[01:14] <Robot101> thom: are you going to fosdem? :)
[02:02] <Kano> hi, when will be a kernel based on 2.6.20 final?
[02:02] <mjg59> Soon
[02:02] <Kano> you should check for working nvidia drivers... there was a change since rc6
[02:03] <mjg59> The most recent kernel is post-rc6 anyway
[02:04] <Kano> but not final?
[02:04] <crimsun> mjg59: when the latest patch_sigmatel changes are merged, feedback from testing on your Intel Mac would be appreciated
[02:04] <Kano> why dont you use only libata?
[02:04] <mjg59> crimsun: Don't have one right at the moment, but when I do again I will do
[02:04] <mjg59> crimsun: cjwatson has the one I used to have
[02:04] <crimsun> mjg59: ah ok, many thanks
[02:05] <mjg59> Kano: Because not all libata drivers are complete
[02:14] <Kano> which aren't?
[02:17] <Kano> isnt it a bit stupid to have both the same time
[02:17] <Kano> piix 0x00008086 0x0000245b
[02:17] <Kano> ata_piix 0x00008086 0x0000245b
[02:18] <Kano> both for same hardware
[02:29] <Kano> with that there are 229 pci ids with more than one driver
[02:29] <Kano> grep drivers/ata modules.dep |sed -r 's/.*\/(.*)\.ko:.*/-e \1/'|sort -u|xargs grep modules.pcimap|awk '{print $2" "$3" "$4" "$5" "$6" "$7" "$8}'|sort -d|wc -l
[02:29] <Kano> 229
[02:32] <Kano> or 31 modules which share pci ids
[02:33] <Kano> grep drivers/ata modules.dep |sed -r 's/.*\/(.*)\.ko:.*/-e \1/'|sort -u|xargs grep modules.pcimap|awk '{print $2" "$3" "$4" "$5" "$6" "$7" "$8}'|sort -d|sed -r 's/(.*)/-e "\1"/'|xargs grep modules.pcimap|cut -f1 -d' '|sort -u|wc -l
[02:34] <Kano> do you really think thats usefull
[02:35] <kylem> of course they share ids. they both support the same hardware.
[02:35] <Kano> but they could not work the same time
[02:35] <kylem> so?
[02:35] <Kano> only one of em
[02:36] <Kano> maybe better find pci ids which are only supported by old ide 
[02:45] <Kano> grep drivers/ide modules.dep|sed -r 's/.*\/(.*)\.ko:.*/-e "^\1 "/'|sort -u|xargs grep modules.pcimap|awk '{print $2" "$3" "$4" "$5" "$6" "$7" "$8}'|sed -r 's/(.*)/-e "\1"/'|xargs grep modules.pcimap|awk '{print $2" "$3" "$4" "$5" "$6" "$7" "$8}'|sort|uniq -u|sed -r 's/(.*)/-e "\1"/'|xargs grep modules.pcimap|sort -u|wc -l
[02:45] <Kano> 71
[02:46] <Kano> uniq ids only for old driver, but these could be matched by generic patterns maybe too
[02:49] <Kano> which are preffered?
[02:49] <Kano> in case of dups
[02:54] <Kano> any special probing sequence?
[02:54] <xhaker> crimsun, noticed your upload of alsa-lib, and I have something I feel you're the one to ask.
[02:54] <xhaker> crimsun, what alsa version for the final feisty release?
[03:01] <crimsun> xhaker: which part of alsa are you asking about?
[03:01] <crimsun> xhaker: alsa-lib will be 1.0.13 with any appropriate bugfixes backported from hg
[03:02] <crimsun> xhaker: alsa-kernel (via linux-source-2.6.20) will be a mash of current hg and 1.0.14rc2
[03:03] <crimsun> xhaker: -utils will be 1.0.13 with any appropriate bugfixes; -tools will be 1.0.14rc; -firmware will be 1.0.14rc
[03:55] <xhaker> thanks crimsum, i'll wait and see, looking forward to 2.6.20 final and new alsa
[07:36] <carebear666> if you ever try to kick me or ban me i will get my uncle(irc opp) after you and by the way i have a nuke and a mail bomb on the way asshole
[07:40] <carebear666> if you ever try to kick me or ban me i will get my uncle(irc opp) after you and by the way i have a nuke and a mail bomb on the way asshole
[09:48] <concept10> anyone around
[09:58] <thom> Robot101: yes
[09:58] <mjg59> Win
[10:11] <glatzor> morning mvo_ Riddell
[10:11] <mvo_> hey glatzor!
[10:14] <glatzor> mvo_: Riddell: I several issues in software-properties and python-apt. would be nice if you you upload them mvo_
[10:15] <mvo_> glatzor: merging python-apt now
[10:15] <glatzor> mvo_: the update-manager separation branch is also up-to-date
[10:16] <mvo_> glatzor: I will merge that one next and upload them 
[10:17] <glatzor> mvo_: thanks a lot
[10:18] <mvo_> glatzor: cheers! thanks for working on the branch!
[10:19] <glatzor> mvo_: software-properties currently has got some waiting_for_implementation issues regarding the mirror support (if you choose a mirror from the chooser it won't be used). but i haven't had the time to fix it. I think that is more important to get the KDE stuff in
[10:20] <glatzor> mvo_: perhaps I find the time to the this on my today's train ride :)
[10:20] <glatzor> to do this
[10:20] <jsgotangco> mvo_, glatzor hi
[10:20] <glatzor> hey jsgotangco!
[10:20] <jsgotangco> how is the german g-a-i connection hehe
[10:21] <mvo_> glatzor: agreed
[10:21] <glatzor> mvo_: jsgotangco: i have to prepare for work.
[10:21] <mvo_> glatzor: ok, see you later!
[10:23] <jsgotangco> ok later
[10:40] <glatzor> Riddell: I am going to split show_distro in init_distro and show_distro soon
[10:47] <elkbuntu> ogra_, *ping* please pm me when you're around. thanks
[10:54] <cjwatson> crimsun: happy to help test patch_sigmatel if you tell me what it is and how I can test it ;-)
[10:55] <Tonio_> hi, I uploaded kdepim on friday afternoon, upload is still in the queue, is there a reason for this ?
[10:57] <cjwatson> Tonio_: looks like Tollef forgot to let the contents of the queue through after unfreezing feisty
[10:57] <cjwatson> Tonio_: I've done it now
[10:58] <Tonio_> cjwatson: hehe, okay ;), that can happen :)
[10:59] <webben> Does anyone know if there are any statistics available on which packages are downloaded from the Ubuntu (or Debian or indeed any) repository?
[11:00] <webben> I'd be interested to find out how many downloads there are of some of the alternative browsers and Emacspeak.
[11:04] <mvo_> webben: the closest thing we have is poopcon.ubuntu.com
[11:05] <webben> mvo_, thanks :)
[11:09] <elkbuntu> um, i believe he means popcon.ubuntu.com
[11:10] <elkbuntu> not sure i want to know if the former even exists
[11:11] <Chipzz> is feisty frozen wrt new translatable strings already?
[11:11] <TheMuso> cjwatson: What is the absolute latest that something can be seeded before ff?
