[12:09] Is there a way for me to convert my daily system to use all the default artwork? [12:16] hornbeck: right click on the desktop, and select the Ubuntu Default Desktop [12:16] it just did it by its self [12:17] crashed my x for some reason [12:17] than came back up with the new artwork [12:17] who are the people? [12:17] models [12:18] figured [12:18] the default desktop for new installs has the ubuntu logo [12:18] yeah I have it up [12:18] nice brown [12:18] I like it [12:18] but the preview had the calendar selected, so everyone who had the preview and even looked at the desktop options, gets the calendar now [12:19] man for some reason when it crashed it launched another x display [12:19] hmm [12:19] we need to tell them somewhere that more is available [12:20] more artwork, that is [12:20] wiki? [12:21] for instance :) [12:21] there needs be readme like doc on the wiki I suppose. to explain how to get over certain quirks that might get fixed or noticed on the release. [12:22] yeah [12:22] we need to just figure out a way to organize the wiki alittle better [12:22] need a sitemap on it [12:22] sabdfl: congratulats to the idea with the pics :) [12:22] yes, we do. firstly, the documentation areas needs to become the documentation develop area, [12:23] as it's more of a collectiong of docs and stuff. I am going to organize the /UDP page, [12:23] so it will look more as a starting point for people seeking the latest documetation which is not on *.ubuntu.com [12:23] I think we should have one main page for docs that links off to new docs [12:23] (yet) [12:24] well at least we agreed on that [12:24] ok cool. [12:24] amu: thank you [12:24] sivang do you want to start putting that together [12:25] sivang we need to do a site survay to see where all the docs are [12:25] I'm on to it \ [12:25] ok [12:25] yes [12:25] but let's wait for the full response [12:25] everybody, please let us know if this is annoying the RC stuff here..:) [12:26] we might do it on #ubuntu-docs [12:27] :) [12:27] I think we are fine right now [12:28] sivang: did you see the new start page for firefox? [12:28] hornbeck : yes I did, mark sent me the final. [12:29] cool [12:29] hornbeck : however, it doesn't include the gfx, what's on there gfx wise? [12:29] sivang: just the logo at the top [12:29] we need to get some how-tos for the site also [12:29] cool [12:30] or tutorials that is [12:30] agreed [12:30] however, we might want to save all this decisions for until the meeting? [12:30] I am making a list of things to bring up at the meeting [12:30] so a larger crowd would attend and comment, suggest, take up tasks etc? what do you think? [12:30] stuff I think needs to be done [12:30] great [12:31] I am just thinking outload :) [12:31] send over when done. [12:31] did you ever get tomboy working? [12:31] cause it is all in tomboy notes [12:31] I did on the laptop, [12:31] I'll install it here. [12:31] ok, so would you rather a ooo file [12:32] why not it's already installed here :) [12:32] ok [12:32] 2.49 eta for RC testing... [12:34] I am so happy people have keep with my format on the hardware support === ggi [~ggi@ggi.base.supporter.pdpc] has left #ubuntu-devel ["Leaving"] [12:51] can anybody tell if I might get better throughput using rsync instead of the torrent? [12:52] sivang: probably rather the other way around [12:52] sivang: notes on ideas on the way. [12:52] tomboy really needs a export function [12:52] hornbeck : thanks [12:52] no prob [12:53] Mithrandir : you mean rsync is better? Looks like it, the torrent only use 50% of my bandwidth [12:53] sivang: no, the torrent should be better. [12:54] Mithrandir : I've had better results with the rsync untill now. maybe the torrents have dropped (I've used to get those at 98Kb/s) due to heavy load. [12:55] the torrents should go up with load, not doewn [12:56] s/doewn/down/ [12:56] oh right, cause you have more uploaded.. [12:56] yup [12:58] upstream seeds [01:03] I added the info about the doc meeting to my blog [01:03] since I know there are some ubuntu users who read it [01:04] opps that was supposed to be pm to sivang [01:04] no prob :) === nasdaq4088 [sdfsd@tkp-ip-nas-1-p82.telkom-ipnet.co.za] has joined #ubuntu-devel === lamont_r [~lamont@168-103-92-87.dnvr.qwest.net] has joined #ubuntu-devel [02:30] moof [02:53] moof? === lamont_r is between a private lesson and the evening's jujutsu class [03:11] did we hear any more from alex yet? === lamont_r goes to class. on again in a couple hours. [03:12] what's the word on netgear wireless cards? [03:12] prolly 802.11g [03:16] "I'd be upset if my splash screen suddenly featured a tribute to Ronald Reagan, or a reminder that Jesus loves me." [03:16] ...ah, the memory of gimp splash screens past [03:46] hhaa. iso donwloaded. going to test RC on my new SCSI fileserver :-) === hazmat [~hazmat@c-24-15-10-12.client.comcast.net] has joined #ubuntu-devel [04:21] Keybuk: 'The Politically Incorrect Release' [04:29] got some errors on the SCSI AIC-7xxx driver on boot up, but post install seems to be running smoothly [04:30] also got "pciehp" module insertion error [04:30] "Operation not permitted" [04:31] ok, not to download live cd [04:31] *now [04:58] ooh, Python 2.4 [05:04] Keybuk : word has it you're the Python lead and expert for Ubuntu, how do you get accepted to the Python team? :) [05:06] Keybuk : I meant, how can someone can accepted to that fine high lever programming team? :) [05:07] to be honest, I haven't really put a huge amount of thought into it yet [05:07] I'm mostly involved with other Canonical projects at the moment, and spending only relatively little time on Ubuntu [05:08] Keybuk : Oh, I see. ARCH, rozetta and freinds? === azeem_ [~mbanck@socks-out.lrz-muenchen.de] has joined #ubuntu-devel [05:08] the general focus will be getting Python scripting support into every application [05:09] and brushing up and evangelising that which already exists [05:09] yeah, I'm working on the really kick-arse arch stuff :p [05:10] Keybuk : does it menifest on arch's CVS? or , coming to think of it, ARCH's arch archive? :-) [05:11] nah, there's no public manifestation of what I'm working on yet; soon, hopefully [05:12] Keybuk : canonical had spawned a private tree of the code, and would flush it back when done? [05:13] nah, it's not directly related to the arch code; more exciting things you could do with arch :o) [05:14] Keybuk : sounds very apptetiting :) === lamont_r tries to import a disk with sound juicer, and discovers that 'music player' doesn't find any of the .ogg's... [05:15] why is that, I wonder? [05:16] maybe I should finish my upgrade? [05:18] lamont_r : i try to upgrade my dpkg's failing on me...:) [05:18] ouch [05:19] 202 upgraded, 1 newly installed. [05:22] morning guys [05:23] mdz: you around? [05:24] morning fabbione [05:24] up so early? [05:25] feeling better? === hornbeck [~hornbeck@adsl-68-92-227-78.dsl.okcyok.swbell.net] has joined #ubuntu-devel [05:25] fabbione: yes [05:25] sivang: i was ok yesterday too :-) [05:25] fabbione: how is the move going? === lamont_r heads for home. [05:25] mdz: portmap. i can upload it, but didn't you have the packages already? [05:26] fabbione: yes, I have a .changes ready to upload [05:26] mdz: i decided to stay another day here at the old house and have the office on the floor [05:26] mdz: it will give my ISP another day to fix the adsl line at the new house [05:26] ah [05:26] mdz: do you want me to do another .changes or will you upload it directly? [05:27] mdz: in anycase tomorrow i must move out. either to the new house or to the other office [05:27] fabbione: I thought it might be good to wait some time after the RC before uploading new packages [05:27] mdz: make sence [05:28] mdz: but we need to get these new packages tested as well [05:28] fabbione: but on the other hand, I would like for it to get as much testing as possible before final [05:28] :-) [05:28] mdz: i think you should upload it [05:28] fabbione: I think we are both very confident that it does not break the desktop [05:28] mdz: i am confident [05:28] my only concern is users who are running NFS and such [05:28] desktop is working fine here [05:28] you tested client or server or both? [05:28] mdz: yes [05:28] client is 100% ok [05:28] I have been without sleep for a while, so excuse my poor recall :-) [05:28] server needs to be reconfigured in case [05:29] mdz: it's ok.. blame bugzilla that deleted all the nice report i did yesterday [05:29] I wrote some notes in bugzilla [05:29] fabbione: justdave fixed that, by the way [05:29] mdz: yup [05:29] fabbione: do you think there is any _simple_ way we can make the NFS server Just Work if the user installs it? [05:30] I guess the answer is, "wait for hoary" [05:30] when hopefully we can get rid of fam, and follow the plan I outlined in bugzilla [05:30] jdub: what are the odds on replacing fam for Hoary? [05:30] yes i read the comments [05:30] i agree with them [05:31] gamin seems to be active [05:31] so hopefully it will be a viable option [05:31] ok, I guess I will upload [05:31] i mean.. who implements a nfs server has at least a minimum knowledge of filysystems [05:31] and what to look for [05:31] mdz: so the GetModeLine isn't OK now? [05:31] it's not that simple for a newbie to write an exportfs [05:31] but I am not aware of any other unix which ships a portmap listening only on localhost by default [05:31] mdz: too bad for them :-))) [05:32] daniels: we have a list of serious issues to deal with; let's leave trivial items for Hoary [05:32] daniels: what GetModeLine? the 099l ?? [05:33] I don't object to the patch, but we need to change as little as possible between now and final [05:34] mdz: let's discuss X in a few minutes ok? [05:34] fabbione: ok [05:34] i am preparing an interdiff [05:34] but working from the floor doesn't make me any faster ;) [05:34] I would like to give gzip a good workout [05:35] but there is nothing to upgrade :-) === fabbione uploads X to make mdz happy [05:35] har har [05:35] mdz. you can simulate ;) [05:35] dpkg-getselection > log [05:35] apt-get --reinstall install < log [05:35] or something like that [05:36] daniels: do you realize that xscreensaver writes a line to stdout (and thus ~/.xsession-errors) whenever it activates anyway? [05:36] there is a lot of work to be done on eliminating unnecessary disk access, but it is a Hoary project [05:37] fabbione: yeah, 099l [05:37] mdz: ok [05:37] if you try to reinstall many packages at once, apt tends to get confused [05:37] mdz: well .. that's how i saved my server once [05:37] mdz: / was lost in lost+found [05:38] eek [05:38] found dpkg via md5sum [05:38] and the minimum pieces required to run apt [05:38] impressive [05:38] also via md5sum [05:38] that is a lot of pieces [05:38] glibc and nss and libapt and the methods [05:38] mdz: not if you have /var/lib/dpkg/info intact [05:38] script to check md5sums === azeem [~mbanck@socks-out.lrz-muenchen.de] has joined #ubuntu-devel [05:39] and to move files around according to them [05:39] fabbione: I noticed in my testing that when you upgrade portmap, it does not restart :-/ [05:39] once you get the minimum working [05:39] zack [05:39] I do not know if the bug exists in Debian [05:39] hmm it did here [05:39] I don't have any more Debian systems to test [05:39] wait a sec [05:39] every time in my test upgrades, it would still be listening after upgrade [05:39] because it was still the old process [05:40] dpkg - warning: downgrading portmap from 5-4ubuntu4 to 5-4ubuntu3. [05:40] * Stopping portmap daemon... [ ok ] [05:40] Installing new version of config file /etc/init.d/portmap ... [05:40] * Starting portmap daemon... [ ok ] [05:40] * Restoring old RPC service information... [ ok ] [05:40] same when upgrading [05:42] I will test in a moment when I upgrade my NFS server :-) [05:43] once the package is built [05:43] mdz: back in 5 minutes to discuss X === fabbione needs to visit the thinking room [05:43] Mithrandir: ping? [05:43] asleep I suppose === azeem [~mbanck@socks-out.lrz-muenchen.de] has joined #ubuntu-devel [05:52] only Mithrandir, elmo and Kamion seemed to have been able to reproduce the gzip bug [05:52] Mithrandir gave me an account on his system, but his IP seems to have changed [05:53] re [05:54] http://people.ubuntulinux.org/~fabbione/diff [05:54] mdz: ^^ [05:54] reading [05:54] ubuntu24 to ubuntu25 [05:54] fabbione: do you have tiff prepared? [05:54] there are a lot of noisy bits like changelog, CHANGESETS, TODO [05:55] mdz: i woke up when i said goodmorning here [05:55] ah [05:55] mdz: i am going to do it after we discuss the diff [05:55] it won't take too long since the patches are in bugzilla [05:55] don't worry about it [05:58] i only have one pending change to the FAQ [05:58] but i guess that's not an issue ;) [05:59] the doc updates of course I have no problem with [05:59] we need to talk about xkb, v4l and #2328 [05:59] I read #2328, but from only the bug, it is impossible to tell either the problem or the fix :-) [05:59] ok [06:00] mdz: 2328 is a simple fix [06:00] it's at the end of the diff [06:00] it is hard to see with such a long line; what is the change? [06:00] - db_get xserver-xfree86/config/monitor/mode-list [06:00] + if [ -n "MAXRES" ] ; then [06:00] + RET="MAXRES" [06:00] + else [06:00] + db_get xserver-xfree86/config/monitor/mode-list [06:01] mdz: is in xserver-xfree86.postinst.in and .templates [06:01] the end of the diff is a template change [06:01] yes and the bit before it [06:01] there are 2 problems basically [06:01] fabbione: shouldn't that be $MAXRES? [06:01] daniels: correct [06:01] because "MAXRES" will always be strlen() != 0 [06:02] fixed [06:02] mdz: Carlos specific problem is that his resolution [06:02] was not known at @60Hz [06:03] so basically debconf was returning 640x480 @ 60Hz [06:03] as default value [06:03] but this is a specific fix [06:03] reviewing the logic [06:03] i noticed that can happen for all the unknonw resolutions [06:03] now if MAXRES exists it means that we