[12:00] Kamion, sivang, https://bugzilla.ubuntu.com/show_bug.cgi?id=1231 [12:03] is there any way to play a Vorbis stream from firefox, e.g. as background audio for a web page? === T-Gone is now known as T-Bone [12:04] quick review of https://bugzilla.ubuntu.com/show_bug.cgi?id=2123 please? [12:04] yuval: thanks, I noticed your updates; I'll have a look tomorrow [12:04] (since most of my other RC stuff is out of the way now) [12:05] yuval: it seems to be a d-i bug too [12:05] he_IL.UTF-8 UTF-8 [12:05] he_IL ISO-8859-8 [12:05] but languagelist has: [12:05] Hebrew;he_IL;he_IL;he;IL;he_IL:he:en_GB:en;kbd=iso08.f16(utf8) [12:06] ... in fact, why don't I just have a look tonight ... [12:06] Kamion: I checked debian installer and there is no problam there. [12:06] really? languagelist seems to be the same [12:07] ubuntu have a more update base-config, maybe something changed. [12:07] Kamion: that patch looks reasonable at plain reading [12:08] thom: actually there's a minor bug in the LENOF macro (insufficient parentheses at one point), although it doesn't affect this case [12:09] yuval: don't see anything relevant in the base-config changelog === thom takes that as his cue to go to bed [12:10] g'night [12:12] thom: but thanks [12:12] will upload now [12:12] np [12:12] Kamion: What is he_IL;he_IL;he;IL;he_IL:he:en_GB:en ? [12:12] i'm almost at matchsticks to hold eyes open here :-) [12:13] yuval: too complicated to explain here, look in the languagechooser source ... [12:13] I'll do that. [12:14] I'm trying it out with the first two he_IL entries changed to he_IL.UTF-8 [12:15] not that I speak Hebrew, but fortunately I can drive the installer almost blind :) [12:15] lol [12:16] Kamion: how confident are you in this archive-copier change? [12:16] mdz: seems to work when I test package-cache-names at the command line [12:16] Kamion: it's tempting to revert to copying everything, considering the late hour [12:17] that seems overkill; this is a really easy change [12:17] Kamion: Do you know how "Hebrew" looks in Hebrew? [12:17] yuval: not even that === Capri [~makolb@mnch-d9ba4683.pool.mediaWays.net] has joined #ubuntu-devel [12:18] Kamion: ok, if you're confident, it's OK with me [12:18] mdz: if you're really uncomfortable with the change to C code I can fix it in debian/postinst, but that's harder to test [12:18] (since I have to boot d-i to test it in that case) [12:19] Kamion: it looks bigger than it is, due to the renaming [12:19] I think it's fine [12:21] uploaded [12:22] I'll be back in half an hour. I go out with my dog. [12:25] Kamion : hand me over an iso when you set it up, I'll test it on my USB kbd [12:25] Kamion : however if the languagle list is the same like debian's , I don't see why it should break. [12:27] that does puzzle me [12:27] I mean, you havn't touched anything on the lists right? [12:28] we are somewhat behind on languagechooser/countrychooser/base-config, and the interactions there are subtle [12:28] BTW : how can I svr or else get U-I's source? [12:28] (and quick to anger) [12:28] sivang: from the source packages in the archive. we have no revision control yet. [12:28] ok [12:29] apt-get source u-i/d-i ? ;) [12:29] it's easy for me, since I uploaded most of it and can just look in ~/src/ubuntu/ :-) [12:29] no, individual source packages [12:29] so e.g. apt-get source languagechooser [12:29] oh [12:29] much better ;) === sivang feels again the need for 3mb downstream broadband.. [12:30] for iso testing..;) [12:31] Kamion : the languagechooser scripty is the whole source? === T-Bone would rather have the opposite of his current setup: 6mb dl/ 0.5mb up. Much better to upload big packages and run servers ;) [12:32] sivang: it has complex interactions; if you're interested in figuring out how d-i works it's probably much easier to check out the upstream source from subversion, at the moment [12:32] Kamion : oh, you removed all the remarks and stuff? :) [12:33] no [12:33] ok [12:33] but it's easier to navigate the upstream tree since it's