[11:11] <seb128> Chipzz: no
[11:13] <Chipzz> mvo_: sorry for bugging you again, but could you take a look at that patch I pointed you at a couple of days ago? IMHO, it's nice polish, and the patch is only a couple of lines really
[11:13] <cjwatson> TheMuso: Thursday
[11:14] <Chipzz> seb128: is there even such a thing as a string freeze? :)
[11:14] <cjwatson> Chipzz: https://wiki.ubuntu.com/FeistyReleaseSchedule
[11:15] <Chipzz> oh, still a month away
[11:15] <Chipzz> I expected it to be sooner really ;)
[11:16] <seb128> feisty is still not opened on rosetta anyway
[11:16] <seb128> so no real point to freeze strings
[11:17] <pitti> carlos: how did the experiment go about rosetta + feisty?
[11:19] <carlos> pitti: not as good as it should, we are still doing some tests to see how to do it faster
[11:19] <fabbione> morning
[11:19] <fabbione> ogra_, pitti: ping?
[11:19] <pitti> hi fabbione 
[11:19] <fabbione> hey pitti
[11:20] <TheMuso> cjwatson: Thanks.
[11:20] <fabbione> pitti: could you be so kind to look at bug #83059? I am not sure if it's gnome-power-manager or hal that's exporting info at random
[11:20] <Ubugtu> Malone bug 83059 in Ubuntu "Hibernation button only appears after second time in Feisty Herd 3" [Undecided,Unconfirmed]  https://launchpad.net/bugs/83059
[11:20] <cjwatson> TheMuso: (but pushing the deadline is deprecated)
[11:20] <seb128> fabbione: that's already fixed
[11:21] <fabbione> i have seen it happening on powerpc too (while we were in Olso)
[11:21] <seb128> fabbione: could you try pinging right people about bugs?
[11:21] <seb128> fabbione: ogra is maintaining that package
[11:21] <pitti> seb128: he did
[11:21] <fabbione> seb128: i did ping both ogra and pitti...
[11:21] <pitti> seb128: good morning, btw *hug*
[11:21] <fabbione> seb128: because i was not sure what the fault was
[11:21] <seb128> hey pitti :)
[11:21] <fabbione> seb128: and i was asking for information..
[11:21] <TheMuso> cjwatson: I understand that the sooner the better is preferable. Just wondering as espeak is approved for main, but as far as I know is not seeded. At this point, I'd rather wait till Wednesday to seed it, as there is a new release due on Tuesday that I'd like to get in first.
[11:21] <TheMuso> While its still in Universe.
[11:22] <seb128> fabbione: did you install the update package from friday which ships the autostart desktop again?
[11:22] <seb128> fabbione: it looked like you were trying to get people fixing bugs you point again
[11:22] <fabbione> seb128: this bug is from hw-cert on herd-3. machines don't get upgraded. we certify the release
[11:23] <fabbione> seb128: it would be nice if you could speak with your manager about that.
[11:23] <seb128> about what?
[11:23] <heno> TheMuso: why does that matter, isn't that just a bug-fix release?
[11:23] <fabbione> seb128: about me reassining bugs to people
[11:23] <TheMuso> heno: Not sure.
[11:23] <seb128> fabbione: I don't need to, I'm happy to fix support bugs, I'll not fix random bugs you point though
[11:23] <heno> TheMuso: I would suggest seeding it before Feature Freeze and then address bugs afterwards
[11:24] <fabbione> seb128: it's not a random bug.. it comes from the support center
[11:24] <seb128> well, then make clear the procedure
[11:24] <heno> I'm in late on this conversation, but I assume that's the deadline cjwatson was referring to
[11:24] <fabbione> seb128: could you be so kind to explain to me what is not clear?
[11:25] <fabbione> seb128: so that we can actually fix that?
[11:25] <seb128> it looks like you point random bugs to random people and try to make them fix them
[11:25] <Robot101> thom: cool :)
[11:25] <fabbione> seb128: ok sec..
[11:25] <seb128> nobody told me I was supposed to fix your bugs, so I'll no do until I'm asked to do so
[11:26] <webben> The README and FAQ for popcon.ubuntu.com are both down. Are the voting figures effectively votes received /this/ week? Or are they on a longer cycle?
[11:26] <ogra_> fabbione, thats fixed with my last upload, did you upgrade after herd3 ? 
[11:27] <fabbione> ogra_: as i said to seb128 , we don't upgrade the machines. we certify the milestone
[11:28] <ogra_> oh, ok
[11:28] <ogra_> well, its broken on the milestone 
[11:28] <cjwatson> TheMuso: I'm with heno on this
[11:29] <TheMuso> o kok fine
[11:29] <heno> we'll have enough madness around FF without adding this ;)
[11:30] <cjwatson> fabbione: but now we know that the milestone is broken, so would it hurt to test the fix?
[11:30] <fabbione> cjwatson: of course we can test the fix. I was only looking for somebody to tell me where the bug was because i am not familiar with that part of distro
[11:30] <heno> cjwatson: can I just request the seed change of you here, or in some other format? eSpeak seeded for the CD, Festival un-seeded
[11:31] <ogra_> fabbione, the autostart file wasnt installed so g-p-m isnt run ... if you want to reproduce: if you start the power preferences applet g-p-m is started in the background ... if that fixes the logout screen, thats the bug
[11:31] <cjwatson> heno: festival doesn't seem to be seeded?
[11:31] <cjwatson> heno: anyway, seems like a good time for you to learn how to make seed changes yourself :)
[11:31] <fabbione> ogra_: thanks for the details. can you please add it to the bug in LP so that cr3 can double-test
[11:31] <fabbione> ?
[11:31] <cjwatson> heno: oh, you aren't in ubuntu-core-dev
[11:31] <fabbione> (i can't copy paste from this machine atm)
[11:31] <ogra_> oki
[11:32] <heno> cjwatson: no, I think I've tried pushing seed changes before without luck
[11:32] <cjwatson> heno: http://people.ubuntu.com/~cjwatson/germinate-output/feisty/rdepends/ALL/festival
[11:33] <cjwatson> heno: ^-- indicates that festival is in main due to dependencies, not due to being explicitly seeded
[11:33] <cjwatson> specifically the libgnome-speech3 dependency
[11:34] <TheMuso> cjwatson: dholbach has gnome-speech ready to be uploaded with a dependncy on libespeak1.
[11:34] <TheMuso> As far as I am aware when I was talking to him last.
[11:35] <heno> cjwatson: so libgnome-speech3 depends on festival? 
[11:35] <TheMuso> heno: Are we dropping festival completely?
[11:36] <heno> right, so libgnome-speech should depend on eSpeak and recommend festival I guess
[11:36] <heno> Festival should stay in Main IMO but be dropped from the CD
[11:37] <TheMuso> heno: If we are going to recommmend festival, gnome-speech's festival driver will probably ahve to be in its own package.
[11:38] <heno> TheMuso: why, can you not install that driver without installing festival?
[11:38] <TheMuso> Gnome-speech can have a festival driver present and festival not be available, but I think it would make things confusing if there was a festival option there, but no speech was heard.