are either probing or guessing [06:04] and it contains the same value as the template (theoretically) [06:04] this change is very uncomfortable to me [06:04] mdz: it's simple [06:04] it's difficult to explain [06:04] it had one bug already [06:05] mdz: i didn't finish all the tests yet [06:05] it's not like i wouldn't have spot it [06:05] the problem with changing anything in the detection logic at this point is that we can never re-test all of the hardware that has been tested [06:05] we have had hundreds or thousands of tests at this point [06:06] mdz: it doesn't really change the logic [06:06] it fixes it [06:06] mdz: if you see the script carefully: [06:06] MAXRES = something [06:06] db_set template $MAXRES [06:06] later: [06:06] db_get template [06:06] yes... [06:06] and than we compare what's in there with the list [06:06] now [06:07] limitation: [06:07] the template might not be aware of that MAXRES [06:07] debconf returns the default of 640x480 @ 60Hz [06:07] the frequencies detection goes banana [06:07] the if then else fi [06:08] simply avoids a limitation to the debconf template [06:08] and it is not intrusive. [06:08] under what circumstances is MAXRES set? [06:08] i am not "growing" the template like i did for the other resolution list [06:08] MAXRES is set only when we cannot detect the frequencies via xresprobe [06:09] # if we are here we did not detect the frequencies, but we have the resolution. [06:09] # get the highest resolution. [06:09] so MAXRES is the answer to the mode question [06:09] so this changes the behaviour for (potentially) every circumstance where the user is asked the mode question [06:10] nope [06:10] because the mode list is all known and mapped to mode-list [06:11] so if the user is promped for the resolution it can only enter well known values of MAXRES [06:11] that will match later on the values into the frequencies case [06:11] i wonder if fixing this would allow high quality displays to do above 100hz freqs :) [06:11] sivang: not in your case [06:12] it's another bug here we are talking about [06:13] mdz: what you need to look at is this: [06:13] # set a sane default for mode-list [06:13] MAXRES="$MAXRES @ 60Hz" [06:13] db_set xserver-xfree86/config/monitor/mode-list "$MAXRES" [06:13] yes [06:14] the problem happens only if MAXRES is not known to the template [06:14] currently, that falls back to the default in the template if $MAXRES is not in the choices [06:14] the default is 1024x768 @ 60 Hz [06:15] mdz: apparently no, if you try to set the template and than db_get [06:15] you get the first value of the list [06:15] that might be a frontend bug [06:15] I agree [06:15] that is a bad debconf behaviour [06:15] frontend ;) [06:17] does it look too dangerous to you? [06:17] otherwise there is another simple way to implement it [06:17] if that can make you feel better [06:17] i can db_get the template [06:18] if the MAXRES is != null and the 2 are the same than ok [06:18] otherwise prefer MAXRES to db_get [06:18] same effect as you see it in the patch [06:18] kust written in a different way [06:18] s/kust/just/ [06:19] here is my concern [06:20] well [06:20] is there any possibility of this affecting any case other than when 640x480 @ 60Hz would be the result? [06:21] how can we test this? [06:21] nope [06:22] mdz: simply asking carlos to test the new package [06:22] he has the testcase ;) [06:22] I am not so concerned about the bug [06:22] and just try to reprobe on well known machines [06:22] it seems to affect very few people [06:22] it is the regression testing [06:22] mdz: well not really a few... [06:22] which cases do we need to test for regressions? [06:23] there were a bunch of people having the 640x480 problem [06:23] where xresprobe fails? [06:23] mdz: where xresprobe returns a resolution but not a frequency set [06:23] you can even wrap xresprobe to simulate that for you [06:23] just write a bogus sh script to return: [06:24] rabidbt: blabla [06:24] ops [06:24] identifier: blabla [06:24] freq: [06:24] res: 1024x768 [06:24] or something like that [06:24] that will trigger the MAXRES [06:24] but you need to use a not known resolution [06:24] no discard my last comment [06:25] it can be also on well known resolution [06:25] it doesn't really matter [06:25] mdz: looks like the preview set calendar as the default, so anyone upgrading a preview to current bits gets people. [06:26] lamont: if that is happening to _anyone_ who upgrades from preview, there is a new and interesting bug somewhere [06:26] it is known to happen to anyone who even breathed on the background preferences dialog [06:27] fabbione: it is all a question of how much testing we can get for it [06:27] mdz: I didn't change background preferences prior to today, and when I pull that up, it says 'ubuntu monthly calendar' [06:27] lamont: you don't have to change them; you only have to look [06:27] I see. [06:27] anyone who has a ~/.gnome2/backgrounds.xml or whatever [06:27] will get a surprise artwork change [06:28] I see [06:28] ok. [06:28] and does removing the file make things better, or worse, I wonder? [06:28] removing the file should give you the new default [06:28] which is the gradient+logo background [06:29] fabbione: thanks for the explanation, I understand the change much better now [06:29] fabbione: can we talk about xkb? [06:29] mdz: so which all prism drivers do we have now? [06:29] mdz: of course. [06:29] lamont: I don't think we add any prism drivers, but we do add the firmware for prism54 [06:29] mdz: we have 2 problems with XKB [06:29] so we have whatever is in the 2.6 tree [06:30] mdz: so prims54 should "just work", or should be "trivial to enable"? [06:30] lamont: should 'just work' post-installation [06:30] mdz: for the testing i can simulate all the cases replacing xresprobe. that's not a problem [06:30] may or may not work in the installer, not sure of the status of that one [06:30] mdz: it was already in my TODO. it's not like i tend to push untested changes [06:31] mdz: let's go with xkb when you are ready [06:31] mdz: what about liveCD? [06:32] lamont: the live CD has a completely different kernel [06:32] fabbione: I need 10 minutes for workrave [06:32] mdz: ok [06:32] wrists are killing me [06:32] mdz: i need to help my gf a few minutes so it fits perfectly ;) [06:32] we publish a livecd RC? [06:34] looks like the answer is 'no' === lamont checks and discovers that his warty mirror will fit on a dvd. Just barely [06:48] well, about 20MB to spare.. :-) [06:50] lamont: I thought you weren't able to build one which booted [06:50] lamont: did you and alex find a solution? [06:50] mdz: haven't heard from alex since he headed for home. [06:51] lamont: please keep on him [06:51] but he had uploaded one, and it looks like we held off on calling that an RC. [06:51] he's on the top of my list. [06:52] the one that he built was based on day-old stuff, rather than the RC packages [06:53] right [06:54] have there been any post-rc uploads yet? [06:54] fabbione: back [06:56] me too [06:56] xbk problems [06:57] there are 2 issues that i am aware of at the moment [06:57] the first is that some layout are "multi layout" capable [06:57] no clue what that means, but that's how it works [06:57] basically these multilayout requires extra definitions in a bunch of files [06:58] the last changes we imported are to support a number of extra multi layout thingy [06:58] and it includes a cleanup in the direction [06:58] (that was between ubuntu23 and ubuntu24) [06:58] apparently there is a regression [06:58] what kind of testing can we do? [06:59] for the Super key [06:59] mdz: no idea really... [06:59] other than testing the minimal stuff [06:59] like the windows key and stuff like that [07:00] it also means that you need to set the layout on the specified language to be able to see if it works [07:00] and need to be familiar with it [07:00] so i am not completely sure how to test stuff in that direction [07:00] the only person that seems to understand xkb is denis [07:00] (and upstream of course) [07:00] now [07:01] regarding the regression is a chicken egg problem at this point in time [07:01] either i revert the changes done between ubuntu23 and ubuntu24 [07:01] but we lose a big bunch of layouts [07:01] or we try to update to the last changes Denis did. [07:01] Also Denis commented a lot in the bug reports [07:01] that most of these problems are also application problems [07:02] and he explained some workaround that I am NOT happy to implement [07:02] and daniels agreed too [07:02] like using the alt:whatever_win option in X [07:03] i truly believe that the best thing to do is keep going on Denis direction [07:03] also becuase the same things will hit Debian soon [07:03] and we will not be too different in that area [07:04] this sounds very risky [07:04] in this area? [07:04] we do not know how to test if the bugs are fixed, we do not know how to test for regressions, and we do not understand the system very well. on all points we basically trust denis [07:04] everything is risky [07:04] mdz: that's what we are doing in debian too [07:05] fabbione: debian is not releasing in a few days [07:05] they will have weeks or months to shake out the bugs [07:05] mdz: that's not my point :-)) [07:05] that neither Branden or I have the knowledge to say what Denis is doing is correct or not [07:05] we cannot afford to take the same risks as Debian unstable [07:05] so he has kinda of "licence to kill" [07:06] mdz: i don't disagree [07:06] what i am saying is that there are 2 options [07:06] we kill the changes [07:06] we keep going on them [07:06] why can we not leave it alone? [07:06] first option: we kill the layouts for a bunch of coutries (including ca) [07:07] second option: we at least try to improve the situation [07:07] i am not leaving it alone because we know that there are bugs and it is not working for everybody [07:07] the fact of killing layouts does not make me happy [07:08] specially when something like ca is involved [07:08] .ca <- [07:09] maybe get the changes on a seperate package , have it downloaded and tested by the reporting parties, just like after preview went out? [07:11] fabbione: how does the gnome keyboard capplet get information about the layouts? [07:11] mdz: my best guess is that at the first startup it checks X config [07:11] mdz: later it stores its own info [07:12] reason why you can see the popup like: "hey you little punk! you changed the X config keyboard setup!!!" [07:12] fabbione: it has text descriptions for all layouts [07:12] which I do not see in the xkb data [07:12] mdz: it depends how it is implemented [07:13] this is another thing to be concerned about [07:13] what i am sure about is that gnome doesn't know the xkb protocol upside down [07:13] and that's why there are inconsistencies [07:14] what i am concerned about is why gnome needs its own keyboard layer, instead of using what is there already [07:15] but that's off topic now [07:15] do you understand my concerns about these changes? [07:15] yes i do [07:15] there are so many unknowns, and we could end up with a serious bug in the release that we can do nothing about === x4m [~max@74.151-200-80.adsl.skynet.be] has joined #ubuntu-devel [07:16] mdz: we can simply bounce the problem to the gnome people :) [07:16] they will not be able to fix it in warty either [07:16] and tell them to write their application compliant to the X protocol [07:16] and they will tell that X has to be gnomified :P [07:17] so i mean i can leave it as it is now [07:17] but i don't think it will change anything [07:19] + * Replace level3(ralt_switch_multikey) by level3(ralt_switch) in all [07:19] + layouts so that symbols on the 4th column may be accessed with [07:19] + Shift+Alt_R or Alt_R+Shift. (Closes: #270235) [07:19] I have no idea what that is about [07:19] and it is only a severity: normal bug in Debian [07:20] I cannot say if the change is correct [07:20] I cannot test it [07:21] how can I approve this kind of change for late inclusion in the release? [07:22] mdz: was there a 'how to get the old gdm splash' writeup somewhere? [07:22] mdz: i am not trying to sell you a prodocut or something.. don't misunderstand me [07:22] lamont: FAQ [07:22] mdz: i dunno that shit either. i am in the same position as you are. both cases we lose. [07:22] dih [07:22] doh [07:22] mdz: either because we don't have the fix or because we might introduce a bug [07:23] fabbione: personally, I still prefer a whitelist to MAXRES [07:23] given a choice of two sets of bugs, I choose the known over the unknown every time [07:23] largely because timings can often be totally absolutely bong [07:23] e.g. 2777x2700 or whatever it was [07:24] mdz: btw, there are about two people in the world with xkb knowledge -- denis and svu@fd.o [07:24] mdz: ok. at what level do you want me to revert? [07:24] fabbione: let's go over the changes one by one [07:24] daniels: and Ivan Pascal [07:24] fabbione: what about this v4l stuff? [07:24] mdz: it fixes the support for the v4l2 bttv [07:25] I didn't even know that X used v4l [07:25] mdz: see the debian bug and the diff. [07:25] what applications use this interface? [07:25] xawtv? [07:25] & CO? [07:25] xawtv opens /dev/video [07:25] or uses Xv [07:25] Xv is for output [07:26] we are talking about input [07:26] if you have a video card with a bttv iface [07:26] (like i ghave on my machine) [07:26] i get the v4l loaded [07:26] otherwise i don't see tv ;) [07:26] fabbione: yeah, but after looking at his code, I'm convinced he doesn't know too much :P [07:26] fabbione: so it is for video cards with built-in capture? [07:27] mdz: yes [07:27] mdz: i saw it loaded also for normal tuners, but i cannot confirm 100% if [07:27] it is used by them [07:28] I do not think it is used for a standalone capture card [07:28] X would not be involved at all [07:28] I have a bttv card [07:28] I do not even load the v4l module [07:28] ok [07:28] fabbione: is your card broken without this patch? [07:28] it does on mine