktop package, though xubuntu-desktop is installed [10:22] I'll report that separately [10:22] heno: have you enabled universe ? [10:22] heno: you need universe [10:22] * heno tries that [10:23] I can prepare a fix for hardy-proposed now, but it will take a bit until its is available (because hardy-proposed is not open yet) [10:24] if there is a bugnumber I will comment on it (we probably need to talk to the xubuntu devs as well if we just should unconditionally enable universe for people with xubuntu-desktop installed) [10:25] I wouldn't want to add new strings, but then enabling universe without confirmation seems to be not ok too :/ [10:26] mvo: xubuntu is now in universe, so the user will have to turn it on for upgrade anyway [10:26] seems to work now [10:26] I don't see why you would need a confirmation as turning it on is the only way you can upgrade [10:29] my only concern with that is, that because universe has a different level of (security) support, they may not want to upgrade [10:30] but yeah, most of the ciriticla stuff for security upgrade is in main anyway [10:30] so its not a big issue === seb128_ is now known as seb128 [10:32]