[11:39] <cjwatson> TheMuso: that should just be uploaded, and then espeak can be promoted
[11:39] <cjwatson> heno: yes
[11:39] <cjwatson> heno: if festival is to stay in main, it needs to be added to the supported seed. dholbach should be able to do that
[11:40] <heno> ok, and we should probably remove gnopernicus from main now, It's ot much supported anymore
[11:41] <heno> cjwatson: ok, I'll email dholbach with these requests
[11:46] <ogra> *comes
[11:53] <Mithrandir> siretart: gxine is main and now build-depends on xulrunner, which isn't in main.   I suggest you either revert your change or talk to pitti about xulrunner in main.
[11:53] <pitti> yet another firefox copy in main? only if ffox itself uses xulrunner
[11:54] <ogra> switch it !!
[11:57] <ogra> :)
[11:59] <seb128> xulrunner to main? waouh :)
[11:59] <seb128> *g*
[12:00] <ogra> :)
[12:00] <asac> hello :)
[12:00] <ogra> elkbuntu, i'm on it ...
[12:00] <elkbuntu> ogra, thanks.
[12:03] <Chipzz> pitti: firefox not using xulrunner is an upstream problem, right?
[12:03] <pitti> Chipzz: I don't know anything about xulrunner
[12:04] <asac> Chipzz: not exactly I would say :) ... mozilla plans to switch to xulrunner for firefox 3
[12:04] <pitti> hey asac!
[12:05] <asac> hello pitti 
[12:05] <Chipzz> asac: yes. WAY too late
[12:05] <pitti> slomo: ping
[12:05] <asac> yes ... for feisty there will be no xulrunner
[12:05] <Chipzz> they should have switched for firefox 2
[12:05] <slomo> pitti: pong
[12:06] <pitti> slomo: not sure whether you got my q last week: do you actually get apport reports from mono program crashes in current feisty?
[12:06] <pitti> slomo: whenever I kill -SEGV a mono program, mono intercepts the signal, prints out its stuff and then just exits the program
[12:06] <pitti> slomo: apport isn't even triggered
[12:06] <pitti> slomo: (context: I wanted to add the mono blacklisting)
[12:07] <Chipzz> asac: I don't get it anyway... isn't postponing that creating more of a maintenance nightmare anyway?
[12:07] <slomo> pitti: hmm, i didn't get a new crash since some time... but when testing now with -SEGV it just crashes... something must've changed in apport then (as mono wasn't changed in months) :/
[12:07] <asac> Chipzz: maybe ... anyway, its not supported by mozilla.org; so what can we do?
[12:07] <pitti> slomo: but as I said, apport is not even triggered. hmm
[12:08] <pitti> slomo: ok, so I leave this at it is for now
[12:08] <slomo> pitti: probably the best until someone touches mono to add apport magic :)
[12:08] <asac> so how can I get the power to change all firefox bugs? e.g. severity?
[12:08] <pitti> asac, siretart: easiest is IMHO to build gxine against firefox, not xulrunner
[12:08] <Chipzz> asac: hrrrm, I was somehow under the mistaken impression that you were related to mozilla development ;)
[12:10] <pitti> asac: you probably need to become member of the bugsquad team; can anyone confirm? cjwatson?
[12:11] <asac> Chipzz: related .. yes ... but not deeply involved atm ;).
[12:11] <Adri2000> only members of the QA team can change severity
[12:12] <asac> Adri2000: package maintainer too, right?
[12:13] <Adri2000> ubuntu-dev and ubuntu-core-dev can since they are members of ubuntu-qa, otherwise I don't think so
[12:14] <pochu> pitti: to change importance you need to be a member of ubuntuQA, and to be a member of ubuntuQA you need to be a member of the BugSquad :)
[12:14] <Fujitsu> pochu, you don't need to be a member of the bugsquad.
[12:14] <pochu> oups: Adri2000: didn't see your message ;)
[12:14] <pochu> Fujitsu: not? I think yes :)
[12:15] <Fujitsu> No.
[12:15] <pochu> but sfllaw knows it :)
[12:15] <pochu> Fujitsu: then?
[12:16] <Fujitsu> Being a member of bugsquad isn't a requirement. ubuntu(-core)-dev are implicitly members of ubuntu-qa, among others.
[12:18] <heno> pochu: it's the most common way, but not a requirement. People who contribute reliably to the ISO testing team should also qualify for Ubuntu QA
[12:19] <heno> I agreed this with sfllaw in Oslo
[12:19] <pochu> didn't know :)
[12:19] <heno> pochu: are you in Ubuntu QA yet?
[12:19] <pochu> this is from the wiki: In order to join:
[12:19] <pochu>     *
[12:19] <pochu>       sign up: with the BugSquad.
[12:19] <pochu> heno: not, but I would like to join in the next hug day :)
[12:19] <pochu> heno: February 14th :)
[12:20] <heno> didn't I encourage you do join, I meant to
[12:20] <heno> pochu: you should join the bugsquad, it's a great team
[12:20] <pochu> heno: I'm a bugsquad member :)
[12:21] <heno> pochu: but if you just sign up to Ubuntu QA I'll see that you are approved
[12:21] <pochu> ?
[12:21] <gnomefreak> heno: he shouldnt have a problem at all
[12:21] <heno> I've already agreed your case with Simon
[12:22] <pochu> which case? mine? I can't understand you :(
[12:25] <heno> pochu: yes, just apply on Launchpad for Ubuntu QA
[12:25] <heno> pochu: your contributions to organising ISO testing are enough
[12:26] <heno> no need to wait for hug day
[12:26] <pochu> heno: ok :) I haven't done that because I read on the wiki about triaging and the hug day
[12:26] <pochu> however, I will come to the hug day :D
[12:26] <Mithrandir> heno: sorry about not seeing your questions before today, but I haven't been watching work channels, I was hacking on photo stuff all weekend.
[12:27] <heno> Mithrandir: sounds sensible. We'll just leave it as it is now I think
[12:27] <pochu> heno: about the iso team, some people still reports in the forum. I've asked them to report the real bugs to Launchpad, but don't know if I should ask them to make their reports to the iso tracker. Should I?
[12:28] <Mithrandir> heno: yes, I just need to fix up the script (and put it in bzr somewhere so you can get at it)
[12:28] <heno> Mithrandir: need to think about how this will hold up/scale to more demanding release phases
[12:28] <heno> Mithrandir: cool, thanks
[12:28] <pochu> heno: also I've read your mail to the -devel list. If you need some help, just tell me what can I do
[12:29] <heno> pochu: ok, thanks. I think you are right to point people at filing Launchpad bugs They could also file ISO test comments (I've now posted Herd 3 tracker bugs)
[12:30] <pochu> heno: ok, I'll do it
[01:00] <seb128> doko: do you plan to ask for a sync of the new ttf-dejavu from Debian?
[01:01] <doko> seb128: IMO we should update to the new upstream release
[01:01] <sfllaw> doko: Would it be possible to make a ttf-dejavu-lgc version?
[01:01] <seb128> doko: they have a newer version, is there yet another new version upstream?