and i have a gfx card with bttv integrated [07:29] mdz: no because mine is a v4l [07:29] the patch fixes v4l2 [07:29] that is the new protocol/standard [07:29] mdz: but yes.. mine works [07:29] very little uses v4l2 [07:30] fabbione: the 1152x768 thing is just adding a mode to the template choices, right? [07:30] yup [07:30] mdz: yes [07:30] mdz: that's how i spotted also the other more generic problem [07:30] so carlos was affected by both of those bugs? [07:31] mdz: yeps [07:31] the missing entry in the template [07:31] and the other thing we discussed before [07:32] it can definetely be improved.. but not for warty [07:32] there were other reports of #2328? [07:33] even carlos did not report this until today :-/ [07:34] mdz: mostly on irc and mailing list [07:34] i don't have all the references at the moment [07:34] but i remember them for sure [07:36] the other bugs we are fixing for final are melting laptops, breaking katie, security issues, and reverting one patch in hal [07:37] yes i read the mail [07:37] the X issues don't seem to be of the same severity to me === ph_ [~ph@pD9E6B877.dip.t-dialin.net] has joined #ubuntu-devel [07:38] ok === jdub [~jdub@home.waugh.id.au] has joined #ubuntu-devel [07:39] mdz: let's drop ubuntu25 and that's it. [07:39] people will have to wait for X.org [07:39] fabbione: I think that is for the best [07:39] mdz: i agree only for one thing [07:39] and possibly the xkb stuff [07:40] but for the other fixes i don't [07:40] in anycase we need to release [07:40] so people will get what's in now [07:42] I know it's hard to know about the bugs and not fix them, but sometimes that is what we need to do [07:46] fabbione: please put 099l and the XKB stuff in a TODO stuff somewhere [07:47] daniels: it's in bugzilla already [07:47] 099l isn't [07:47] daniels: the XKB stuff we will go 100% upstream for X.org [07:47] but I'll file that bug [07:48] not for 6.8 [07:48] i can kick it upstream, but not for a 6.8.x release [07:49] but if we use my modular approach, and use xkbdata from fd.o, I can kick it upstream there [07:52] mdz: are we supposed to get a libxvidcore4-dev somewhere/time? a fair chunk of multiverse is blocked on it, if we care. [07:52] lamont: marillat archive? [07:53] fabbione: it's more to get it shoved into the archive... [07:53] or dsl [07:54] EWINDOW === pitti [~martin@195.227.105.180] has joined #ubuntu-devel [07:54] lamont: i know that i get that package from marillat :-)) [07:55] lamont: that's it.. [07:55] i didn't follow multiverse too much [07:55] ENOTIME === lamont only follows it to the extent that it sends him mail. [07:55] Package: libxvidcore4-dev [07:55] Source: xvidcore [07:55] Maintainer: Christian Marillat [07:56] probably it wants more stuff [07:56] multiverse/doc/xvidcore_1:0.9.2-0.3: Installed [optional:uncompiled] [07:57] and therein lies the issue. [07:57] Morning, guys! === vorlon [vorlon@dsl093-039-086.pdx1.dsl.speakeasy.net] has joined #ubuntu-devel [07:59] ok, why does ubuntu's main archive include ndiswrapper-utils if there are no ndiswrapper module packages? :-) === sivang [~pooh@80.179.66.20.forward.012.net.il] has joined #ubuntu-devel [08:00] just did an rc install. [08:00] vorlon: er, the ndiswrapper module is in the main kernel packages [08:00] daniels: bah, that's cheating! [08:01] apart from d-i no saying "no enough room on drive" when choosing to install to not enough room partition, everything went swift [08:01] it's just ingnores your partitioning actions [08:01] vorlon: did you try 'sudo modprobe ndiswrapper'? :) === lamont sleeps [08:02] daniels: no, it never occurred to me that it'd be part of the stock kernel package. Allow me to express my jealousy on behalf of sarge. ;) === sivang [~pooh@80.179.66.20.forward.012.net.il] has joined #ubuntu-devel [08:03] vorlon: heh, thanks. graciously accepted. ;) [08:03] lamont: dunno about xvidcore, ask elmo [08:04] vorlon: hey [08:04] hey vorlon === vorlon waves to the Night Crew. === mdz yawns [08:05] yeah, I'm reppin' the night crew at 1407 === daniels makes flashy hand signs. === sivang nearly drops infront of the new artwork === lamont makes a note to pass some questions on to silbs in the morning [08:07] darn. my lumix camera still doesn't work.. [08:07] and it's a standard SD flash storage cam [08:08] fabbione: can you take care of gzip? it needs for Tollef to verify the fix since he can reproduce it, and then I think it can be uploaded (more testing is appreciated of course) [08:08] I will need to sleep very soon [08:08] mdz: anything else to fix? i'm currently wandering through the acpi fan/tz bug [08:08] daniels: I posted a list to u-d [08:08] ok, ndiswrapper question solved, now all the system needs is a good tool for managing wireless connections for roaming users. ;) [08:08] basically, severity: critical bugs [08:08] so you did, and I even replied to it [08:09] daniels: #2341 could certairly use regression testing [08:09] mdz: yeah, I'll keep on that [08:12] mdz: sure. go and get a good night rest [08:12] thanks [08:12] mdz: i am doing regression testing on libtiff now [08:12] mdz: pong [08:12] Mithrandir: oh, good [08:12] Mithrandir: I need you to test #1854 [08:12] dhcp-103-140.idi.ntnu.no [08:13] I'll test [08:14] sivang: please try http://www.piware.de/hal_0.2.98-1ubuntu9_i386.deb, maybe it works with your cam [08:14] ok, I'll log out and leave it to you [08:14] pitti : ok, didn't know you were awake..:) [08:14] sivang: I got up at 6am, because I went to my gf yesterday evening [08:15] sivang: just returned home === Mithrandir thinks it has been sillily little traffic on the torrent overnight, ~7G i386, ~2G amd64 ~0.15G ppc. [08:16] pitti : ok going to test your packages === pitti crosses his fingers [08:16] Mithrandir: yes, the response to RC has been understated === pitti notices that one can't type with crossed fingers [08:16] at least in part due to lack of slashdotting [08:17] ok, I'm working on kernels to fix the thermal/fan issue [08:17] new initrd-tools, then rebuilding kernels to deal with that === azeem [~mbanck@socks-out.lrz-muenchen.de] has joined #ubuntu-devel [08:19] mdz: seems good so far, doing a bit more testing before letting you off the hook [08:19] daniels: please go ahead and attach a proper debdiff to the bug; I don't think there is one already [08:19] Mithrandir: do your worst [08:19] pitti : should I have done anything but dpkg - and reconnect the camera? [08:19] sivang: should work (if it works) [08:20] pitti : didn't :( [08:20] sivang: damn! [08:20] mdz: will do [08:20] fabbione: portmap just did it again [08:20] tcp 0 0 *:sunrpc *:* LISTEN [08:20] Preparing to replace portmap 5-4ubuntu3 (using .../portmap_5-4ubuntu4_i386.deb) ... [08:20] * Stopping portmap daemon... [ ok ] [08:20] Unpacking replacement portmap ... [08:20] Setting up portmap (5-4ubuntu4) ... [08:20] Installing new version of config file /etc/init.d/portmap ... [08:20] sivang: http://wiki.ubuntulinux.org/FrequentlyAskedQuestions, point 8 [08:20] * Starting portmap daemon... [ ok ] [08:20] * Restoring old RPC service information... [ ok ] [08:20] tcp 0 0 *:sunrpc *:* LISTEN [08:22] oh, my /etc/default/portmap is wrong [08:22] is the one in the package correct [08:22] ? [08:22] pitti : there a bug# to attach to? [08:22] sivang: just mail the stuff to me [08:22] no, it is totally buggy [08:22] pitti : sure, no probe [08:22] mdz: the one in the pkg has ARG="-i 127.0.0.1" [08:23] really? [08:23] there is one out there with OPTIONS= in it [08:23] # By default, listen only on the loopback interface [08:23] ARGS="-i 127.0.0.1" [08:23] rather than ARGS [08:23] are you sure that is the one in the archive? [08:23] mdz: this is your package [08:23] pitti : hal sees it, btw [08:23] fabbione: the one I uploaded to the archive was wrong [08:23] sivang: that's good [08:24] I am fucking up left and right, need to sleep [08:24] mdz: ok go to sleep [08:24] fabbione: will you fix it? [08:24] mdz: i will upload asap the portmap with ARGS [08:24] great, thanks [08:24] good night all [08:24] mdz: libtiff is ready for upload too [08:24] mdz: regression tests are ok [08:24] night mdz [08:24] mdz: and the new package spawns proper errors [08:25] mdz: new hal package is on http://people.ubuntulinux.org/~pitti/utopia/ [08:25] mdz: shall I anounce it separately or do you want to copy it in your repo? [08:26] "Programmers don't die! they just GOSUB without RETURN" [08:26] 11 upgraded, 601 newly installed, 0 to remove and 97 not upgraded. [08:26] Need to get 415MB of archives. [08:26] that should be a nice test. [08:27] mdz: night [08:27] mdz: sleep well, dude [08:27] pitti : sent [08:30] mdz: fixed portmap and fixed tiff are up [08:30] mdz: good night dude :-)) [08:32] my DSL hates me for downloaing the kernel stuff, and my laptop will hate me for compiling it [08:32] the moving company is here [08:32] brb [08:36] pitti : any luck? [08:39] mdz: I'm happy. [08:41] pitti : at first glance the code looks ok, though it lacks sufficient documentation to my taste [08:42] sivang: which code? [08:43] icky, debootstrap + apt-get install ubuntu-desktop breaks: [08:43] Unpacking lsb (from .../archives/lsb_2.0-1_i386.deb) ... [08:43] dpkg: error processing /var/cache/apt/archives/lsb_2.0-1_i386.deb (--unpack): [08:43] trying to overwrite `/lib/lsb/init-functions', which is also in package lsb-base [08:43] dpkg-deb: subprocess paste killed by signal (Broken pipe) [08:43] pitti : network-admin [08:43] sivang: the code looks good???? It's a mess! [08:43] Mithrandir: that's bad! [08:44] sivang: well, not a mess, but a little more synchronous wouldn't have hurt [08:44] how come we didn't see that with cd install? [08:44] fabbione: I don't know [08:44] pitti : well, I did find it _very_ hard to see where it really sends requests to the backends, but then again one cannot disregard my newbieness on this :) [08:44] Mithrandir: ok. let me see if it happens here too [08:45] Mithrandir: did you use any special command to bootstrap? [08:46] or just debootstrap -> chroot apt-get install ? [08:46] pitti : I have been trying to get a hold of where this _actually_ happens. in regard to #2221 [08:46] fabbione: yes, just debootstrap + apt-get install [08:46] ok [08:47] (naturally creating a sources.list between) [08:47] yeah :-) [08:48] probably my fault, I had ftp.se in sources.list; will have to try again [08:48] anyway, need to get some sleep. nighterz all [08:48] i am checking now [08:48] almost completed the bootstrap [08:49] apt-get install now [08:55] *art-stop-daemon: nothing in /proc - not mounted? [fail] [08:55] there is something wrong with the error message [08:55] it's lacking a char [08:55] and a space [08:57] ftp.ubuntu.com is slooow [08:57] only 1MByte/sec [08:57] heh [08:57] Mithrandir. [08:57] wonder how much auckland is pushing out [08:57] it looks ok here [08:58] fabbione: ook, good, checking myself as well [08:58] ok [09:01] Mithrandir: i confirm.. it's ok here [09:02] ok we are still at 9 RC bugs [09:03] that's 9 too many :/ [09:03] 1854 is fixed, just pending upload [09:04] Mithrandir: if it works i guess you can go with it [09:06] nah, mdz has the fixed package in rc-bugs-fixes (or what it's called), so he [09:06] 's handling it. === Mithrandir runs off for class [09:09] Mithrandir: later mate [09:14] BUGGER. [09:14] kernel ftbfs. [09:14] cp -p debian/patches/00list-17 debian/patches/00list [09:14] cp: cannot stat `debian/patches/00list-17': No such file or directory [09:14] make[1] : *** [debian/monolith/patch-2.6.8.1-17] Error 1 [09:14] make[1] : Leaving directory `/home/daniels/canonical/kernel/linux-source-2.6.8.1-2.6.8.1' [09:23] mdz: we will almost certainly replace fam for hoary. [09:24] jdub: with gamin? [09:24] yeah [09:24] <|trey|> daniels: yes, says that on the HoaryHedgehog release plans :) === |trey| [~trey@ip68-230-75-109.ph.ph.cox.net] has joined #ubuntu-devel [09:35] ah right, figured out the correlation === plovs_work [~plovs@195.13.248.82] has joined #ubuntu-devel [09:39] ooook [09:39] since there are only RC bugs left and assigned [09:39] it's time to start to do something more interesting ;) [09:45] wiki still does not work, //htdocs/warthogs/css/common.css was not found on this server [09:45] moinmoin not correctly configured? [09:50] might be missing: Alias /wiki/ "/usr/local/share/moin/htdocs/" in /etc/httpd/httpd.conf === seb128 [~seb128@ANancy-151-1-6-116.w83-194.abo.wanadoo.fr] has joined #ubuntu-devel [09:51] morning [09:52] hey seb128 ! [09:53] hello fabbione [10:03] Hi seb128! [10:03] hello pitti [10:11] daniels: nobody other than Danzer commented on the way i proposed to split the includes [10:11] daniels: i am definetely going to divide them [10:11] i need to give it more thought [10:12] daniels: ok. for me it's either solution 3 or solution 4 [10:12] daniels: given that we will hit that problem [10:12] daniels: we don't know yet [10:13] i like the way I split it up :) [10:17] daniels: i don't trust the way you did :P [10:29] fabbione: heh, why not? [10:32] daniels: just shitting on you :) [10:33] haha === plovs_work [~plovs@195.13.248.82] has joined #ubuntu-devel [10:33] daniels: did you book the flight and the hotel? [10:34] i kinda need to know when you will arrive to come and pick u up at the airport [10:34] fabbione: flight booked, haven't found a hotel yet [10:34] daniels: ok. do you want me to search one for you? [10:34] fabbione: i arrive on the 31st of nov, probably late [10:34] 31st of Nov?? [10:34] fabbione: if you could find one, that would be great. it's kinda hard to find one just in english :) [10:34] er [10:35] time doesn't really matter [10:35] october! [10:35] ok [10:35] sure i will take a look [10:35] do you have any special requirements? [10:35] not really, no [10:35] internet access probably preferred [10:36] daniels: i will see what i can find [10:36] budget/day ? [10:36] there seems to be a strong correlation between having an english-language website and being bloody expensive ;) [10:36] thanks a lot === mvo [~egon@suprimo-215.ping.de] has