all in one piece [12:33] erk, he_IL.UTF-8 doesn't even work a little bit for me [12:33] Kamion: btw, i've been unfortunately delayed on the ia64 side, but I haven't dropped d-i stuff there. It's just a 1week delay, most likely [12:37] Kamion : you on the USB kbd? [12:38] sivang: that's not relevant for this part of the bug [12:39] sivang: don't get too distracted by the keyboard thing; it's connected, but it won't be the thing causing text to be unreadable [12:40] oh , _that_ bug [12:40] :) [12:40] I thought we were discussing the one that fails right when d-i starts. [12:40] *that makes d-i fail on start [12:41] no [12:41] hm [12:41] well, there seem to be two conflated things in that bug [12:42] I'm talking about Yuval's report of corrupted text in the second stage [12:42] which is really a different bug and probably shouldn't be part of #1231 [12:42] right. [12:43] I never went on with the hebrew install fearing it would break other thigns, so I never got to see this ;-) [12:43] sorry... he_IL.UTF-8 doesn't change anything? [12:43] moreover, I don't think I could survive a complete hebrew d-i install ;-)) [12:44] sivang: Why not? [12:44] yuval: it's corrupted differently; I'm trying to investigate [12:47] yuval : ah, it's a long story. I am so used to english on computers by now, it freaks me sometimes people have "enabled" setups. I'm using english to correspond even with my closest freinds, so I don't really need it. Plus if you have ever tried KDE/GNOME with hebrew, it's not the most pleasent experience trying to figure out where everything is put. issue of taste and experience. [12:48] yuval : using a bank's site or anything close does require me to have the suitable fonts for hebrew display on the browser, however all interactions with them are numbers and english password letters - so everything's fine. [12:49] oh, we don't have jfbterm in base, that might explain it ... [12:49] yuval : But I sure like to see Ubuntu works well with it, just for the sake of hebrew users. [12:50] maybe it would be easier to attack the Greek bug first [12:50] boy it uses iconv [12:50] Kamion : there's a related greek one? [12:53] I think so, if only I can find it [12:54] maybe I'm mixing up Debian and Ubuntu bugs [12:55] oh, there it is, #1683 === elmo_ [~james@george.kkhotels.co.uk] has joined #ubuntu-devel [01:00] Kamion : I can't explain why, but base-config sure attracts me. reading the sources is the _only_ way to get aroud it? [01:01] I don't understand; could you rephrase? [01:03] Kamion : ahhm. sorry. I meant, in general reading the package's srcs would be the best way to understand base-config and it's inner workings? [01:03] that is the case for software in general [01:04] Kamion : true, however having a pre background about it can help understanding it. When I do so, I usually come up with new questions, that usually are left unanswered unless I bug someone on IRC ;-) [01:04] Kamion : "When I do so" = when I read the source === sivang [~pooh@80.179.82.182.forward.012.net.il] has joined #ubuntu-devel [01:31] do we have a script to automagically set up nvidia? (think i've seen that on the wiki somewhere) === sivang [~pooh@80.179.82.182.forward.012.net.il] has joined #ubuntu-devel [01:44] daniels: OK, I'm a substantial way through investigating this locale problem, so I can take the Greek bug back if you like; it is indeed a subtle set of interactions ... [01:46] we need languagechooser to set up /etc/environment in order for /etc/init.d/console-screen.sh to work in order for UTF-8 to have any chance of being set up correctly in order for ... [01:47] having hebrew, greek and other display correctly for ... :) [01:47] quite [01:48] ow, gums bleeding, this can't be good [01:48] think I'll go to bed [01:48] heh [01:48] have a good night [01:48] Good night Kamion === justdave_ [~dave@24.247.63.44.gha.mi.chartermi.net] has joined #ubuntu-devel [01:53] Good night (01:53 in israel) === T-Bone gently pokes lamont, just in case [01:55] I think he's still away, he