[01:02] <sfllaw> doko: That builds the Latin-Greek-Cyrillic only versions of these fonts?
[01:02] <sfllaw> Apparently, the Chinese character sets for DejaVu do not look very good.
[01:02] <siretart> Mithrandir: pitti: I have to apologize, I didn't notice that gxine was promoted to main. Darren did a lot of work to fix gxine issues with gtk 2.10. I tried to get the package so that we can sync it from debian, but obviously, this won't work
[01:02] <geser> Mithrandir: Hi, now that herd3 is out have you some time to look at the build failure of xmms2? It builds in a pbuilder, during lucas' rebuild and on Debian's buildds but not on ours.
[01:04] <pitti> Checking for working C compiler... no
[01:04] <pitti> OPTION CC MUST POINT TO A VALID C COMPILER!
[01:04] <pitti> haha
[01:04] <Mithrandir> geser: I have no idea about it, really.
[01:05] <Mithrandir> geser: upload a new version where you run the configure with -x or its equivalent.
[01:06] <doko> seb128: didn't see that; yes, 2.14 sounds fine
[01:06] <seb128> doko: ok, can you do a sync request then please? ;)
[01:07] <pitti> seb128, doko: I can sync it right away if you want to
[01:07] <doko> sfllaw: looks good, but one week before ff? if you do have the time please go ahead
[01:07] <doko> pitti: please do
[01:07] <seb128> pitti: please do if doko is ok
[01:07] <pitti> oh, hm, Mithrandir seems to do syncs ATM
[01:07] <seb128> ok, he is, then do ;)
[01:08] <sfllaw> seb128: Are we sticking with bug 81608?
[01:08] <Ubugtu> Malone bug 81608 in kubuntu-default-settings "Please use DejaVu Sans Condensed as the default font" [Undecided,Unconfirmed]  https://launchpad.net/bugs/81608
[01:08] <sfllaw> If so, I will upload a version that does LGC.
[01:09] <seb128> sfllaw: well, looks like that need discussion, maybe somebody should mail the ubuntu-devel-discuss list about it?
[01:09] <sfllaw> seb128: Are you planning to revert?
[01:09] <seb128> revert what?
[01:10] <seb128> we didn't change anything yet
[01:10] <sfllaw> Oh, I thought you had uploaded something...
[01:10] <sfllaw> Never mind.
[01:19] <Keybuk> when you change your home IP range ... remember to update the ACL of your DNS server
[01:19] <StevenK> Hah
[01:19] <StevenK> Done that myself
[01:24] <zul> hey
[01:29] <pitti> hi zul
[01:34] <Hobbsee> heya pitti 
[01:34] <Hobbsee> :D
[01:35] <Hobbsee> pitti: requestsync needs to be upgraded not to break when a package isnt in the ubuntu archive yet. 
[01:35] <Hobbsee> (ie, when a package is new to ubuntu)
[01:35] <pitti> Hobbsee: specifying a '0' basever (3rd argument) doesn't work as a workaround?
[01:39] <Hobbsee> pitti: didnt try
[01:39] <Hobbsee> (where that's the old versoin)
[01:40] <pitti> Hobbsee: that's just some debug output of sync-source (in particular, the current overrides)
[01:40] <Hobbsee> pitti: oh right.  so it can be ignored.
[01:41] <Hobbsee> pitti: it seems not.  
[01:49] <fabbione> ogra: sorry do you have the bug number for the missing g-p-m file? it's not in the changelog
[01:50] <fabbione> (so i can mark duplicate etc. etc.)
[01:51] <heno> cjwatson: which package will you be uploading the braille script with, brltty, casper? I'e has dome updated work from Dave and would like to compare and sync
[01:51] <Mithrandir> Hobbsee: https://launchpad.net/ubuntu/+source/kimdaba/+bug/83162 ; a small reason would be nice.
[01:51] <Ubugtu> Malone bug 83162 in kimdaba "please remove kimdaba from the archive (feisty)" [Undecided,Unconfirmed]  
[01:52] <cjwatson> heno: brltty I think
[01:52] <cjwatson> I did a chunk of it on Friday but got stalled for a bit - I'll take it up again today or tomorrow
[01:53] <Hobbsee> Mithrandir: i'm positive i put one in there.
[01:53] <pochu> heno: I'm interested in the braille feature, is there any spec, or something?
[01:53] <Hobbsee> Mithrandir: oh bugger it
[01:53] <heno> cjwatson: ok, thanks. feel free to send me whatever you have by mail as well to look over
[01:53] <Hobbsee> Mithrandir: maybe only the original bug did.  done.  upstream renamed it
[01:54] <Mithrandir> Hobbsee: ok, what's the new name?
[01:54] <Hobbsee> Mithrandir: kphotoalbum
[01:54] <Mithrandir> thanks
[01:54] <heno> cjwatson: or better, yet: I'll just send you what Dave sent me
[01:54] <Hobbsee> Mithrandir: :)
[01:55] <heno> pochu: https://wiki.ubuntu.com/Accessibility/Specs/BrailleSupport
[01:55] <pochu> heno: thanks!
[02:15] <tkamppeter> I have a question about the uplaod of a new package and its availability. See https://launchpad.net/ubuntu/+source/m2300w/0.51-0ubuntu2
[02:16] <pochu> tkamppeter: #ubuntu-motu
[02:18] <cjwatson> pochu: it's fine here
[02:18] <vernes> so, this channel is only for anouncements?
[02:18] <tkamppeter> One can see on the left that m2300w was successfully built on all platforms, but on http://archive.ubuntu.com/ubuntu/pool/universe/m/m2300w/ only the source packages show up and not the binaries. How long will it take until the binary packages appear?
[02:18] <pochu> cjwatson: ok :)
[02:18] <cjwatson> vernes: no
[02:19] <cjwatson> vernes: this channel is for discussion and coordination among Ubuntu developers
[02:19] <vernes> ah, ok
[02:19] <cjwatson> tkamppeter: undetermined - it's in the NEW queue
[02:19] <Hobbsee> s/still//g
[02:19] <tkamppeter> The last message got broken, here it is again: One can see on the left that m2300w was successfully built on all platforms, but on http://archive.ubuntu.com/ubuntu/pool/universe/m/m2300w/ only the source packages show up and not the binaries.  How long will it take until the binary packages appear?
[02:20] <cjwatson> tkamppeter: you can see the NEW queue here: https://launchpad.net/ubuntu/feisty/+queue
[02:21] <cjwatson> tkamppeter: is the split of m2300w-ppds really worth it? that package is tiny
[02:21] <tkamppeter> Thanks, cjwatson, some weeks ago I asked here where I can see the NEW queue and no one answered me.
[02:22] <cjwatson> tkamppeter: there are some much bigger files in m2300w - if there's to be a split at all it should be those files that are split out, if they're architecture-independent; if they're not, I'd just put the lot in one binary package
[02:25] <tkamppeter> The split is because other printer driver packages were built in a similar way.
[02:26] <tkamppeter> But it seems that the package was in general accepted this way because the source was accepted.