joined #ubuntu-devel [10:36] oh everybody or almost speaks english [10:36] so that's not a problem [10:36] dunno, probably below EUR100 [10:36] ask the people approving the budget ;) [10:37] but yeah [10:37] daniels: ok [10:37] i don't know how cheap/expensive hotels are there though [10:37] neither do i :) [10:38] heh [10:39] i just realized that i didn't read d-d ml for like 3 weeks and there was not even an interesting message [10:39] i should consider to unsub from it [10:40] heh [10:40] i haven't been subscribed for ages now [10:40] i'm not missing it === rburton [~ross@82-44-126-41.cable.ubr03.croy.blueyonder.co.uk] has joined #ubuntu-devel [10:51] hey rburton [10:51] hi fabbione [10:53] hum, I've some files corruption and crashes, seems to be a memory problem :/ [10:53] bbl === doko [doko@dsl-084-057-050-163.arcor-ip.net] has joined #ubuntu-devel === Sledge__ [~steve@80.46.37.4] has joined #ubuntu-devel [10:59] morning === |trey| [~trey@ip68-230-75-109.ph.ph.cox.net] has joined #ubuntu-devel [11:07] moo === pitti [~martin@195.227.105.180] has joined #ubuntu-devel === seb128 [~seb128@ANancy-151-1-6-116.w83-194.abo.wanadoo.fr] has joined #ubuntu-devel [11:15] marnin' [11:15] i feel so much more human after 13hours sleep [11:16] thom: word [11:16] thom: ready to get back into the firefox groove? ;) [11:16] i busted into 12 lsat night [11:18] i was checking that apache2 patch last night and it took me 5 attempts to type RENEGOTIATION [11:18] which was not ideal [11:18] 9 hours sleep, not what I needed but it'll do for now [11:19] thom: hey man, I typoed VERBOSITY into VEROBSITY [11:19] thom: and I counted eight attempts to type my GnuPG password before I nailed it [11:19] heh === fabbione kills more needs of X11 ;) [11:39] daniels: xconfig is now perfectly able to bootstrap without the need of external includes [11:40] cool [11:40] not even the evil ../../exports/include [11:42] i hate ../../exports/include [11:43] hm. lots of people suggesting forums instead of lists, but IME, forums scale far, far, far, far, far worse than lists. [11:44] forums suck donkey balls [11:44] personally i'd create a forum but say that ubuntu support is primarily on the lists [11:45] then i'd get the IPs of people who use the forum, find out where they live, and send them hate mail === rburton appears to be in a bad mood today [11:45] daniels: why are you doing new kernel packages to fix the thermal/fan problem? [11:45] thom: new initrd [11:45] when the trivial fix is a two line flip of mkinitrd [11:46] uh. doesn't the initrd get built on the fly? [11:46] daniels: uh, you know the initrd is generated at install time [11:46] holdasec, gotta clean up, apparently. (/me is sick of being dragged away from work) [11:46] Kamion: indeed it is [11:46] run the installer with a hacked initrd-tools dropped in === mjg59 is entertained by the response to the artwork [11:48] hi Matthew [11:49] xmkmf [11:49] imake -DUseInstalled -I/usr/X11R6/lib/X11/config [11:49] xmkmf [11:49] imake -DUseInstalled -I/usr/share/xorg/config [11:49] Mithrandir: see what i mean ;) [11:49] fabbione: huh? wrt nvidia-settings? [11:50] as long as the forums/lists are bidirectional gated [11:50] and only for the users list [11:50] Mithrandir: yes [11:50] and doesn't break threads, include the author's name and such. [11:50] fabbione: what you mean I need to run both xmkmf and imake? [11:50] and does in reply to right [11:51] Mithrandir: no no... when i asked you to switch from a generated Makefile to use xmkmf [11:51] Mithrandir: one upload and it's done automatically with the new settings [11:52] oh, ok. === rburton reads ubuntu-users archive, rofl at keybuk's black death comment [11:57] thom: cool, I suck. [11:58] Kamion: should I have a bash at a CD image with a hacked initrd-tools and ask the ACPI bongsippers to drop it in? [11:58] thom: ubuntuforums.org has a bidirectional list<->forums gate [11:58] daniels: sounds plausible to me; new package, edit dists/warty/main/binary-i386/{Packages,Packages.gz} and dists/warty/Release [11:58] daniels: what? just put the initrd-tools somewhere and then do "apt-get --reinstall linux-image-foo" [11:59] Kamion: phat [11:59] the rune to make the ISO is "mkisofs -r -V 'Ubuntu 4.10 i386 Bin-1' -o warty-install-i386-hacked.iso -cache-inodes -J -l -b isolinux/isolinux.bin -c isolinux/boot.cat -no-emul-boot -boot-load-size 4 -boot-info-table wherever-your-new-tree-is" [11:59] thom: yeah, because we need a higher barrier to entry when convincing people to test :P [11:59] --reinstall install [11:59] Kamion: i am in awe of your d-i skillz [12:00] daniels: WHATevah. but the two main whiners are spiv and keybuk, so they're gonna do what i'm suggesting i'd bet [12:00] thom: fair cpo [12:00] also, fair cop [12:00] "in addition ..." [12:00] "AMONG our spelling errors" [12:00] heh [12:02] daniels: Intel's download for the extreme graphics drivers seems to include support for putting different displays on the different pipes - do you know if this is supported by the free drivers? [12:04] mjg59: i don't believe so. is intel's download sourceful? [12:06] Nope [12:06] I'd always assumed that there stuff was just rebuilt source, but it doesn't sound like it [12:06] ftp://aiedownload.intel.com/df-support/7485/ENG/readme.txt and http://downloadfinder.intel.com/scripts-df/filter_results.asp?strOSs=39&strTypes=DRV%2CARC&ProductID=922&OSFullName=Linux*&submit=Go%21 [12:10] ALTERNATIVES="b2m ctags ebrowse emacsclient etags grep-changelog rcs-checkin " [12:10] for i in `echo ${ALTERNATIVES} | grep -v emacs` [12:10] LICENSE. You may copy the Software onto a single computer for [12:10] your personal, noncommercial use, and you may make one back-up [12:10] copy of the Software, subject to these conditions: [12:10] spot the OBVIOUS ERROR [12:10] mjg59: dude. [12:10] Kamion: empty ${ALTERNATIVES} [12:11] daniels: nope [12:11] Kamion: well, it will be after grep has finished with it [12:11] daniels: yep :) [12:12] obviously was *meant* to remove all the words in $ALTERNATIVES that don't contain 'emacs', but ... [12:12] daniels: It's obviously derived from the same code [12:12] mjg59: i'm not overly keen on chancing it, at any rate [12:12] Kamion: yeah. oops [12:12] daniels: Oh, hang on - Xorg has the same code [12:13] Man. Now I need Xorg. [12:13] mjg59: (note that the i915 stuff was done by Tungsten, on contract to Intel, so Intel could get a licence) [12:13] mjg59: bahbahbah [12:13] New in 6.8.0 [12:14] yeah, the i915 merge was done about 3 months ago, maybe 4 [12:14] obviously it has phat i8xx features as well [12:14] hey, this could be a good thing for me, too [12:14] mjg59: if you want it you must have it! (in hoary) === daniels ramps it twenty spots up his priority list. [12:14] sabdfl: dude, Hoary will clean my teeth and make my coffee [12:15] phew, that's a relief. not the coffee. [12:15] Haha === mjg59 discovers that the secret to mad phat support is to get all the developers to own identical hardware to you [12:16] daniels: does it fix the crackful i8xx crash that we were seeing with xresprobe? :P [12:17] thom: you think this laptop's seen Xorg? [12:17] mjg59: (and mark) [12:17] sabdfl: heh :) === Kamion notes idly that we're releasing Array CD 1 in four days [12:18] according to HoaryHedgehog/ReleaseSchedule [12:18] Kamion: ... [12:18] Kamion: isn't hoary ... empty? [12:18] Kamion: yeah, i have the sneaking suspicion that might slip [12:18] I suppose this might depend slightly on hoary existing [12:19] Kamion: hey man, the suite's there, don't knock it [12:19] hrm. i really should buy a copy of Hackers and Painters [12:19] i could upload X to it with the change to make it not write a log message every time the screensaver comes on if you wanted it ;) [12:19] http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&rd=1&item=5130553701&ssPageName=STRK:MESE:IT - 100 AU$? [12:19] BONG. [12:20] (Not that I'm not tempted anyway) [12:20] mjg59: but I am. bastard. [12:20] Oh, I'm massively tempted [12:20] I've got an Amiga Unix machine [12:21] But I'm using dodgily obtained install media that's not entirely up to date (as much as any software that was killed in 1992 can ever be up to date, but still) [12:21] $30 postage? [12:21] From New Zealand [12:21] still [12:24] ha ha. [12:24] gcc ICEd during the kernel compile anyway [12:24] oh man. [12:25] <3 gcc [12:27] gcc [12:28] xn--d0ha04i gcc [12:29] Haha [12:29] (my setup is not utf-8-compliant, because screen hates me) [12:29] I was bored enough yesterday that I wrote an xchat script so my quotes are automatically made pretty [12:29] do you want to package X.Org instead? :P [12:29] No [12:29] oh my god. [12:29] Hahahahaha === Kamion installs idn so he can decode daniels' punycode [12:30] My christ, what have I started? [12:31] uh. *was* that punycode? [12:31] ah, idn -u manages it === Kinnison [~dsilvers@host81-153-126-219.range81-153.btcentralplus.com] has joined #ubuntu-devel [12:34] Morning [12:35] Kamion coerced me into joining here. [12:35] about time === Kinnison has said that he'll join the desktop team if (and only if) colin writes lucille [12:36] writing Lucille would be pretty rad [12:37] fsvo. rad [12:37] :-) [12:37] heh [12:39] <|trey|> What is Lucille? [12:39] <|trey|> (link?) [12:39] Never mind [12:39] <|trey|> :( === seb128_ [~seb128@ANancy-151-1-12-161.w83-194.abo.wanadoo.fr] has joined #ubuntu-devel [12:43] Kinnison: see the bug in cscvs native protocol ? === Kinnison [~dsilvers@host81-153-126-219.range81-153.btcentralplus.com] has left #ubuntu-devel [] [12:51] |trey|: lucille is the tool we are writing that will let us publish ubuntu packages more effectively [12:51] <|trey|> hmm :) [12:51] i don't think Kinnison was talking to you with the "never mind" [12:51] crosstalk :-) === |trey| [~trey@ip68-230-75-109.ph.ph.cox.net] has joined #ubuntu-devel === |trey| points at the last section of the Topic before RC bugs :) === ..[topic/#ubuntu-devel:thom] : Ubuntu development -- general discussion on #ubuntu | Happy Ubuntu Artwork week! | 9 RC bugs to go === Kamion tries to clean up his last-minute hacks on the cdimage machine [01:05] thom: could I have gpg on little, please? signed archives and all that ... [01:08] done [01:11] ta === doko [doko@dsl-084-057-017-069.arcor-ip.net] has joined #ubuntu-devel [01:32] interestingly, while there were way more torrented downloads of PPC for the preview, amd64 is about double for the RC [01:32] so this means amd64 users reinstall each time they download a new iso [01:41] duh.. do we want the daily d-i images to still go in? [01:42] I guess so, but we should add "disable warty daily d-i build" to the final's release TODO list [01:42] elmo: right [01:43] probably won't need it between now and release, but I haven't quite got a feel for what mdz is going to want me to change [01:58] i suspect it just means that someone in the amd64 community pointed out the rc [02:00] amd64 people are more likely to be early adopters as well, as it's a fairly new architecture, so you'll get tinkerers who, well, tinker with their machines [02:01] pr0n lovers [02:01] yeah, and the amd64 is so pr0n loving and heard about the artwork, so... [02:01] ;P [02:02] anybody know when we can expect to see gfx cards that do >8bit per channel? [02:02] so you could get true truecolor? [02:02] isn't truecolor 8bits per rg and b? [02:02] we really need more [02:03] you have 32 bits as well, since most computers are faster at working with 32 bit increments than 24 bits. [02:03] apparently the film industry uses specialised monitors and gfx cards that do 16 bit depth per channel [02:03] ok [02:03] a hoary question... [02:03] personally, I don't need more, but then, I'm not making movies. :) [02:04] could we bring the machine up sanely without networking? [02:04] and have services discover networks as they are brought up? [02:04] the pause during dhcp is... irritating [02:04] sabdfl: that is how NetworkManager (and by extension, fedora core 3) works [02:04] problem with ifplugd (or an equivalent) is broken hardware. [02:05] is this something we could aim for with hoary? [02:05] oh, you mean just running the dhcp client in the background? [02:05] together with dependency-based init? [02:05] sabdfl: networkmanager is a release goal [02:05] so yeah, we'll be looking at it [02:05] Mithrandir: i think that runs the risk that a service like apache comes up before dhcp has bound to an address [02:05] thom: great, thanks [02:05] sabdfl: doesn't matter if it listens to 0.0.0.0 [02:05] dependency based init would be nice too [02:06] Mithrandir: or 127.0.0.1 for that matter [02:06] sabdfl: for 127.0.0.1 it matters, since then it only listens on loopback, but if you listen on 0.0.0.0, it'll automatically listen on newly created interfaces. === pitti [~martin@195.227.105.180] has joined #ubuntu-devel === sivang [~pooh@80.179.66.20.forward.012.net.il] has joined #ubuntu-devel [02:16] is there anyway we could make the wallpaper be more bright? I wanna see the models better .. :) [02:17] pitti : any more LUMUX troubleshooting we can do to make it work? [02:17] *LUMIX [02:18] sivang: If it works on your sid kernel, then you could file the information as a bug against the Ubuntu kernel [02:18] pitti : ok, no probe. === pitti [~martin@195.227.105.180] has joined #ubuntu-devel === rabidbt [~rabidbt@66.45.74.16] has joined #ubuntu-devel === bob2 [rob@bob2.user] has joined #ubuntu-devel === Kamion [~cjwatson@host81-153-126-219.range81-153.btcentralplus.com] has joined #ubuntu-devel === daniels [daniel@fooishbar.org] has joined #ubuntu-devel === chrisa [~chris@nullcode.org] has joined #ubuntu-devel === mjg59 [mjg59@cavan.codon.org.uk] has joined #ubuntu-devel === amu [amu@bofh.debian.net] has joined #ubuntu-devel === m_tthew [matt@iorek.ice-nine.org] has joined #ubuntu-devel === tseng [~tseng@thegrebs.com] has joined #ubuntu-devel === thom [~thom@amnesiac.heapspace.net] has joined #ubuntu-devel === Kamion [~cjwatson@host81-153-126-219.range81-153.btcentralplus.com] has joined #ubuntu-devel === amu_ [amu@bofh.debian.net] has joined #ubuntu-devel === rabidbt [~rabidbt@66.45.74.16] has joined #ubuntu-devel [02:41] F S UID PID PPID C PRI NI ADDR SZ WCHAN STIME TTY TIME CMD [02:41] 0 D lamont 644 1 0 76 0 - 629 mempoo 00:05 ? 