said he had more of a couple of things to do ;) [02:05] mdz: so i have some interesting bugs to figure out [02:06] ubuntu, on pia's oldish p2 toshiba satellite is incredibly slow [02:06] to install and to run [02:06] sid runs at full speed [02:06] is DMA disabled? [02:06] however, running sid with the ubuntu kernel demonstrates the slowness [02:06] nup [02:06] doesn't seem to be i/o related [02:07] everything runs slowly [02:07] what kernel did you use under sid? [02:08] I had a similar machine (though k6 rather than p2), and ubuntu performed similarly to debian [02:08] she's currently using a 2.4 kernel; will have to try a different 2.6 when she comes back from cairns [02:08] did you try the -686 kernel? [02:09] hrm [02:09] i think she might have [02:09] but i'll check that too [02:13] jdub: you have a powerpc machine handy? [02:13] (running ubuntu) [02:13] someone please try installing libopenhbci-dev on powerpc [02:16] hrm, it's totally out of date ;) === npmccallum [~npmccallu@69-162-252-7.ironoh.adelphia.net] has joined #ubuntu-devel [03:06] "Disable challenge-response authentication?" [03:10] jdub: Kamion [03:10] jdub: #1586 [03:11] ahr === natem [~natem@69-162-252-7.ironoh.adelphia.net] has joined #ubuntu-devel === npmccallum [~npmccallu@69-162-252-7.ironoh.adelphia.net] has joined #ubuntu-devel === sjoerd_ [~sjoerd@fire.ipv6.luon.net] has joined #ubuntu-devel === azeem [~mbanck@socks-out.lrz-muenchen.de] has joined #ubuntu-devel [04:16] Kamion: sure, you can have it; seems you figured it out while I was asleep :) === mdz [~mdz@4.13.160.19] has joined #ubuntu-devel [04:25] rc bugs dropping like drunken college chicks [04:25] way to go. [04:48] jdub : i have had witnessed horroble slowness with ubuntu using kerlen 2.6.8-386 on a dell inspiron 8200 [04:49] jdub : I tried swithicng to 686 kenrel - didn't help either [04:51] jdub : this is a 1.8Ghz machine, 7200 rpm 256MB ram [05:12] past 5AM here, time to get some sleep. Bye all. === mdz [~mdz@69-167-148-207.vnnyca.adelphia.net] has joined #ubuntu-devel [05:54] morning guys [06:05] mdz: permission to upload fontconfig for 2122, s/|/, in debian/control ofr laptop-detect [06:06] jdub? [06:06] or any other devel === azeem_ [~mbanck@socks-out.lrz-muenchen.de] has joined #ubuntu-devel [06:14] fabbione: yes, that's trivial, no need to ask [06:15] usual peer-review ;) [06:15] never lose the good habits :) [06:16] mdz: i have a wacom driver for the submitter to test and some keyboard layout almost pending (waiting Denis Barber to commit) [06:16] the latter ARE important for us to have [06:16] ok [06:19] moof [06:19] hey lamont === lamont is going to be very sore tomorrow. [06:24] lamont: uh what happened? [06:24] must. learn. to. fall. better. [06:24] oof. [06:26] ahi [06:28] fabbione: do you need a ppc login to build a driver for that guy? [06:30] morning [06:31] mdz: oh right... he has a ppc [06:31] mdz: no i can build in the chroot [06:31] ok [06:34] mdz: 2113.... should we add cupsys-driver-gimpprint* to supported? [06:34] (works for the user...) === lamont thinks actually desktop would be good.. [06:35] I think it's too late for that === lamont has issues with adding the drivers yet again... [06:37] certainly want them for hoary... === lamont could probably be talked into adding bfc-s200-hack.ppd to the cupsys package... :-) [06:41] ubuntu-users is way too high traffic now [06:43] yep [06:51] 12MB in the last half of Sept, 7MB in the first 6 days of Oct. sheesh [06:55] mdz : I'm occasionaly stung when I see you respond almost every user with support request. How do you keep up? ;) === sivang has long given up on reading it as a whole. trying only to notice relavent parts for him to support/help. === fabbione wisper to sivang that mdz is a manager and therefor sit on the chair *almost* doing nothing :P [06:57] sivang: mdz is a bot [06:57] that's the truth [06:57] we created an AI living form even able to walk on walls [06:57] like spideman [06:57] he's so fast also, [06:57] spiderman even ;) [06:58] amazing [06:58] :)) [06:59] are we going to have to bug system tp accept http://bugzilla.u.c/bug# ? [06:59] *the [06:59] sivang: it already does [06:59] fabbione: :-P [06:59] oopst [06:59] ok. [06:59] mdz : thanks :) [07:00] mdz : are you going to leave fabbione get away with this? :-) [07:00] sivang: my revenge is swift and without warning [07:00] when he least expects it [07:01] the black helicopters will hover overhead... [07:01] haha [07:01] ahhah [07:01] boy, i'm after a sleepless night, almost dropps and can't stop [07:01] fabbione : remeber the drug? [07:01] I think I am in it bad.... [07:04] sivang: yes i do :-) [07:04] mdz: a patch for 2111 is very very simple [07:04] mdz: the driver code is nice and dandy ;) [07:04] why do I get the feeling you are not entirely serious? [07:05] mdz: i am serious [07:05] it's a 3 lines change [07:05] right now is: [07:05] use_bios = TRUE; [07:05] it will be: [07:06] if CARD == foo || CARD == bar { use_bios=FALSE [07:06] and so on [07:06] the line after will load the override (if any) from XF86Config-4 [07:06] as simple as this ;) [07:07] there are only to ProSavageDDR models defined/known in the driver [07:07] that makes things easier [07:07] I hate NFS [07:07] oh I hate it so === daf [daf@muse.19inch.net] has joined #ubuntu-devel [07:11] mdz: just repeat after me 'NFS stands for "Not a File System"' [07:11] night everybody - hitting bed after 20hrs of uptime ;-) wish I had some auxilary power source to connect to.. [07:11] NFS stands for "Not a File System" === sivang wishes he had a pesronal human docking station. [07:13] I am finally caught up on email [07:13] from taking sunday off [07:13] -users is out of control [07:13] mdz: definitely. No slow s-curve of adoption here... [07:14] i don't think we expected such a boom [07:14] did we? [07:14] fabbione: hoped is the word, I think. [07:15] aw, ubuntu-traffic didn't make LWN this week [07:16] we hoped for a boom... but i don't think we were ready for one of this size.. that's my impression at least [07:16] mdz: so do you want a new cupsys upload with the file then, or we defer moving cupsys-driver-gimpprint's move til hoary and downgrade the bug? [07:17] lamont: that bug isn't RC anyway, is it? [07:17] if it is, that's wrong === lamont checks again [07:17] i want to thank you guys, my gnome-login-bonobo-hang-error was no longer there this morning, thanks! [07:17] 3 of the remaining 14 bugs are powerpc bugs :-/ [07:18] normal [07:18] doko: what is the status of tetex-base? [07:18] (#2066) [07:20] mdz: removing the references in the documentation to the removed .html file. should be done in some minutes [07:20] ok [07:48] fabbione: pong [07:50] jdub: too late :-) === lamont prepares to head for bed. [07:52] good night lamont === lamont wonders if there is anything he can do to help progress on any of the RC bugs [07:55] anyrate, sleep until morning.. === Mithrandir waves [08:19] is "reject_nondigits" something that debconf uses? [08:20] no [08:20] never mind [08:22] jdub: this new firefox is supremely buggy [08:27] mdz: should we consider switching to gcc 3.4 for it? Works on AMD64. :) [08:28] seems to work on x86 now for the most part as well [08:28] besides a pretty major b0rkage with sse2 [08:29] tseng: do you have a p4? [08:29] i have a p4 and a p-m [08:29] could you try reproducing 1854? [08:30] i have no HT [08:30] mdz (or someone else): somebody wants to review tetex-base (#2066) before upload? chinstrap:~doko/tetex/ [08:30] and no apt-get lameness on gzip either [08:30] tseng: ok. :/ [08:30] doko: attach a diff to the bug? [08:30] tseng: it doesn't show without HT, it seems. [08:30] Mithrandir: my friend joe is using an HT p4 w/ no serious problems either [08:31] Mithrandir: smp 686 kernel [08:31] mdz: ok, but debdiff doesn't show removed files. === pitti [~martin@195.227.105.180] has joined #ubuntu-devel [08:31] tseng: hmm, ok. I find it somewhat strange that I'm the only one able to reproduce it. [08:32] ah, you can? [08:32] yes, it's I who filed