[02:27] <tkamppeter> So for me it was strange that the binary was correctly built but not uploaded.
[02:34] <cjwatson> binary packages with new names are *always* held in the new queue
[02:35] <cjwatson> always always always
[02:35] <cjwatson> tkamppeter: the person processing the source probably didn't notice. I'm noticing now, and recommending that it be done differently
[02:35] <cjwatson> I'll accept the binaries anyway for now, but I recommend that you rethink the split
[02:36] <Mithrandir> or it wasn't split when the source was uploaded.  And it's hard to judge sizes when all you have is the source package.
[02:43] <Tonio_> pitti: ping ? maininclusionqueue show kde-style-polyester has been approved, but https://wiki.ubuntu.com/MainInclusionReportKdeStylePolyester doesn't...
[02:44] <Tonio_> pitti: is it really being proceed ? In case yes I will change the kubuntu-desktop seed
[02:44] <Riddell> Tonio_: pitti didn't handle it, iwj did
[02:45] <hunger> Is there any iptables-based firewalling script that still works on feisty?
[02:45] <Tonio_> Riddell: okay
[02:45] <Riddell> Tonio_: and I've added it to the seeds
[02:46] <Riddell> Tonio_: so now we go back to pitti to ask him to promote it :)
[02:49] <Nafallo> hunger: firestarter
[02:49] <hunger> Nafallo: firehol is still borked:-(
[02:49] <hunger> Nafallo: Have you tried that recently?
[02:50] <Nafallo> hunger: I run it atm.
[02:57] <pitti> iwj: right, if you approve a MIR, you need to change the wiki page, too
[02:59] <pitti> Tonio_: it has been promoted earlier by Mithrandir 
[02:59] <iwj> pitti: Err, didn't I ?
[03:00] <iwj> I meant to update both the page for the review itself and the queue page.
[03:00] <pitti> apparently not
[03:00] <iwj> Let me fix that.
[03:01] <iwj> I must have forgotten to save the page or something.
[03:01] <pitti> doko, seb128: I just synced ttf-dejavu
[03:01] <geser> Mithrandir: I've uploaded a new xmms2 which prints now the config.log if scons fails
[03:02] <geser> Mithrandir: but I can't make much sense from the build log: http://librarian.launchpad.net/6221835/buildlog_ubuntu-feisty-i386.xmms2_0.2DrHouse-3.1ubuntu2_FAILEDTOBUILD.txt.gz
[03:02] <Tonio_> pitti: yes Riddell told me thanks :) just the MIR page missed the info so I wanted to be sure :)
[03:04] <iwj> pitti: Fixed, sorry.
[03:04] <pitti> iwj: no worries, thanks
[03:06] <seb128> pitti: thank you
[03:10] <Chipzz> geser:'
[03:10] <Chipzz> gcc -o .sconf_temp/conftest_0.o -c -O2 -Wall -g .sconf_temp/conftest_0.c
[03:10] <Chipzz> /bin/sh: -c: line 0: syntax error near unexpected token `('
[03:10] <Chipzz> it's executing gcc, which appears to be a symlink to /bin/sh or something?
[03:11] <geser> Chipzz: check the next line in the log
[03:13] <Mithrandir> geser: where does that env call come from?
[03:13] <Chipzz> oh, right
[03:14] <geser> Mithrandir: I don't know
[03:14] <Mithrandir> it seems to be unhappy about the HASH(0x82db558)="" which has snuck into the environment
[03:15] <pitti> BenC: hm, it seems we get invalid core dumps pretty often now -- since the kernel already reaps the crashed process before writing to stdin is finished, may it be possible that parts of the core dump get lost?
[03:15] <Chipzz> heh
[03:15] <BenC> pitti: it's possible
[03:16] <BenC> pitti: I can look into the wait4 stuff I did before
[03:20] <doko> BenC: you did ping me yesterday
[03:20] <BenC> doko: You have an atheros wifi, right?
[03:20] <doko> BenC: yes, non-working :)
[03:20] <BenC> doko: Can you test the latest (6.4) lrm in feisty and tell me if it works?
[03:20] <tepsipakki> could someone sync nfs-utils from experimental?
[03:21] <pitti> tepsipakki: did you test the new version?
[03:21] <seb128> tepsipakki: if you tried it open a sync bug and get somebody confirming it
[03:22] <tepsipakki> pitti: hmm, actually no
[03:22] <tepsipakki> seb128: I was about to
[03:25] <cjwatson> FYI, "UbuntuSpec:launchpad-spec-name" now works on wiki.ubuntu.com
[03:25] <cjwatson> which saves all the [https://blueprints.launchpad.net/ubuntu/+spec/launchpad-spec-name launchpad-spec-name]  nonsense
[03:25] <pitti> yay
[03:25] <BenC> cjwatson: nice
[03:25] <seb128> nice
[03:26] <geser> Mithrandir: looks like the env call comes from Scons
[03:28] <tepsipakki> why is it that on my systems none of the scripts in /etc/network/if-up.d are run for eth*? I'm pretty sure it works for some
[03:30] <geser> Mithrandir: but I don't know where the HASH came from. does perhaps the buildd set it?
[03:30] <doko> BenC: it's not yet in the archive
[03:31] <BenC> It's built, maybe you can get it from launchpad
[03:31] <Mithrandir> it's from sbuild, it's a perl thingy.
[03:32] <cjwatson> looks like a stringified hash reference
[03:32] <cjwatson> $ perl -le '%foo = (); print \%foo'
[03:32] <cjwatson> HASH(0x100158b4)
[03:32] <BenC> doko: http://librarian.launchpad.net/6220771/linux-restricted-modules-2.6.20-6-generic_2.6.20.1-6.4_i386.deb
[03:33] <BenC> doko: or for amd64 http://librarian.launchpad.net/6220744/linux-restricted-modules-2.6.20-6-generic_2.6.20.1-6.4_amd64.deb
[03:33] <geser> and how do I avoid it?
[03:33] <cjwatson> get somebody to fix that sbuild bug? :)
[03:33] <cjwatson> probably infinity
[03:37] <tepsipakki> mvo: "software-properties-gtk has no installation candidate" ;)
[03:37] <tepsipakki> oh it's in the queue
[03:38] <bddebian> Heya
[03:41] <cjwatson> pitti: woo, thanks
[03:41] <pitti> cjwatson: thanks for your suggestions
[03:45] <seb128> * pitti sings 'Another one bites the dust' and marks increase-hwdb-participation as implemented
[03:45] <seb128> bddebian: usually reading a few lines back give you an hint
[03:45] <seb128> ;)
[03:46] <doko> BenC: seems to work
[03:47] <bddebian> seb128: I don't stay connected to keep a scrollback :)
[03:47] <pitti> seb128, ogra, bddebian: thanks :)
[03:47] <fabbione> mjg59: ping?
[03:47] <seb128> bddebian: that was the first line after you joined ...
[03:47] <seb128> bddebian: if you ignore the part line
[03:47] <ogra> fabbione, i duplicated the g-p-m bug, thanks for notifying ...