00:00:02 growi [02:41] hrmpf [02:41] is 6 hours a long time to be blocked on mempool? [02:42] g'day [02:42] g'day amu [02:42] I think I get to reboot.. [02:42] hey amu [02:43] seb128: how do i convince gnome that i have french installed? [02:43] i've reconfigured locales to add fr_FR* [02:44] hai lamont, fabbione [02:48] thom: what are you trying to do ? [02:48] thom: start a session in french ? just an app ? [02:49] i want to start a session in french [02:49] well, i don't want to. i need to :-) [02:49] thom: crazy person. [02:51] brb [02:51] seb128: if i set the language as french in gdm (running in xnest) gnome tells me that no such language is installed [02:52] Language fr_FR does not exist, using System default [02:52] but it lets me choose it in gdm [02:54] thom: dpkg-reconfigure locales and select fr_FR@euro or fr_FR.UTF-8 ? === netjoined: irc.freenode.net -> sendak.freenode.net === pitti [~martin@195.227.105.180] has joined #ubuntu-devel === bob2 [rob@bob2.user] has joined #ubuntu-devel === daniels [daniel@fooishbar.org] has joined #ubuntu-devel === chrisa [~chris@nullcode.org] has joined #ubuntu-devel === mjg59 [mjg59@cavan.codon.org.uk] has joined #ubuntu-devel === m_tthew [matt@iorek.ice-nine.org] has joined #ubuntu-devel === tseng [~tseng@thegrebs.com] has joined #ubuntu-devel === thom [~thom@amnesiac.heapspace.net] has joined #ubuntu-devel [02:58] Guys, why do you continuously disappear and appear again? [02:59] 14:57 -!- Netsplit burroughs.freenode.net <-> irc.freenode.net quits: rabidbt [02:59] 14:57 -!- Netsplit burroughs.freenode.net <-> irc.freenode.net quits: pitti, [02:59] netsplits. [02:59] freenode sucks, film at 11 [02:59] oh, scary === tseng [~tseng@thegrebs.com] has joined #ubuntu-devel [03:00] ah, from your POV rabidbt and pitti disappeared? [03:00] thom: so, I was saying ... you have the fr_FR@euro locale generated with dpkg-reconfigure locales ? [03:00] yep [03:01] i generated all the fr_FR locales after it didn't work the first time [03:01] pitti: yes, and a fair number more. [03:02] and all of them give the same error [03:02] brand new user, too === rabidbt [~rabidbt@66.45.74.16] has joined #ubuntu-devel [03:02] pitti: IRC is client <-> server (<-> server)* <-> client [03:02] pitti: if one of the links in the middle fails, you get netsplits. [03:02] Kamion: thanks for that explanation! [03:03] yeah, and the IRC servers are organized in a tree, so you can have small splits if just a server splits off, or huge splits if the root loses some of its links. [03:03] https://bugzilla.ubuntu.com/show_bug.cgi?id=2294 please review patch :-) [03:04] thom: use if [ "$results" ] rather than if [ ! -z ... [03:05] or -n if you must [03:05] want to change as little as possible [03:05] and decide on [ ] or test, not both. :) [03:05] yeah [03:05] i'll clean up post warty [03:05] but i want that patch in warty [03:06] looks good to me === Mithrandir grabs his "APPROVED" stamp and stamps thom, err, the patch. [03:06] heh [03:10] seb128: note that i did this post install, does gnome do anything locale related at install time that i'd need to kick? [03:13] no [03:14] I've already added some locales to reproduce bugs [03:14] just dpkg-reconfigure, pick the locale [03:14] and then select it in the gdm menu [03:14] ok, it could be an x nest [03:14] issue === thom logs out fully [03:14] brb [03:18] right [03:18] strange that reportbug sends mail to ubuntu-users [03:18] as it turns out, gnome can't detect languages until you restart gdm entirely. ho-hum [03:18] rburton: nothing else we can really do with it [03:18] i guess not [03:19] anyone would think that we didn't have a bugzilla hacker employed to write us an xml-rpc version, but anyway [03:20] hehe [03:24] seb128: OOI, if you install mozilla-firefox-locale-fr, does it give you a french firefox from the menu entry? [03:24] i'm trying to reproduce #2359 === ..[topic/#ubuntu-devel:thom] : Ubuntu development -- general discussion on #ubuntu | Happy Ubuntu Artwork week! | 10 RC bugs to go [03:25] thom: lemme test, I use epiphany [03:26] thanks [03:26] and yes, i know :-) [03:27] with firefox 0.99+1.0PR.1+revertedto0.9.3-0ubuntu1 that's ok [03:28] I'm upgrading it to see if that's still ok with the current version [03:34] thom: yes, works fine with the current version too [03:34] thom: do you have the problem on your box ? [03:38] nope [03:47] if i wanted to install ubuntu on a machine with a cd drive and can't be arsed with netboots, is there an easy way? [03:47] i.e. a floppy bootstrap [03:51] sorry, CD and netboot are all we do at the moment; USB will probably happen shortly after warty, too [03:51] dunno about floppies yet, they're quite a pain to support well [03:52] k [03:52] Mithrandir: [X] SATISFACTORY [ ] UNSATISFACTORY [03:53] Kamion: any quick guides on netboot? not done that before... [03:53] or would it be quicker to find a cd-rom drive :) [03:58] rburton: there's a pxeboot.tar.gz which is supposed to set it all up in one shot for you; never actually tried it myself, I must admit [04:02] Kamion: where is that file? === sivang [~pooh@80.179.66.20.forward.012.net.il] has joined #ubuntu-devel === Keybuk [scott@descent.netsplit.com] has joined #ubuntu-devel [04:04] rburton: /dists/warty/main/installer-i386/current/images/netboot/ on the mirror of your choice [04:04] new iso mirror with a 1,7TByte connection http://source.rfc822.org/pub/mirror/releases.ubuntu.com/ [04:04] amu: could you put that on http://wiki.ubuntu.com/Archive? [04:05] Kamion: thanks [04:06] Kamion: yap [04:10] daniels: ping? [04:15] the pxeboot works great [04:15] just follow the steps on the debian wiki [04:15] ah debian wiki [04:16] good plan [04:16] rburton: i installed my X40 that way ;-) [04:16] wiki.debian.net, right? [04:16] yeah === rburton prods web server [04:16] you pretty much untar pxeboot.tar.gz in your tftproot [04:17] yeah, it just useful to know the runes for the dhcp setup [04:17] it's called netboot.tar.gz in Debian now I think [04:17] then stick 'next-server ip-of-tftp-server;' and 'filename "pxelinux.0";' in the dhcp host bit for the machine you want to boot [04:17] righto [04:17] so its a dhcpd + tftpd thing [04:18] yup [04:18] cheers [04:20] host elite_wired { [04:20] hardware ethernet 00:b0:d0:a4:40:9a; [04:20] fixed-address elite.netsplit.com; [04:20] next-server descent.netsplit.com; [04:20] filename "pxelinux.0"; [04:20] } [04:21] Keybuk: cheers [04:21] assuming you don't already have dhcp setup, you'll probably need a subnet { } gubbins as well [04:22] we've got dhcp already === seb128 [~seb128@ANancy-151-1-12-161.w83-194.abo.wanadoo.fr] has joined #ubuntu-devel [04:25] heh, whoops [04:25] almost as bad as rebooting the wrong machine -- was logged into my desktop upstairs (sitting on my laptop at the lounge), it's 12:30, and accidentally put Aphex Twin on at earsplitting volume [04:25] took me a second to notice because it was normal volume for my earphones [04:26] kinda twigged when I could feel the bass vibrations though :P [04:26] see, I bet your family didn't appreciate the funny side of being woke up to screams of "I WAAAAAAAAAAAAAAAAAAAAAAAAAANT YOUR SOUL" [04:26] 'how to win friends and impress your neighbours' [04:26] bob2: luckily it was Girl/Boy Song, not o/~ i would like some milk from the milkman's wife's tits [04:26] daniels: "man, these headphones *rock*" [04:26] rburton: heh [04:26] 'GREAT SOUND!' [04:26] ah === carlos [~carlos@69.Red-80-33-181.pooles.rima-tde.net] has joined #ubuntu-devel [04:27] daniels: the milkman's wife is totally on the top of his... === sivang away [04:28] bob2: HAHA [04:28] that adds a whole new perspective to that whole new film [04:30] I'm still emotionally scarred by it [04:30] ya know, I've just realised -- our desktop-icon-less desktop is perfect [04:30] you can click on it and type "captain" without any fiddling [04:30] captain? [04:31] Kamion: just do it [04:31] captain? [04:31] doesn't it give you an easter egg about campd being lonely? [04:31] daniels: u suck [04:31] yeah [04:31] uh, that would require me to go upstairs :) [04:31] oh dear [04:32] how many calls does he get? [04:32] a few :) [04:32] yeah [04:32] luckily he's a relatively good-humoured guy [04:32] don't you dare [04:32] even if he has thrown his laptop through a wall once [04:32] sabdfl: Applications -> Run Application ... "free the fish" is slightly less amusing [04:33] oh dear, it's wanda [04:33] though it gets *really* annoying after a while [04:33] if you click on it it goes the hell away [04:33] mercifully [04:33] daniels: and comes back later [04:33] you have to kill gnome-panel to stop it [04:33] dude [04:33] SWEET MOTHER OF GOD [04:33] 6 days out from release and now you have fish on EVERYONE'S DESKTOP [04:33] arrgh! [04:33] make it go away! [04:34] if that came back at the wrong moment, I would seriously put an FUCK OFF FISH axe through my crt [04:34] sabdfl: Applications -> Run Application ... "killall gnome-panel" :) [04:34] daniels: it *used* to be on all the time, in GNOME 1.4 era; but would only appear at *really* rare times [04:34] it's only come back once and it's even more infurating than the typing monitor spawning at the exact wrong moment [04:34] Keybuk: so much ARGH bong [04:34] one of those random times was while Nat was doing a big presentation === daniels executes his panel. [04:34] Keybuk: copped it [04:35] Keybuk: hey, btw -- new deb up [04:35] 'and for my next trick, I'll actually swap the load order!' [04:35] daniels: you learned to use vi this time? [04:36] Keybuk: i blame a mixture of being distracted by a visitor, the X40's keyboard, and, um, RSI [04:36] RSI ... not JD ? [04:36] or something [04:36] ? [04:36] Jack Daniels [04:36] ... [04:37] so far you've connected repetitive strain injury, a UK Debian developer, and some really bad whisky [04:37] i just finished the bottle of Red Eye -- it hasn't kicked in, so make the connection for me [04:37] daniels: now I just have to link them to the drawing room and I WIN!!! [04:38] Keybuk: bah, rubbishing me in multiple media [04:38] i have to look stupid on IRC, the lists, and Bugzilla :P [04:38] you do exceptionally well at it, I must say :o) [04:39] though that new initrd-tools is much better [04:39] thankyou dear [04:39] nice to know I'm appreciated for my talents (specifically, court jester) [04:40] hey, that's my job! === bob2 is happy to hand that mantle over to daniels [04:40] a double-act! [04:41] bob2: no, you're just the sideshow [04:41] duh. [04:41] i'm the main attraction :) [04:42] hm, having a NEEDINFOer would be nice [04:42] e.g. I NEEDINFO from this person, so it shows on their list also [04:42] guilt-tripping Keybuk from afar [04:42] Dr Daniel and his Amazingly Astounding X Packages. [04:42] daniels: hrm? [04:43] Keybuk: you need to test 2341, duh :) [04:44] daniels: just have, fans are on [04:44] bob2: Marvel as be breaks mdz's machine! Be astounded as he does yet another bloody XKB symlink! Gasp as mdz looks at him threateningly! [04:44] well, three of them are on [04:44] Keybuk: phat [04:44] three? that's halfway to a Hoverbook :) [04:44] 15:32 < iDunno> thom: who's responsible for the 'artwork' for gdm and the gnome startup? [04:44] 15:36 < Dave> erm [04:44] *sigh* [04:44] 15:36 < thom> da boss [04:44] 15:37 < Dave> thom: its horrid [04:44] daniels@brenna:~/canonical/initrd-tools% ls /proc/acpi/fan [04:44] daniels@brenna:~/canonical/initrd-tools% [04:45] your x40 is called brenna? [04:45] although there's definitely one going; I just put my ear up to it and it blew a crapload of air into my ear. go figure. [04:45] the x40's fan is hardware controlled afaik, not acpi controlled [04:46] bob2: brenna is tycho's girlfriend. i've been through tycho (fooishbar.org), gabe (fd.o), twisp (pda), catsby (old laptop ... maybe I'll change to catsby when I reinstall tonight), brenna (x40), kara (unused -- gabe's wife), div (a Myth box I installed for a mate) [04:46] thom: that kinda sucks [04:46] oh, and I called the laptop I had for LCA fruitfucker2000 === Keybuk was playing the other day with having different trip_points for battery and AC [04:46] Keybuk: not really; it just blows out the side anyway [04:47] Keybuk: so it's not like the OmniBook, where it FROZE YOUR FRIGGING LEG [04:47] has anyone tested lamont's live CD, then? === daniels took to having a small script to just disable it every five seconds, as its only use was to make your right leg so cold it got really, really sore. [04:47] daniels: yeah, but fans use power, so it's nice to be able to control them yourself [04:48] Keybuk: ah, I don't worry about power consumption, you see [04:48] for FAN in /proc/acpi/fan/*; do echo 0 > $FAN; done ... cold leg [04:48] for FAN in /proc/acpi/fan/*; do echo 3 > $FAN; done ... cold leg [04:48] uh, that second should be warm leg :p [04:48] yeah [04:48] daniels: ahhhh [04:49] but my left leg was always boiling (sometimes to the point of sweating profusely) whatever I did [04:49] I'm scared my ibook will set fire to my desk in the summer [04:49] the fan blew straight on to my right leg (the idea was to have it elevated, and run under the laptop to ventilate the left side), so it didn't actually cool the left side. at all. [04:49] hammering the disk makes it too hot to have on your lap [04:49] Keybuk: given that the x40 appears to have a single fan and the hardware appears to get it right, i'm not too fussed [04:49] bob2: same with a powerbook [04:49] the x40 also very, very rarely gets hot [04:49] in winter you can compile kernels to keep yourself warm [04:50] the only time I've ever actually noticed it -- and I wasn't actually uncomfortable -- was after it had been building X on AC power (thus un-SpeedStepped) for a couple of hours [04:50] Kamion: ever owned an ia64? ;) [04:50] nope :) [04:50] Kamion: hehe, the powersupply is good for chilly toes [04:50] the CD drive generates a fair bit of heat, actually [04:50] taking the CD out is generally useful [04:52] bob2: yeah, I used to love that about the OmniBook [04:52] bob2: until one day where I burnt the underside of my toes and lost some skin [04:52] after that, I'd just go find some socks [04:52] ARSE [04:53] socks for your arse? [04:53] daniels is "unique". [04:53] Kamion: just realised that as my music is all track%d.cdda.ogg when ripped now, not %d-audiotrack.ogg, my small tracknumber script (a horrible sed hack) has been putting on track numbers of 'tr' for a couple of CDs [04:53] and there are no good tag editors [04:54] the only one I could find was catbus or something, and it's *abysmal*. [04:54] alias tn='for i in *.ogg; do echo TRACKNUMBER=`echo $i | sed -e "s/^\(..