it. [08:32] :) [08:32] foo() { echo "" } [08:32] bar=$(foo) [08:32] what can cause something like that to hang? [08:32] and I'm hunting for somebody else who can confirm it, if nothing else to prove I'm not on serious amounts of crack. [08:33] in a debconf environment [08:33] this happens on any apt-get upgrade? [08:34] tseng: no :/ ; apt-extracttemplates seems to hang randomly about 10% of the time. [08:34] mmmm, random [08:35] of course, stracing the problem makes it go away. [08:35] is apt-extracttemplayes heavily threaded? [08:35] which probably means it's some sort of a race issue [08:36] I don't know, mdz could tell? [08:36] Mithrandir: can you try to call sync between each call to apt-extracttemplates? [08:36] well mdz alluded to nptl there [08:36] we fought some funny bugs with nptl and apps making alot of threads just crapping out [08:37] but it didnt seem to affect distros that ship a hybrid glibc [08:37] fabbione: same problem. [08:37] tseng: can I list the threads somehow, or do they still show up as separate processes in ps? [08:38] not under nptl [08:38] they look like a single process [08:38] Mithrandir: do you have any idea for the sh script above? [08:38] I doubt apt-extracttemplates uses many, if any, threads. [08:38] but that doesnt explain much about the non-SMP kernel [08:38] fabbione: are you sure you aren't talking to debconf? [08:38] ya.. [08:38] the race idea is valid also [08:39] Mithrandir: it's the config.in for xserver-xfree86 [08:39] since it only happens once in about ten times. [08:39] Mithrandir: it loads debconf that's clear [08:39] Mithrandir: it hangs in a function similar to the one i wrote above [08:39] Mithrandir: see https://bugzilla.ubuntu.com/attachment.cgi?id=355 [08:40] fabbione: yes, but the foo=$(bar) might try to read from debconf since stdout is redirected. [08:40] mdz: done [08:40] tseng: and I have to kill -9 the gzip processes. [08:41] hmm, weird. [08:41] if I reproduce the problem, it doesn't seem to go away until I start another process (ls, dmesg, whatever) [08:41] oh, it does [08:41] it's just slow === Mithrandir scratches head. [08:52] doko: debdiff seems to show the removed file OK [08:53] doko: the diff looks fine to me [09:01] doko: will you upload it? [09:02] mdz: done === ..[topic/#ubuntu-devel:doko] : Ubuntu development -- general discussion on #ubuntu | 13 RC bugs to go [09:08] mdz: the libnet-ph-perl upload discussed in email: does the control file section need to be changed to universe/* [09:13] doko: no, it is overridden [09:13] ok, uploading [09:22] mdz: now down to 12 RCs? === ..[topic/#ubuntu-devel:mdz] : Ubuntu development -- general discussion on #ubuntu | 12 RC bugs to go [09:23] 1080 is getting downgraded as well; it's way too late to mess with fixing bug-buddy [09:24] doko: you forgot to close #2066? === doko_ [doko@dsl-084-057-047-164.arcor-ip.net] has joined #ubuntu-devel [09:34] jdub: hmm, I guess that message is going to show up for everyone who installed the preview and then upgrades to final [09:35] that kind of sucks [09:35] um, guys [09:35] mdz: wvdial is being interactive on me during the initial installation [09:35] mdz: can't we stop it doing that? debconfization should be two hours' work for somebody who knows debconf reasonably well [09:35] Kamion: pitti fixed that yesterday, I thought [09:36] maybe it didn't make the CD build? [09:36] ah, maybe it missed my CD [09:36] I think I have yesterday's here [09:36] Kamion: #2069 [09:36] OK, fine [09:37] but hey, SHINY GREEK TEXT [09:37] kamion: should the pmac sound problem be fixed on yesterday's powerpc cd? [09:38] Kamion: any idea what's going on with #1997? [09:38] doko_: doko: you forgot to close #2066? [09:38] doko_: should think so, yeah [09:39] mdz: absolutely no idea; I'm tempted to ask the submitter to try yet another daily just to be sure [09:39] but the CD he claims to have really should have all the required code [09:40] Kamion: it doesn't look like we've confirmed that the module is being loaded === ..