[03:48] <bddebian> Oh, heh, lost in the shuffle
[03:48] <pitti> nevermind, guys, that was really a trivial one
[03:48] <fabbione> ogra: thanks to you.. i just couldn't find the reference..
[03:48] <fabbione> ogra: i still asked Marc to verify it with a dist-upgrade.. to be extra sure
[03:48] <ogra> ok
[03:49] <BenC> doko: Cool, thanks
[03:50] <pitti> BenC: oh, if you touch the apport kernel bits: can you export the process' core file size ulimit in an env var?
[03:50] <BenC> pitti: Sure
[03:50] <pitti> BenC: Keybuk and I discussed that last Friday, and I got a complaint about not writing cores to people's cwd if ulimit > 0
[03:50] <pitti> BenC: (for non-packaged programs)
[03:50] <BenC> pitti: Makes sense
[03:51] <pitti> BenC: so while this dynamic decision seems a bit hackish, it's the best approximation of what we actually want; WDYT?
[04:04] <pitti> iwj: autofingers strike back :)
[04:04] <cjwatson> udebs were here first. :)
[04:05] <iwj> cjwatson: Can I ask you a question about udebs ?  I'm doing udev-device-mapper and it turns out that the udev scripts for dm devices are going to call a symlink to dmsetup.
[04:05] <iwj> (a) Is that relevant at all to udebs ?  (b) If so, is it OK ?  (c) If so, what should I do to the dependencies in the udebs to make sure dmsetup is installed if it's required ?
[04:06] <iwj> (Don't ask why it's a symlink to dmsetup; that's the way I found it.)
[04:26] <davmor2>  bug 83324 this bug has been partially resolved now in that hardware database item has been added to app>system tools  but I still think that the item needs to go in control center alongside hardware information.
[04:26] <Ubugtu> Malone bug 83324 in hwdb-client "hardware datatbase is requested but in wrong location" [High,Confirmed]  https://launchpad.net/bugs/83324
[04:29] <pitti> ogra: ^ that spec didn't talk about *where* to add the menu item, so I just took whatever already was in the .desktop
[04:30] <ogra> pitti, which is fine imho ... i start to hate the control-center more and more every day ...
[04:30] <bddebian> heh
[04:30] <ogra> it slows me down a lot by just waiting for a grouping window to come up instead of providing direct access to the function i want ....
[04:32] <chrisj> cjwatson:Ping, about the partitioner re-write
[04:32] <ogra> but since i have a general objection against that usability drawback, i'm probably the wrong person to judge at all
[04:32] <mvo> can we get a hwdb-client that can be translated?
[04:32] <davmor2> I only suggested that it go into control panel as it is already in the hardware information app which is in the control center
[04:32] <ogra> mvo, i thought pitti fixed that last upload
[04:32] <cjwatson> iwj: (a) yes, though it would be worth running through partman's LVM and MD support afterwards to make sure it still works; (b) yes, sounds fine; (c) Depends: dmsetup-udeb
[04:32] <cjwatson> chrisj: i
[04:33] <cjwatson> hi
[04:33] <ogra> at least there was a lot in the changelog
[04:33] <cjwatson> (damn keyboard)
[04:33] <pitti> mvo, ogra: no, I fixed the handling of the po/ files, I didn't actually fix any code
[04:33] <pitti> ogra, mvo: unless the reason was a missing .pot file; I fixed that
[04:34] <mvo> pitti: unfortunately that is not enough
[04:36] <chrisj> cjwatson: heno said to ping you about lending a hand.  Has anything be done yet or is it a blank slate? 
[04:38] <cjwatson> chrisj: taking to /query
[04:39] <iwj> cjwatson: Great, thanks.
[04:40] <iwj> I think some of partman's lvm support may become obsolete, since things will get activated automatically.
[04:51] <DaSkreech> Hello
[04:51] <DaSkreech> Why is there no more progress bar on the Bootsplash?
[04:53] <cjwatson> iwj: a little, but probably not lots. Most of it is keeping track of what's there vs. what should be there.
[04:53] <cjwatson> having VGs be brought up automatically will be interesting
[04:53] <Nafallo> DaSkreech: atleast you got a bootsplash... :-)
[04:54] <DaSkreech> Nafallo: Yours is broked? :)
[04:54] <Nafallo> wonders even
[04:54] <DaSkreech> The progress bar?
[04:54] <cjwatson> iwj: one thing to make sure of is that it's possible to deactivate and delete PVs etc. in partman
[04:54] <cjwatson> without them coming straight back up again
[04:54] <Nafallo> DaSkreech: no. my bootsplash :-)
[04:55] <cjwatson> DaSkreech: erm, I don't believe it was intentionally removed ...
[04:55] <DaSkreech> cjwatson: It's very very distracting
[04:55] <cjwatson> if it's gone, it's some kind of accidenet
[04:55] <cjwatson> accident
[04:55] <cjwatson> how can the lack of something be distracting?
[04:56] <DaSkreech> without that and kernel messages it's impossible to tell if the OS is actually booting
[04:56] <cjwatson> that's not "distracting", it's "annoying" or something
[04:56] <DaSkreech> after 10-12 minutes you just kind of guess nothign is happeneing
[04:56] <cjwatson> it's a bug, file it
[04:56] <DaSkreech> no it's distracting
[04:56] <cjwatson> I guess this is distracting as my stepson uses the word :-)
[04:57] <DaSkreech>  I used to be able to turn on the computer and just kind of glance at it to see messages streaming past
[04:57] <DaSkreech>  now I HAVE to sit and watch the screen
[04:57] <cjwatson> the messages were intentionally removed, but the progress bar wasn't
[04:57] <cjwatson> it should still be there
[04:57] <DaSkreech> I know but I can jimmy the messages back in :)
[04:57] <DaSkreech> I don't know how to restore the progress bar
[04:57] <DaSkreech>  Is there a way?
[04:57] <cjwatson> it's not supposed to be gone. *file a bug*
[04:57] <iwj> cjwatson: Right.
[04:58] <cjwatson> also it shouldn't take 10-12 minutes for the computer to boot; that in itself is a problem
[04:58] <DaSkreech> I'm doing that now
[04:58] <iwj> cjwatson: We're only going to be running vgchange on block device addition, not removal, so we should be OK.
[04:58] <DaSkreech> cjwatson: Right but I don't know it's a problem till 10-12 minutes have gone
[04:58] <DaSkreech> cause there is no progress bar
[04:59] <DaSkreech> I can't tell how far it's gone or without watching the clock how long this has been going on
[04:59] <cjwatson> do you have a custom kernel or something?
[04:59] <DaSkreech> no Herd Live CD
[05:00] <DaSkreech> so a) it's a new kernel and b) it's a live cd
[05:00] <cjwatson> well, it worked fine in testing, AFAIK
[05:00] <DaSkreech> Yeah I find random computers simply won't boot from a Ubuntu live CD for a release for soem reason
[05:00] <cjwatson> so since it's a bug, nobody can answer your question of "how do I get the progress bar back?" without diagnosing and fixing the bug
[05:01] <DaSkreech>  some will throw a kernel panic as soon as you hit enter some pretend to boot then just stop doing anything
[05:01] <DaSkreech> Of course without a progress bar it's much harder to figure that out
[05:02] <DaSkreech> cjwatson: ok I was just wondering if there was some hidden option like foolnewbsintothinkingitswindowsxp
[05:02] <cjwatson> no.] 