\).*$/\1/; s/^0//;"` | vorbiscomment -a $i; done' [04:54] daniels: the new musicbrainz library has an accoustic fingerprint tagger [04:54] so in theory it will tag without asking you [04:56] daniels: easytag is not bad IMHO [04:56] daniels: I have used it for quite a while [04:56] rburton: nice [04:57] daniels: it can rename files after tags, or the other way round; quite flexible [04:57] rburton: the only problem I have with SJ is that (aside from never having seen musicbrainz support ever activate) is that I rip a lot of compilations [04:57] and for that I want %d - Artist Name - Track Name.ogg, whereas I just want %d - Track Name.ogg for single-artist stuff [04:57] so I have my own Python script that DTRT according to a very small file I write [04:58] daniels: hrm [04:58] rburton: that and quality support for oggenc :P [04:58] i want to say "use a decent media player" so you don't see the filenames [04:58] daniels: sj in cvs has profile support [04:58] interesting problem though [05:03] rburton: oh? [05:03] rburton: yeah, I keep meaning to, but mainly because of the compilation issue, end up back at the command line [05:03] rburton: rhythmbox isn't quite usable on 1024 -- just not enough screen real estate. i use it on 1600 tho. [05:04] daniels: tried muine? [05:04] rburton: yeah, rather liked it, but just didn't scale [05:05] i guess the album view doesn't work too well with a huge album list [05:06] hm, I kept meaning to true muine [05:06] its not bad [05:06] I'll just wait until I have enough time to write my twisted music playing thing [05:07] rburton: although I did like how it was pretty smart about compilations [05:07] ok, can other people please try the new initrd-tools to ensure they get no regressions? [05:07] i'll be trying it here when I reinstall, but just to make sure [05:21] http://ubuntuforums.org/viewtopic.php?t=442 [05:21] however that will be badly skewed by the (purely anecdotally observed) connection between Gentoo users and web forums [05:22] hm, interesting [05:22] isn't Mepis Yet Another Sid Snapshot LiveCD Thingy? [05:23] probably [05:23] hmmm [05:23] they tend to be non-free with their additions though, AFAIK === lamont [~lamont@mix.mmjgroup.com] has joined #ubuntu-devel [05:25] mdz: permission to upload initrd-tools? only change is the reversal of fan/thermal load order. [05:26] lamont: thanks for getting kismet going === seb128 [~seb128@ANancy-151-1-12-161.w83-194.abo.wanadoo.fr] has joined #ubuntu-devel [05:28] azeem: bong :\ [05:28] bob2: np. amd64 is still missing, though. [05:36] cdrdao isn't part of the base install? === bob2 watches daniels get trolled about X [05:38] lamont: hah, I'll bitch again if/when I justify one of them :) [05:38] bob2: when you do bitch, send me the patch to snacc to make it run on 64-bit, and I'll be happy. :-) [05:39] then again, there might already be a patch for that. hrm. [05:39] heheh [05:42] arse, gtk themes are compiled against gtk-x11 etc. [05:44] what does it mean when growisofs goes into an uninteruptible sleep on 'mempoo' and never comes back, I wonder. [05:45] other than 'time to finish the upgrade to RC'... [05:48] Kamion: how did you invoke alextreme yesterday? email? [05:48] lamont: yes [05:48] lamont: took an hour or two IIRC [05:48] ok. I'll do that after I finish rebooting.. [06:00] oh man. people seem to actually own ISA sound cards. [06:09] blink [06:09] an upgrade, just brought up a nautlius window [06:09] elmo: hal restarted. thought we fixed that bug. [06:09] elmo: does sudo /etc/init.d/dbus-1 restart, trigger it? [06:10] gah. why can't I compile gtk with multiple gdk frontends at once, eh? [06:10] daniels: yeah, dbus-1 restart triggers it here [06:10] I just got two spanking new windows [06:11] man, if there was ever an example of 'don't put off till tomorrow what you can rewrite today', it's apt-cacher. sweet jesus. [06:11] hack city. [06:11] Keybuk: bah. what version of hal? [06:11] 0.2.98-1ubuntu8 [06:11] 0.22-1ubuntu2 of dbus-1 [06:11] dbus-1's version is unimportant, it's just that it runs hal from event.d :) [06:12] daniels: if you do dbus-1 restart on your machine, do you get a / window ? [06:12] Keybuk: not a / window, no [06:12] don't think so [06:13] doesn't happen for me with ubuntu8 [06:13] This was not a hal bug, but a g-v-m one [06:13] it only happens for me once, the first time [06:13] pitti: oh? [06:14] pitti: oh right, yeah === daniels takes another hit. [06:14] Keybuk: g-v-m version? [06:14] https://bugzilla.ubuntulinux.org/show_bug.cgi?id=2169 [06:14] gah, apparently 'cleaning up' in this house means 'move Daniel's thumbdrive somewhere he can't find it'; that's like the fourth time this week [06:14] 1.0.2-0ubuntu5 of g-v-m [06:14] bong [06:15] oh [06:15] you've been running your session for a while :) [06:15] daniels, Keybuk: actually this should be fixed in ubuntu5 [06:15] killall gnome-volume-manager && gnome-volume-manager &| [06:15] does it still occur? [06:15] pitti: well, it just occured now [06:15] Keybuk: damn, g-v-m now should know about device states... [06:15] Keybuk: yes, but is the *running* g-v-m (i.e. the one started when your session was) ubuntu5? [06:15] daniels: just checking that now [06:18] I'm trying to remember how to get the exact start time of a process [06:18] 09:15 < daniels> killall gnome-volume-manager && gnome-volume-manager &| [06:18] unless you're really attached to your current g-v-m instance :P [06:19] one is as good as the other [06:19] it looks to me like I'm running ubuntu5 now [06:20] descent scott% killall gnome-volume-manager && gnome-volume-manager &| [06:20] descent scott% umount: only root can unmount /dev/hda1 from /boot [06:20] umount: only root can unmount /dev/hda3 from / [06:21] after killing, it seems ok, I can restart dbus *shrug* [06:22] right, so it *is* fixed [06:22] elmo: does that work for you too? [06:23] (and this, ladies and gentlemen, is why Windows makes you reboot after any install :p) [06:23] heh === Keybuk tries that to see whether it happens again [06:25] daniels: restarting dbus doesn't make it happen again for me [06:26] elmo: phat, thanks [06:27] no I mean even without the killal stuff :) [06:28] oh === Keybuk [scott@descent.netsplit.com] has joined #ubuntu-devel [06:28] heh, wack [06:28] yeah, seems ok now [06:29] I love how with the same number of processes running, the free memory is much higher after a reboot [06:33] and the free disk space has gone up as well, shows this thing really did need a reboot [06:34] heh === Mitario [~michiel@sikkes.xs4all.nl] has joined #ubuntu-devel === lamont [~lamont@mix.mmjgroup.com] has joined #ubuntu-devel [06:50] mdz, seb128: did we get gnome-applets 2.8.1 or 2.8.1.1 into Ubuntu? [06:50] the latter looks like it fixes a pretty silly crash [06:50] so gnome-background-proerties should do something after it's windo is created, right? (like add buttons, fill in the big white box I see, etc...) [06:51] much happier when root runs it... [06:52] Keybuk: 2.8.1 [06:53] Keybuk: 2.8.1.1 has been released yesterday [06:55] * Fix possible crasher in gweather preferences (Mark McLoughlin) [06:55] * Fix error in Italian translation that caused gweather to not parse the location list (Kjarten Maraas) [06:55] to me that seems reasonable to let in *shrug* up to mdz though [06:56] I've 2 other important fixes too, but mdz said it's too late for changes yesterday [06:56] post-RC is not the time to introduce fixes for non-RC bugs... [06:56] a fix for gok which just crashes when you run it [06:57] and a fix for a data corruption with the ftp method (and perhaps the http one too) in nautilus [06:57] do we want italians to be able to use gweather? :p [06:58] oh, that's probably the error that I've fixed in 2.8.1-0ubuntu2 [06:58] a <<...> instead <...> breaking the Locations.xml [06:58] it was breaking the locations for everybody [06:59] ah ok, so we got that in then === smurfix [~smurfix@smurfix.developer.debian] has joined #ubuntu-devel === maskie [~maskie@196-30-110-242.uudial.uunet.co.za] has joined #ubuntu-devel === __randy__ [~randy@sclab-25-433.sclab.clarkson.edu] has joined #ubuntu-devel [07:24] morning [07:24] mdz: 'morning. how was your sleep? [07:24] delicious [07:24] Hi mdz! [07:24] mdz: looks like you finally got some proper sleep [07:25] Mithrandir: still here? [07:26] mdz: morning === mdz stretches and yawns [07:26] is everyone happy with gzip? [07:27] I'd like to get that uploaded [07:27] mdz: it seems to be working fine for me [07:27] I beat on it fairly hard last night with apt and apt-extracttemplates [07:27] and Mithrandir confirmed it fixed his bug [07:29] mdz: it's not broken anything for me on up i386 [07:30] uploaded [07:31] mdz: permission to upload initrd-tools to fix 2341? [07:31] mdz: (there's now an interdiff on p.u.c/~daniels/thermal-fans/source/) [07:32] daniels: what's this about a kernel ftbfs? [07:33] diff looks exactly right [07:33] how much testing has it gotten? [07:33] mdz: that was me not realising you had to copy 00list-$(debrevision-1) to 00list-$(debrevision) [07:33] mdz: keybuk says it's OK, so do my own two machines (neither of which have fans, mind) === plovs_knoppix [~knoppix@195.13.248.82] has joined #ubuntu-devel [07:34] I made the change on my laptop last night [07:34] the fan module doesn't seem to work now, but I'm not entirely sure whether it worked before either [07:34] mdz: there's still some missing branding in the installation manual, specifically the installation howto (which says stuff like "Debian GNU/Linux warty") [07:34] mdz: ugh, bongtastic [07:34] mdz: can I re-upload debian-installer at some point to fix that? [07:35] can someone with a currently working fan setup (ACPI fan, and working when thermal is loaded before fan) please test the new initrd? download initrd-tools from p.u.c/~daniels/thermal-fans/all/, install it, and run apt-get install --reinstall linux-image-2.6.8.1-3- [07:36] Kamion: can it be the last d-i upload for Warty? [07:36] mdz: probably :-) [07:36] I don't see why it'd break anything. The installer already loads fan before thermal anyway, and before thom added those modprobes to the initrd the acpid init script loaded them and loads fan before thermal [07:36] mdz: would you agree the fix to laptop-detect is rc, and can i upload it? [07:36] I'll do a really good read-through of the installer this weekend [07:36] er, installer manual [07:36] Keybuk: yeah, but um, having testing would be good [07:36] Kamion: maybe for Hoary we should split out the manual into a separate source packagae so that the doc team can beat on it [07:37] i'd test it, but the x40 fans are hardware, not acpi [07:37] daniels: I think the same of T42 [07:37] thom: bug#? [07:37] mdz: the reason why it's there in the first place is that katie/wanna-build falls over if you try to upload a package containing only byhand components [07:37] (or something similar) [07:37] Kamion: sounds like a bug :-) [07:37] basically it doesn't get autobuilt that way [07:38] it could build an installation-guide.deb, no? [07:38] mdz: if /proc/acpi/fan is empty, you have no ACPI fans [07:38] mdz: 2294 [07:38] daniels: correct [07:38] mdz: er, it does [07:38] mdz: but the manual is arch-specific [07:38] Kamion: aren't debs not-byhand components? [07:38] mdz: yes, that's why one of them is there [07:38] mdz: by the way, SiS have fucked us. either we pick a lot of PCI-to-AGP bridges as video cards, or we pick a few video cards as PCI-to-AGP bridges. the former results in a broken configuration, the latter results in the user having to nominate that they have an SiS card. guess which one I picked :) [07:39] mdz: otherwise it would be just the initrd tarballs, and those need to be autobuilt too [07:39] mdz: (:0001 is used for TWO COMPLETELY DIFFERENT CLASSES OF PCI DEVICE) [07:39] Kamion: ah, gotcha [07:39] Kamion: I thought you were saying the doc package would be byhand-only, but you meant d-i itself [07:39] mdz: right. joeyh wrestled with this a lot a while back; the current situation is the resulting compromise [07:39] daniels: ok to upload initrd-tools [07:39] mdz: surely for a doc team we can simply let them commit to the right bit of arch [07:40] mdz: thanks [07:40] Kamion: yes, but we still have to get the stuff uploaded somehow [07:40] anyone know how i could track down a seemingly random hardlock? [07:40] mdz: sure, but d-i gets uploaded regularly anyway for other reasons [07:40] it would be nice to be able to loosen the reins for doc updates without worrying about all the byhand stuff that d-i involves [07:40] 0000:00:00.0 Class 0600: 1039:0745 (rev 01) [07:40] 0000:00:01.0 Class 0604: 1039:0001 [07:40] 0000:00:02.0 Class 0601: 1039:0018 [07:40] tseng: ctrl+scrollock [07:41] mdz: you can't make changes to initrd stuff without new d-i builds [07:41] or sysrq+t [07:41] mdz: I don't really think this is a problem [07:41] anyway, have to go now :) [07:41] ^ the SiS bridge here has an 0600 class [07:42] Keybuk: yes, but 1039:0001 is ambiguous -- it is both a video card, and a PCI-to-AGP bridge [07:42] daniels: ah, so that was a local kernel build that was breaking for you [07:42] *PHEW* [07:43] daniels: yeah, so check the class [07:43] for some reason, discover does totally bong matching (it forces the types of what it detected and what's in the file to equality if the PCI ID is the same, so no way to specify two classes, one ID), and I do *not* want to touch that code. i tried a couple of fixes and it broke stuff. so I just let it go, and both :0001 sis video users can deal. [07:43] mdz: yeah, not an RC bug [07:43] speaking of RC bugs [07:43] discover is broken === ..