[topic/#ubuntu-devel:doko_] : Ubuntu development -- general discussion on #ubuntu | 11 RC bugs to go [09:40] mdz: done :) [09:40] maybe it is in fact being loaded, but something else is wrong [09:40] it's rather hard to tell without a keyboard [09:41] true [09:41] brb [09:41] I thought ohci-hcd was built in on powerpc for just this reason [09:42] it's not; it really shouldn't need to be [09:42] hmm, so there's no keyboard until hotplug runs? that's rather scary [09:42] what happens if fsck fails? [09:43] dunno :) [09:43] mdz: fix in #1683 OK by you, in principle? I need to do more testing [09:43] for Greek it works for me all the way up to a desktop [09:43] incidentally I really must learn Greek, because it is so much prettier [09:44] Kamion: so the bug is that it wasn't generating the locale? [09:44] it wasn't generating it early enough [09:44] ah [09:44] it was being generated in base-config or something, but that's too late; it needs to be there by the time /etc/init.d/console-screen.sh runs [09:44] and the fix is to have prebaseconfig generate the locale in the target? [09:45] if so, sounds fine [09:45] right [09:46] anyway, I need to shut down, taxi coming soon; back in some hours === seb128 [~seb128@ANancy-111-1-23-184.w81-50.abo.wanadoo.fr] has joined #ubuntu-devel [09:57] morning [09:59] pitti: ping === sjoerd_ is now known as sjoerd [09:59] hey sjoerd [09:59] seb128: morning :) === elmo_ [~james@host217-37-231-28.in-addr.btopenworld.com] has joined #ubuntu-devel [10:10] sjoerd: pong [10:11] sjoerd: sorry, I was away for half an hour [10:11] pitti: haha [10:11] sjoerd: ? [10:11] pitti: don't apply davidz version of my patch, it's wrong [10:11] sjoerd: the one for the first hotplug event? [10:11] pitti: that you say sorry for being away half an hour :) [10:11] pitti: yes [10:11] sjoerd: last_seq_number or so [10:13] pitti: it can kill off hotplug processing of hal, if events come at a bad timing [10:13] sjoerd: well, since the current version works great, I rather not change anything unless I find a bug [10:13] pitti: i'll send a mail explainign why, after i finish math homework :) [10:13] sjoerd: but thanks for letting me know. Did you already talk to David about this? [10:13] sjoerd: ah, okay. [10:14] david is probably still sleeping anyway === daniels [daniel@fooishbar.org] has joined #ubuntu-devel === ross [~ross@82-44-126-41.cable.ubr03.croy.blueyonder.co.uk] has joined #ubuntu-devel === azeem_ is now known as azeem [11:15] Mithrandir: hrm, do you know if firefox-locale-no has a PR1.1 release? [11:16] thom: it does, I just haven't made a package yet. [11:16] at least pr1.0 release. [11:18] oh, rock [11:18] we might get it locaalised to one language then [11:18] afaics, french and german hasn't happened yet [11:19] I can look at fixing it up and dropping the package into experimental tonight [11:19] thom: it's so nice [11:20] thom: remind me to buy you a beer some day [11:20] *g* [11:20] told ya [11:21] thom: except mine is superior to yours [11:21] jdub, mdz: IMHO we can downgrade #2131, do you agree? [11:21] thom: they must've fixed the keyboard layout in a later revision :) [11:22] thom: your keyboard LED is white, no? [11:22] no, orange [11:22] ahr, mine's orange too [11:23] I thought yours was white for some reason [11:23] strikes me as kind of odd to put an orange LED there [11:23] when white provides more light and IMO looks better [11:23] ho hum [11:24] dinner time [11:28] Kamion: fun and games with openssh-server this morning [11:30] Kamion: i installed openssh-server a month ago and never touched the config [11:30] this morning'supdate is asking me about passwords being disabled and challenge response... [11:30] seems like that quesiton assumes the default is passwordauth yes, so if your config is "no" then it assumes you must have edited it [11:31] but i think your default is passwordauth no [11:31] so this question is getting asked [11:31] can we fix that please? [11:38] seb128: morning [11:38] Setting up desktop-file-utils (0.9-0ubuntu1) ... [11:38]