[05:02] <DaSkreech> :-)
[05:02] <DaSkreech> Good to know
[05:02] <cjwatson> press F6 at gfxboot and delete the "quiet splash" boot arguments
[05:02] <cjwatson> then try booting with that
[05:02] <Nafallo> ehrm... dudes. should I have /etc/usplash.conf on amd64? :-)
[05:02] <cjwatson> if it's kernel-panicking, you'll see that
[05:03] <cjwatson> DaSkreech: if it panics before the progress bar is set up, then that would explain it
[05:03] <cjwatson> DaSkreech: in that case, please file a bug on linux-source-2.6.20 with the best transcription of the panic that you can manage
[05:03] <DaSkreech> No if it panics on this machine the progress bar goes away and I get Scary Messages (c)
[05:03] <DaSkreech> But this machine actually boots fine
[05:04] <DaSkreech>  It's just annoying that I have to be distracted by a boot process
[05:04] <DaSkreech> When for many years it was done so well
[05:04] <pochu> pitti: I think you should look at this bug: https://launchpad.net/bugs/83324
[05:04] <Ubugtu> Malone bug 83324 in hwdb-client "hardware datatbase is requested but in wrong location" [High,Confirmed]  
[05:05] <pochu> pitti: it's "fixed", however it could be better ;) but feel free to close it if you think it's done
[05:05] <cjwatson> hmm, https://launchpad.net/ubuntu/+source/usplash/+bug/81413 suggests some unzeroed struct members or somethin
[05:06] <Ubugtu> Malone bug 81413 in usplash "[Feisty]  no usplash" [Undecided,Unconfirmed]  
[05:06] <cjwatson> g
[05:06] <iwj> cjwatson: So to test out my crazy new udebs I should drop them in build/localudebs of my unpacked and ready-to-build-an-image d-i tree ?
[05:06] <iwj> Is there an easier way that doesn't involve writing huge cds ?
[05:06] <seb128> pitti: ok, using an old linux version make apport works fine again
[05:07] <seb128> pitti: BTW were is the ignore list and the package crashes counter for apport?
[05:07] <seb128> I want to reset some of them ;)
[05:07] <pitti> seb128: I guess it's the piping that crops the coredump
[05:07] <pitti> seb128: ~/.apport-ignore.xml
[05:07] <seb128> ah, thank you
[05:07] <pitti> or just sudo touch the executable :)
[05:08] <seb128> and there is a "that package already got 3 crash reports today" counter too, no?
[05:08] <Nafallo> thanks cjwatson :-)
[05:09] <tepsipakki> ok, nfs-utils from Debian experimental works fine, bug #83435
[05:09] <Ubugtu> Malone bug 83435 in nfs-utils "please sync nfs-utils (main) from Debian experimental" [Undecided,Unconfirmed]  https://launchpad.net/bugs/83435
[05:09] <tepsipakki> it is a newer snapshot which will (hopefully) soon become 1.0.11
[05:13] <cjwatson> iwj: if you can arrange not to need any of this until you reach the stage of the installer when networking is brought up, you can wget the udebs and udpkg -i them on the fly
[05:13] <cjwatson> iwj: alternatively, you may be able to nuke all the .debs from pool/, which would make writing CDs quicker
[05:14] <cjwatson> (presumably you don't need to go as far as installing the base system)
[05:14] <Nafallo> usplash came back when I removed /etc/usplash.conf :-)
[05:14] <seb128> pitti: and there is a "that package already got 3 crash reports today" counter too, no?
[05:14] <Nafallo> we are not trying to make that pretty for feisty are we? :-)
[05:17] <cjwatson> yes
[05:17] <cjwatson> it just needs ported to new libx86
[05:18] <Nafallo> so not supposed to work now then :-)
[05:18] <Nafallo> good
[05:20] <iwj> cjwatson: This is for the root fs so I'd have to nobble it not to look at any of that, and then I'd have to tediously retype my dpkg -i's when I got it wrong.
[05:20] <iwj> But your suggestion of nuking .debs sounds work a try.
[05:27] <mjg59> fabbione: Hi
[05:29] <cjwatson> iwj: can't be for the rootfs in the installer surely :)
[05:31] <pitti> seb128: that counter is in the report itself, just delete it
[05:32] <iwj> cjwatson: Well, indeed :-).  But doesn't it go and detect hard disks quite early ?
[05:32] <seb128> pitti: ok, so that's not it (I've clean /var/crash), killall -11 rhythmbox produce no crash for whatever reason
[05:32] <iwj> Anyway this is all a bit moot because atm my main initrd generator is evidently broken ...
[05:34] <cjwatson> iwj: yeah, but you could probably nobble it not to ...
[05:34] <cjwatson> iwj: BOOT_DEBUG=3 is often useful if you need to edit stuff before udev starts
[05:34] <iwj> Thanks.  In this case it fails to run udev for some reason.
[05:48] <iwj> I've obviously done something really freaky to this install.  udev is failing to create /dev/{mem,tty0,...}
[05:54] <DaSkreech> cjwatson: https://launchpad.net/ubuntu/+bug/83438
[05:54] <Ubugtu> Malone bug 83438 in Ubuntu "No Progress bar on bootsplash" [Undecided,Unconfirmed]  
[05:55] <fabbione> mjg59: hey.. sorry to bother you.. do you have an idea of what could cause bug #83078 ?
[05:55] <Ubugtu> Malone bug 83078 in Ubuntu "Hibernation on Feisty herd 3 doesn't wake for Sun Ultra 20 Work" [Undecided,Needs info]  https://launchpad.net/bugs/83078
[05:56] <DaSkreech> Thanks
[05:56] <mjg59> Is ultra 20 x86?
[05:57] <mjg59> If so, no, that gives me no useful information
[05:57] <mjg59> Get them to boot without splash quiet
[05:57] <fabbione> mjg59: yes it's x86
[05:58] <fabbione> mjg59: ok thanks a lot..
[06:00] <tkamppeter> pitti, ping
[06:13] <iwj> Grrr.  Every time you do an install it changes the uuid of the swap partition all your other installs are using.
[06:15] <cjwatson> should be fixable by telling it not to format swap
[06:15] <cjwatson> which IIRC is an option these days
[06:15] <wasabi> udev rules question. Trying to figure out how to make evms_activate when ANY block device comes and goes
[06:15] <wasabi> The rules lines are weird. Weird language. Looks likea filters + actions.
[06:15] <fabbione> cjwatson: only if you do manual partitioning AFAIK.
[06:15] <iwj> cjwatson: But if you install an earlier version it breaks your most recent versions.
[06:15] <wasabi> If I just remove the filter, for KERNEL=="dm-*" would that be good enough?
[06:16] <iwj> I think we're pretty crappy at coexistence given that we can't even manage it with ourselves, which makes it hard to justify complaints about M$ antics.