[topic/#ubuntu-devel:daniels] : Ubuntu development -- general discussion on #ubuntu | Happy Ubuntu Artwork week! | 9 RC bugs to go [07:43] Keybuk: no, really?!? [07:43] vendors are allowed to allocate the same id to two devices, provided they have different major classes, if I recall PCI correctly [07:44] Keybuk: that's crap [07:44] Keybuk: i don't think discover has a bridge class anyway, so lots and lots of stuff would break. infeasible to change it. [07:44] we'd just screw up more corner cases than we'd fix [07:45] mdz: comments on #2373? i think it can be closed out-of-hand. [07:45] #define PCI_CLASS_BRIDGE_PCI 0x0604 [07:45] it's reasonably clear what that device is :-) [07:46] Keybuk: yeah, but discover doesn't map this sanely to the 'bridge' class in pci.lst anyway [07:46] and half of them are down as unknown [07:46] tightening matching would break us horribly [07:46] discover doesn't even use the right list files, it has its own [07:46] assume we're talking about discover1 here [07:48] thom: I'm OK with uploading the fix in #2294 [07:48] thom: oh, wait [07:48] thom: that doesn't actually break any maintainer scripts, does it? [07:48] it just makes them think it's not a laptop [07:48] modules.pcimap has a specific 0600 class-match for sis-agp [07:50] mdz: yes [07:51] thom: send me a deb? [07:51] I'm in ultra-conservative asshole mode for Warty [07:53] sure [07:55] mdz: http://people.ubuntu.com/~thom/laptop-detect_0.9_amd64.deb [07:55] ew, it's arch-any? [07:56] mdz: has to be, to get the dmidecode dependencies right :( [07:58] thom: has someone tested it on powerpc? [07:58] mdz: dmidecode? upstream say no way, dunno if anyone has looked at making it work [07:58] thom: I mean laptop-detect [07:59] laptop-detect works fine on ppc [07:59] the patched version? [07:59] the current patch? don't know [08:00] if adding quotes to a variable breaks it, i'll be screamingly terrified [08:00] if you shang us a source package, I'll test it on i386, though I presume your X40 has also felt the love [08:00] thom: i know, it's incredibly trivial and obviously correct [08:00] but my eyes lie to me sometimes [08:00] I don't think this G4 desktop has /proc/pmu [08:01] mdz: er, IIRC it should [08:01] it ought to [08:01] ok, can do then [08:02] mdz@max:~ $ laptop-detect [08:02] mdz@max:~ $ echo $? [08:02] 1 [08:02] mdz: sudo? [08:02] same [08:02] oh, nevermind. [08:03] exit 1 is success in thom's world ;) [08:03] mdz@max:~ $ grep Bat !$ [08:03] grep Bat /proc/pmu/info [08:03] Battery count : 0 [08:03] thom: ok, fire away [08:03] daniels: no. exit 1 is failure, you freak [08:03] it's called laptop-detect, not desktop-detecty [08:03] thom: ber === lamont screams at ubuntu-sounds [08:06] mdz: if I made thinkpad-x40-support mangle /boot/grub/menu.lst to ensure kopt contained acpi_sleep=s3_bios, would I hear the distant sound of baby jesus weeping? [08:06] mdz: (this is just going to be a p.u.c package with some acpi scripts) [08:06] if you have a sound-using app, and you click a button, it fails to open the sound device, since UBUNTU F(^%)%^G SOUNDS grabbed it to make a cute noise. [08:06] daniels: that would be evil [08:07] mdz: suckful [08:07] lamont: open gstreamer-properties, set it to esdsink [08:07] another of those settings upgrade bugs [08:07] esdsink should be the default on new installs [08:07] I just told the desktop to stop issuing sounds on events. [08:07] equally valid [08:07] which is what I had before the last upgrade magically enabled it. [08:08] now that one was correct [08:08] errr [08:08] seb128: ping? [08:08] gstreamer-properties is still defaulting to OSS(!) === lamont goes to stare at the livecd. [08:08] thom: hm, speaking of acpi-support a blueskyism would be to have acpi-support-fallthrough or something with /etc/acpi/* [08:09] thom: so acpi-support-x40 could c/r/p it, and ditto for other setups if people want to make packages [08:09] yeah [08:09] mjg59: gplv2, or v2+? [08:09] mjg59: (your x40 acpi scripts) [08:09] daniels: Hoary :-) === thom is ignoring all the firefox bugs until post warty, they're all trivial, and all likely fixed in 1.0 [08:10] upstream have finally assigned someone to fix the crasher [08:12] thom: er yeah, I don't quite think we'll slip such a change through mdz for warty ;) [08:13] thom: hopefully 1.0 will arrive in time for the hoary freeze [08:13] tar tvzf boot/miniroot.gz [08:13] lamont@mix:/media/cdrom0 $ [08:13] hrm.. I think we might havea a winner [08:14] mdz: it should be, yeah [08:15] lamont: does it have the latest portmap on it? [08:16] mdz: december was the last estimate i saw === mvo [~egon@suprimo-215.ping.de] has joined #ubuntu-devel [08:17] thom: yeah, the last estimate we saw since the one that said 'september' [08:17] *g* [08:19] mdz: that was just looking at the CD that doesn't boot, trying to figure out what it's _supposed_ to look like [08:20] lamont: I think miniroot is a filesystem, not a tar [08:22] mdz: why, so it is. [08:23] mjg59: could I please get a copy of laptop-mode.sh (is it just /etc/init.d/laptop-mode?) and laptop.sh? [08:24] daniels: uh, apt-get install laptop-mode [08:24] pitti: It's hard to say no to #2177 [08:24] daniels: or just which laptop-mode [08:24] pitti: can you make packages available for testing before upload? [08:25] mdz: sorry, was away for a moment. Of course [08:25] mdz: I can put them into my utopia repo [08:25] pitti: are there other packages in there which are not destined for warty? [08:25] mdz: if the patch is wrong, the worst thing is a memory leak [08:25] mdz: no, currently only hal [08:26] thom: right, but mjg59 references laptop-mode.sh [08:26] daniels: same thing [08:26] I don't think it's security-relevant, but it's a big usability problem [08:26] and the fix is a very simple and very-safe one-liner [08:27] daniels: it's only packaged in ubuntu [08:27] thom: righto [08:27] mdz: that's why I begged for approval :-) [08:28] mdz: up [08:28] pitti: what's the sources.list entry? [08:29] mdz: deb-src http://people.ubuntulinux.org/~pitti/utopia/ / [08:29] (for source packageS) [08:29] mdz: same for binary [08:34] Failed to fetch http://people.ubuntulinux.org/~pitti/utopia/gnome-system-tools_1.0.0.orig.tar.gz 404 Not Found === mako_ [mako@micha.hampshire.edu] has joined #ubuntu-devel [08:38] lamont: did you get email from alex? [08:38] lamont: he says he told you the problem was that you need to modprobe loop [08:42] mdz: ooi, when is unstable getting synced for hoary? start of next week? [08:42] just now [08:43] thom: I think it should start as soon as warty is out [08:43] thom: thinkpad-x40-support stuff on p.u.c/~daniels/x40 as a very first crack (it installs OK, seems to work, but it's what I had anyway); if you wouldn't mind giving it a geeze when you've got a sec, that'd be ill [08:43] daniels: sick. [08:43] thom: phat [08:44] mdz: rightyho. might want to fix the release schedule then :-) [08:44] 0445, that sounds like new-netinst-o'clock! [08:44] lamont: can you do a new test build? === daniels does a last rsync of his homedir before he shoves his /home into ext3. [08:45] thombot? [08:46] daniels: laptop-mode.sh is the one that's in the kernel docs [08:46] thom: I need loop on macaroni... where does it's kernel come from? [08:46] The idea is just to call something that switches it on and off [08:46] mdz: pong [08:46] Oh, and GPL 2, ideally [08:46] lamont: custom [08:46] Mithrandir: was just going to interrogate you about gzip some more, but i've uploaded it now === lamont grovels [08:46] mdz: ok [08:47] mjg59: is that /usr/sbin/laptop-mode? [08:47] thom: liveCD's are blocked by lack of loop love. [08:47] mjg59: ok, I'll fix debian/copyright (right now it's 2+) [08:47] mjg59: cheers [08:47] daniels: Probably, yup [08:47] daniels: yes, laptop-mode ships that shell script as usr/sbin/laptop-mode [08:48] rockin' [08:48] lamont: gar and other words [08:49] thom: yeah. other words for sure. [08:49] thom: no pressure and all that, but eta? [08:49] I'll deal with it [08:49] elmo: thanks. quiescing that buildd [08:50] lamont: let me know the moment you have an image up that I can test [08:50] mdz: will do [08:51] http://www.gnome.org/~calum/blog/gnome.org/2004-10/2004-10-13@1328 -- so much crack [08:52] daniels: that'd rock hard [08:52] Drop a file on their head to send it to them. [08:52] mdz: sorry, I did not put the orig.tar.gz there [08:53] thom: yeah, the good sort of crack. but crack nonetheless. [08:53] i find the idea of having jdub's hackergotchi staring out from my panel kind of unnerving :) [08:54] mdz: try again, please [08:56] pitti: I already grabbed it from the main archive [08:57] lamont: done === ..[topic/#ubuntu-devel:daniels] : Ubuntu development -- general discussion on #ubuntu | Happy Ubuntu Artwork week! | 8 RC bugs to go [09:01] pitti: network-admin sees to work fine [09:01] mdz: was running about 20 minutes to build a broken CD... [09:02] mdz: for me, too [09:03] mdz: I also asked my flatmate to test it [09:03] mdz: works for him [09:03] pitti: ok, please upload [09:03] mdz: it'll be my pleasure :-) [09:03] mdz: BTW, if we already are at violating the freeze rules :-) [09:04] mdz: how long do you think should the hal stay in the unofficial repo? [09:04] mdz: I'm not sure what the optimal balance between early testers and left time to the release is [09:05] mdz: the more early we throw the package into the official archive, the more time we have to fix it again and also have more testers [09:05] mdz: OTOH it still should receive one or two days of testing [09:05] mdz: pong === inklingx [~inklingx@u212-239-167-156.adsl.pi.be] has joined #ubuntu-devel [09:06] seb128: on a fresh install, I opened gstreamer-properties, and it says OSS [09:06] seb128: shouldn't it be esd by default? [09:06] yes, should [09:06] I'm looking for it [09:07] thom: ping ? [09:07] seb128: word [09:07] * Set default audiosink to "esdsink" (Warty #1349) [09:07] -- Thom May Wed, 29 Sep 2004 16:20:48 +0100 [09:07] [09:07] how did you make this change ? [09:08] changed the schema [09:08] directly ? [09:08] ie: not patch in debian/patches ? [09:08] changed the schema.in, yeah [09:08] arg [09:08] that's it [09:08] ah, sorry [09:09] I only keep the debian/ dir between versions [09:09] I should have checked that more carrefully [09:09] mdz: fixing right now, the change was made directly in the source and has been dropped on the upgrade [09:15] seb128: btw, uupdate is really nice for such things [09:16] pitti: yes, but usually I just keep the debian/ dir because all the changes are here on the GNOME packages [09:16] so I'm sure to not keep crap out of this [09:16] seb128: I remember packages which did changes which were already fixed upstream directly in the code [09:17] ? [09:17] hrm... that's a bit different [09:17] wrong, but different. [09:17] pitti: oh yeah, if the change are in the upstream devel tree no point to keep them for the new version === lamont refuses to believe that 56MB is enough === smurfixx [~smurfix@run.smurf.noris.de] has joined #ubuntu-devel === smurfixx is now known as smurfix [09:18] Kamion: ping? [09:18] pitti: mixing changes in a big .diff.gz makes the work really harder to do [09:18] Opening input: File too large [09:18] \ [09:18] GAH!~ [09:18] Kamion: there's a user here who cannot install linux-image-386 off the cd with no network [09:21] 12:12 linux-386: Depends: linux-image-386 but it is not installable [09:21] 12:12 Depends: linux-restricted-modules-386 but it is not going to be installed [09:21] Kamion: (that was from d-i; unfortunately they no longer have the machine in d-i, so can't beat any debugging info out of them) [09:21] Kamion: can't see off the top of my head why the l-i-386 on the cd is uninstallable [09:22] Kamion: but it may be that it gets installed when only the network stuff is in the sources.list? [09:22] fabbione: ping? === ..[topic/#ubuntu-devel:daniels] : Ubuntu development -- general discussion on #ubuntu | Happy Ubuntu Artwork week! | 6 (count 'em) RC bugs to go === Mithrandir is a bit angry at himself for not seeing the acx100_pci problem on amd64 earlier. [09:24] fabbione: ugh, and all kernel bugs. Poor herbert... [09:24] pitti: yeah, and all totally wack ACPI bugs [09:25] #1922 seems to be caused by the acpi.sf.net patchset, of all things :\ === ..[topic/#ubuntu-devel:mdz] : Ubuntu development -- general discussion on #ubuntu | Happy Ubuntu Artwork week! | 6 (count 'em) major bugs [09:25] consider RC = severity>=critical for the remainder of the release [09:25] elmo: here? [09:26] mdz: sure [09:26] mdz: zero RC bugs to go [09:26] Mithrandir: isn't #2377 a bug you already fixed? [09:27] Kamion: 12:11 W: Couldn't stat source package list file: warty/main Packages [09:27] (/var/lib/apt/lists/_cdrom_dists_warty_main_binary-i386_Packages) - stat (2 No such file or directory) [09:27] Kamion: hm, upon further inspection -- that's the CD. [09:27] Kamion: *wack*. [09:29] mdz: looks a bit like the one I fixed, but not exactly. [09:29] Mithrandir: the fact that he says it worked before is suspicious [09:30] yeah, as I don't think grub has been changed in the last week, has it? === _rene_ [~rene@dsl-082-083-172-225.arcor-ip.net] has joined #ubuntu-devel [09:30] Mithrandir: no, please ask any necessary questions that I haven't already asked [09:31] <_rene_> mdz: that bibliography bugs is powerpc-specific, yes [09:31] <_rene_> mdz: (of OOo), forgot the number.. [09:31] _rene_: 1903? [09:32] mdz: buy me a couple of 2G sticks and I'll test with my home box? ;P [09:33] <_rene_> mdz: yes [09:33] <_rene_> mdz: debian bug 216848 [09:33] mdz: there's no way that problem didn't exist on Oct 8th. [09:34] <_rene_> and btw, you use GNOME 2.8, right? [09:34] <_rene_> have you fixed debian bug 272638? [09:34] Mithrandir: we don't have time for shipping around hardware; can you debug it with the submitter? [09:34] mdz: I can try, sure. === _rene_ is currently doing this [09:35] mdz: I wasn't serious about it either, though. [09:35] _rene_: unless the bug is RC or independently reported by an Ubuntu user, probably not [09:35] so, 1050803 KB doesn't compress so well into a CD image. === Mithrandir goes to reboot his system to recover the number of ptys [09:35] <_rene_> mdz: ok, it imho is a stopper for me. since you can't open files from GNOME without this [09:36] then again, unmounting proc would help. [09:36] _rene_: oh, yes, we did fix that [09:37] <_rene_> ah, k... [09:37] _rene_: https://bugzilla.ubuntu.com/show_bug.cgi?id=648 [09:38] <_rene_> mdz: p.d.o/~rene/openoffice.org/sysui-mimetypes.diff... [09:39] <_rene_> mdz: made today, where were the patches you promise to give back? ;) [09:39] <_rene_> oh, we should call update-desktop-database, too? [09:39] _rene_: yes [09:40] <_rene_> just run it? with or without args? [09:40] <_rene_> which args? [09:40] <_rene_> :) [09:40] <_rene_> that script unfortunately is undocumented... [09:41] <_rene_> (yes, I know of dh_desktop but that's not yet available in sarge...) [09:42] <_rene_> erm, nm === _rene_ looks in debhelper [09:43] <_rene_> ah, ok [09:45] <_rene_> mdz: expect a 1.1.2-6 with these and some other patches in the next days [09:45] <_rene_> mdz: when are you going to release? [09:48] _rene_: that bug did not apply to Debian, did it? [09:49] I thought it was only with GNOME 2.8 [09:49] <_rene_> yes, but it was reported since $reporter used gnome from experimental [09:49] _rene_: we release on 10/20; we are only accepting fixes for showstopper bugs and safe trivila fixes [09:49] <_rene_> I don't see why it shouldn't be fixed in the deb anyway :) [09:49] _rene_: what's the problem ? [09:50] seb128: #648 [09:50] oh ok [09:50] <_rene_> mdz: oh, any you probably want the openoffice.org-debian-files I am just preparing.. [09:50] yeah, that's GNOME 2.8 material [09:51] _rene_: showstopper bugs and safe trivial fixes only [09:51] <_rene_> mdz: ah, well, I'll let you decide.. [09:51] mdz: permission to make a gok upload ? The rules were in the wrong order in debian/rules, the schemas is not registred and gok just crashes [09:51] _rene_: do you have a changelog? [09:52] seb128: bug#? [09:52] <_rene_> not yet but I can tell you :) [09:52] <_rene_> - minor typo fix in manpafe (\fb -> \fB) [09:52] _rene_: we're in an ultra-conservative posture right now [09:52] <_rene_> - run setofficelang.bin with -f [09:53] mdz: #2141, but I've not attached the patch ... I've just changed the cdbs rules in debian/rules to match all the other GNOME packages [09:53] <_rene_> that latter fixes the problem that OOo doesn't start after initial setup [09:53] seb128: can you attach a patch? [09:53] mdz: ok [09:55] mdz: done [09:55] seb128: hmm? I don't see it [09:55] wrong bug maybe? [09:55] oh, you added it as a comment [09:55] rather than an attachment [09:55] yes [09:55] seb128: ok, looks good, go ahead [09:56] thanks [09:56] since it doesn't seem to work at all, I don't suppose it could get worse from this :-) [09:57] yeah :) [09:57] <_rene_> mdz: p.d.o/~rene/openoffice.org/oo-d-f.patch <-- 2 lines... [09:57] _rene_: I haven't seen the problem where oo.o doesn't start after initial setup [09:57] it seems to work for me [09:57] <_rene_> mdz: erm, ignore that comment in the file. *fixes* [09:57] <_rene_> mdz: oh? [09:58] <_rene_> LANG="C"? [09:58] <_rene_> or something like that? [09:58] _rene_: I do fresh Ubuntu installs on a regular basis, and starting oo.o is one of my test cases [09:58] <_rene_> hmm [09:58] LANG=en_US [09:58] <_rene_> I can reproduce this here at will [09:58] <_rene_> (with de_DE) [09:58] _rene_: is there some way I can test, having already setup oo.o? [09:59] rm -rf .openofice .sversionrc? [09:59] <_rene_> yes [09:59] <_rene_> export LANG="de_DE@euro" or so [09:59] <_rene_> en_US works fine here, too [09:59] <_rene_> probably because there is no UI language to change :) [09:59] confirmed [10:01] _rene_: what does -f do? [10:01] <_rene_> force [10:02] ah, otherwise it exits nonzero and so oo.o fails to start? [10:02] (due to the script exiting?) [10:02] or it changes the way setofficelang works? [10:02] <_rene_> mdz: right. it exits 1 [10:02] <_rene_> mdz: and exits the set -e script :) [10:02] ok, I think we can take -5+1 [10:02] <_rene_> s/exits 1/returns 1/ [10:07] mdz: permission to upload gst-plugins0.8 with the schemas back to esdsink ? [10:08] seb128: yes [10:09] jdub: please tell me we can lose esd for hoary [10:11] thom: don't you have an ipw2100? [10:11] mdz: yeah? [10:11] thom: #2364 [10:13] mdz: #2384 please [10:13] those with bandwidth: http://people.ubuntu.com/~lamont/testing/live.iso if you would be so kind. [10:14] mdz: my ipw2100 is working great, with the -15 kernel [10:14] seb128: the patch seems neither trivial nor obviously correct without context; can you explain it to me? [10:15] thom: please note that on the bug for Herbert [10:15] will do [10:15] thom: you're not up to -16? [10:15] not that it should matter === albireo [~albireo@lns-vlq-6-82-65-50-151.adsl.proxad.net] has joined #ubuntu-devel [10:16] mdz: not looked on the details to be honest, I trust alex (the main gnome-vfs/nautilus devel) for this ... but I can look the detail if you want [10:16] seb128: if warty were released, would you still push this into warty-updates? that is the kind of criteria we need to have at this time [10:17] mdz: no, i have a patched -15.1 with inotify, -16 doesn't fix anything i care about on the laptop [10:17] all other machines are on 16 [10:18] mdz: yes, that fixes a nasty data corruption during transferts === lamont taps his foot [10:19] alex was able to reproduce the bug, tested the fix and commited in 2.8 branch upstream [10:19] mdz: I'll just check what was the problem and what the fix does to be sure and ping you again [10:20] can't hurt to double check :) [10:22] new X????? [10:22] -6ubuntu24, that is... [10:23] or maybe I'm just way behind on things.. [10:23] yeah. I'm way behind. That was from Monday. === lamont crawls back into the hole he crawled out of === Mithrandir crawls into bed. [10:33] <_rene_> mdz: uploaded [10:34] <_rene_> mdz: take it from incoming in a few mins :) [10:34] <_rene_> s/:/: you can/ [10:34] _rene_: we wait for it to be available on ftp.uk.debian.org :-) [10:34] <_rene_> mdz: ah, ok :-) === maskie [~maskie@196-30-110-242.uudial.uunet.co.za] has left #ubuntu-devel ["Leaving"] === lamont has to go get the kids from school. back around 2200 UTC or so. [10:39] although I may login in town and fetch bits. dunno. === azeem [~mbanck@socks-out.lrz-muenchen.de] has joined #ubuntu-devel === daniels attacks another vi mug full of Madagascan Vanilla tea. [10:45] it'll be a few before I leave actually... brb [10:45] daniels: please, think of the children! [10:46] tseng: ? === tseng hides. [10:47] daniels: sorry, i thought you were referencing the vim donate to starving kids deal [10:47] wrong country. [10:48] heh, no [10:48] :help iccf [10:49] may i note that the old logo-only gdm theme isnt even present here anymore? [10:49] i keep hearing "those people arent the default" [10:50] d-i really does hate apt-cacher [10:52] deservedly so [10:52] it's crap [10:52] tseng: you're hearing that in the context of the wallpaper [10:53] the new gdm theme is default [10:53] i see === tseng digs around for the Industrial gdm theme =/ === Keybuk [scott@descent.netsplit.com] has joined #ubuntu-devel [10:58] well, that was nice ... something just decided to make my machine down [11:00] mdz: any news? [11:01] lamont: ENOSPC [11:01] lamont: eta 20 minutes === Capri [~makolb@mnch-d9ba4186.pool.mediaWays.net] has joined #ubuntu-devel [11:01] ARRGH! [11:02] Oct 14 21:51:15 descent kernel: oom-killer: gfp_mask=0x1d2 [11:02] *shrug* aren't log messages great :o) [11:02] yay! === lamont leaves to fetch kidlets, CD. [11:04] ah [11:04] figured it out === Keybuk decides to beat elmo senseless [11:04] good god. it's actually trying to snow or sleet === Keybuk adds "ProxyCommand none" to his chinstrap.warthogs.hbd.com .ssh/config entry [11:05] Keybuk: that's typically helpful, yes [11:06] hrm [11:06] that doesn't help [11:07] urgh [11:07] I needed to add another host entry, adding it to the "chinstrap" one didn't actually turn off ProxyCommand === _rene_ wonders whether he should buy a larger disk and help with testing/developing Ubuntu besides Debian stuff. [11:16] thom: it's the UK, what do you expect? :) [11:16] october is a little premature === daniels looks outside to a bright, mild day; probably low-to-mid 20s. [11:17] all this and it's only 0718. [11:18] thom: it was pretty cold last night :-/ [11:22] hate hate hate libtool, I keep getting app linked with both Xft == broken apps [11:23] well, don't link apps with both Xft then [11:23] it's not libtool doing that to you [11:23] it doesn't conjour dependencies out of thin air [11:24] if you didn't use libtool, apps would still be linked with both Xft and still broken [11:24] yeah, probably, finding the problem is just a pain [11:24] for the moment I can't build gtk+ and control-center on my box [11:25] and now yelp in my jhbuild has the problem too [11:25] I'm decided to track this down, I just need to figure how :) [11:25] apt-get remove libxft1 [11:26] 44 to remove [11:26] The following packages will be REMOVED: [11:26] abiword-plugins capplets eterm galeon galeon-common gdeb gdm gimp gimp1.2 [11:26] gksu gnome-applets gnome-applets-data gnome-applets-dev gnome-apt [11:26] gnome-control-center gnome-netstatus-applet gnome-panel gnome-panel-data [11:26] .. [11:26] x-window-system-core xbase-clients xlibs [11:26] na, not good [11:26] there's still that much stuff built against libxft1? that's total shit [11:27] unless xlibs still deps that, but I'd be pretty surprised [11:27] but that could explain it [11:27] seb128: libtool makes that easy! [11:27] anyway, breakfast time [11:27] just look for it in the .la files [11:28] the deepest one that contains /usr/X11R6/lib/libXft.so.1 in its dependency_libs line is the culprit [11:28] no /usr/X11R6/lib/libXft.so.1 in /usr/lib/*.la [11:29] which ones have -lXft in them? [11:29] $ grep "lXft" /usr/lib/*.la [11:29] /usr/lib/libodbcinstQ.la:dependency_libs=' -L/usr/X11R6/lib /usr/lib/libqt-mt.la -laudio -lXt -lpng -lz -lGL -lXmu -lXrender -lXcursor -lXinerama_pic -lXft -lfreetype -lfontconfig -lSM -lICE -lXext -lX11 /usr/lib/libodbcinst.la /usr/lib/libodbc.la /usr/lib/libltdl.la -ldl -lpthread' [11:29] /usr/lib/libqt-mt.la:dependency_libs='-L/usr/X11R6/lib -lfontconfig -laudio -lXt -lpng -lz -lXrender -lXrandr -lXcursor -lXinerama -lXft -lfreetype -lfontconfig -lXext -lX11 -lm -lSM -lICE -ldl -lpthread ' [11:29] that's all [11:29] *ahem* [11:30] daniels@nanasawa:~/music/Aphex Twin/Richard D James Album% ls -l /usr/lib/libXft*.so [11:30] lrwxrwxrwx 1 root root 15 2004-08-31 11:07 /usr/lib/libXft.so -> libXft.so.2.1.1 [11:30] i don't think you can actually link against libxft1 anymore [11:30] $ ls -l /usr/lib/libXft*.so [11:30] lrwxr-xr-x 1 root root 15 2004-05-31 19:02 /usr/lib/libXft.so -> libXft.so.2.1.1 [11:30] daniels: sure you can, it's in /usr/X11R6/lib [11:30] i think i stopped installing the .so with like xfree86_4.3.0-0ds3v2 [11:30] lrwxrwxrwx 1 root root 13 2004-10-11 12:31 /usr/X11R6/lib/libXft.so.1 -> libXft.so.1.1 [11:31] Keybuk: if you explicitly specify .so.1 ... [11:31] $ ls -l /usr/X11R6/lib/libXft*.so [11:31] lrwxr-xr-x 1 root root 11 2004-04-06 21:23 /usr/X11R6/lib/libXft.so -> libXft.so.1 [11:31] daniels: that's why I asked him to look for that :p [11:31] seb128: dpkg -S /usr/X11R6/lib/libXft.so [11:31] Keybuk: no package !? [11:31] seb128: wtf is that link coming from?? === pitti [~martin@195.227.105.180] has joined #ubuntu-devel [11:31] remove it [11:31] WITH EXTREME PREJUDICE. [11:31] I think a old mess with xfree package in unstable [11:32] then recompile everything you ever built against that libXft [11:32] run, don't walk [11:32] seb128: that be your problem, as soon as you have -L/usr/X11R6/lib, GTK+ will link with that insteadf [11:33] Keybuk: oh ok, that would explain why I only get the problem on my box [11:33] I'm just wondering why this link is staying here [11:33] that'll learn you for using unsupported, experimental, X 4.3 packages :P [11:34] daniels: yeah, I've used them :p [11:34] seb128: there's your problem, clearly! [11:34] I had a good reason, but I don't remember which one now :) [11:34] but why didn't the symlink vanish when he upgraded ? [11:34] they were shiny? :P [11:34] Keybuk: i don't know. sounds like total bong to me. [11:34] daniels: no, IIRC to reproduce some problem with totem ... [11:34] I used the experimental 4.3 packages and I don't have that symlink [11:35] kamion: so what are the rsync options for mirrors? [11:35] rsync://archive.ubuntu.com/ubuntu [11:35] this gets the whole repo, right? [11:35] what gets just the release cd images? [11:36] and is there one that gets both, together? [11:38] daniels, Keybuk: thanks, much better now :) [11:39] "never attribute to libtool that which can be attributed to symlinks that aren't supposed to be there" [11:43] noted :) [11:43] keybuk: is this a rumor that libtools doesn't like `~' in directory names? [11:44] probably wouldn't, no :p [11:44] doesn't like spaces in them either [11:45] so the introduction of `~' in version numbers is pretty useless for packages using libtool ... [11:51] yeah [11:51] mdz: fam and esd are both on the HoaryHedgehog page, slated for replacement ;) [11:51] THIS IS WHAT HAPPENS YOU DON'T ALLOW NEW FEATURES TO BE TESTED UNTIL THERE'S A NEW STABLE RELEASE, FOLKS [11:51] the Wallpaper of Controversial Chaos *giggle* === jdub goes for breakfast [11:58] doko: where is this rumour?