[06:17] <cjwatson> iwj: maybe I need more coffee, but I can't parse your second-last statement
[06:18] <cjwatson> fabbione: if you aren't doing manual partitioning, it won't reuse the existing swap partition
[06:18] <wasabi> So, um. Is it reasonable at this point for somebody (other than me, who can't upload to main) to fix evms? 70-evms.rules should be included in the initramfs, and the udev rule should run on ALL block devices.
[06:18] <iwj> cjwatson: I mean, if you have a working feisty install and then you install dapper, dapper's installer will break your feisty's swap.
[06:18] <fabbione> cjwatson: hmm ok.. i might remember it wrong then.
[06:19] <cjwatson> I guess dapper didn't have the option not to format swap
[06:19] <iwj> wasabi: Err, it's probably best not to touch that right now.  I'm doing the same thing for lvm2 now and I'll do evms when I've got it working.
[06:19] <wasabi> I still wonder why we use swap partitions.
[06:19] <wasabi> I was sort of fond of swapd before I realized it was broken.
[06:20] <iwj> cjwatson: Which means that when we made dapper we thought it was OK to randomly change the uuid of a swap partition and now we think it isn't.  And this isn't the only example of this kind of thing.
[06:20] <iwj> wasabi: There are two rules in 70-evms.rules, aren't there ?
[06:20] <wasabi> Yes. One for dm-*
[06:20] <wasabi> and... um. Some kernel ENV thing
[06:20] <cjwatson> no, when we were making dapper we didn't think about it at all
[06:21] <wasabi> which I just nuked
[06:21] <fabbione> iwj: oh btw... i found a way to get rid of that script that creates the symlinks to lvm in initramfs
[06:21] <cjwatson> I mean, I take your point, but there's not much I can do about dapper
[06:21] <fabbione> iwj: just prefix the calls with lvm
[06:21] <wasabi> Except make a dapper service pack. :0
[06:21] <fabbione> iwj: vgscan -> lvm vgscan
[06:21] <cjwatson> I think we should probably fix feisty so that when it formats swap it extracts the prior UUID of the swap partition and reinstates it after formatting
[06:21] <cjwatson> so that even if you format by mistake it doesn't trash the UUID
[06:22] <iwj> cjwatson: That would be a way for us not to break the promise we made to ourselves in dapper.
[06:22] <cjwatson> in reality, formatting a swap partition (that doesn't have a hibernate signature in it) is just a much faster way of checking it :-)
[06:22] <iwj> fabbione: Oh, hmm.  I'm making the symlinks in a hook script now anyway.
[06:22] <wasabi> Real question though... why swap partitions?
[06:22] <iwj> It's so totally lame that you can't put symlinks in the initramfs!
[06:22] <fabbione> iwj: yes and AFAIK it was the only reason why we were shipping the hook...
[06:22] <fabbione> iwj: so we can just get rid of it by calling lvm $command
[06:22] <iwj> wasabi: Err, are you finding that the trigger isn't working when it should ?
[06:22] <cjwatson> which is in fact why partman started formatting swap by default - checking the swap partition was really slow
[06:23] <wasabi> iwj: doesn't work at all during the initramfs.
[06:23] <wasabi> I have to break and run evms_activate manually.
[06:23] <wasabi> Also, it's not even included. ;)
[06:23] <iwj> wasabi: Yes, but that's nothing to do with the dm-* condition on the second rule.
[06:23] <wasabi> I need to restore that file one sec
[06:23] <iwj> See https://wiki.ubuntu.com/UdevEvms, last section `Notes from the actual implementation'.
[06:24] <lemsx1> is anybody working on the new tzdata file for Dapper/Edgy ?
[06:24] <iwj> wasabi: So yes I'm going to fix the initramfs for evms but I think the udev rules are right.
[06:24] <cjwatson> wasabi: last I checked (a) swap partitions were still more efficient than swap files (b) partman has absolutely no clue of how to set up swap files so it would be a chunk of work to do even if we did want to switch over
[06:25] <iwj> Also if you use swap partitions you can't share them if you dual boot which is actually useful.
[06:25] <wasabi> Hmm. swapd is neat.
[06:25] <wasabi> It's a daemon that auto creates swap files when swap space runs low.
[06:25] <lemsx1> This is very important for us on the East coast of the US: http://wiki.debian.org/TimeZoneChanges and http://packages.qa.debian.org/t/tzdata.html
[06:25] <wasabi> It could be fixed/tuned. And then you could have a nice "Ubuntu is running low on swap space and is increasing it to meet demand" tip like windows.
[06:25] <wasabi> or whatever.
[06:25] <iwj> wasabi: I don't have time now to argue with you about swapd but it's a hideously bad idea.
[06:25] <wasabi> heh. Just want to know why.
[06:26] <wasabi> iwj: I'm looking at the evms rules now. Are you saying the first rule should cause evms to run?
[06:26] <iwj> wasabi: Do you still think the evms udev rule is wrong ?
[06:26] <wasabi> ENV{ID_FS_USAGE}?
[06:26] <iwj> Yes.
[06:26] <wasabi> I need to know what that means. ;)
[06:27] <cjwatson> iwj: apparently the use of cpio --dereference is to save space while building the image - it creates a big symlink farm to the stuff on the system where you're running mkinitramfs and then cpio dereferences them on the fly
[06:27] <wasabi> Ahh. I see. vol_id stuff.
[06:28] <iwj> cjwatson: Yes.
[06:28] <iwj> cjwatson: Really we want   cpio --dereference-only-absolute-symlinks  or something.
[06:28] <wasabi> iwj: Probably works for things where your block device is in fact for raid. But pure LVM?
[06:28] <wasabi> Hmm. Guess that would be dm
[06:29] <iwj> root@samual8:~# /lib/udev/vol_id /dev/hda13
[06:29] <iwj> ID_FS_USAGE=raid
[06:29] <iwj> ID_FS_TYPE=LVM2_member
[06:31] <wasabi> okay, cool. It probalby works... just needs to be in the initramfs. I don't really have anything to test it with that doesn't get handled during the initramfs
[06:33] <iwj> wasabi: Right.
[06:33] <iwj> Damn, debugging a system where udev doesn't start is a PITA.
[06:51] <cjwatson> iwj: swap UUIDs> fix in my local partman-basicfilesystems tree now, although I'll need to test it
[06:51] <iwj> cjwatson: Excellent.
[06:52] <iwj> cjwatson: Would you have any useful tips for a system where udevd doesn't seem to run in the initramfs for some reason ?
[06:52] <cjwatson> obviously dapper installs will still confuse feisty, but I can't do anything about that now
[06:52] <iwj> cjwatson: Yers.
[06:52] <cjwatson> iwj: I'd try cranking udev_log up to debug in /etc/udev/udev.conf
[06:52] <cjwatson> to see if it's being started at all
[06:52] <iwj> Does that help in the initramfs ?
[06:52] <cjwatson> other than that I think I'd be inclined to break=top and step through the initramfs by hand
[06:53] <cjwatson> provided you can get it in place early enough, should do
[06:54] <Adri2000> any archive admin: please give back gaupol (i386, it's an architecture: all)
[06:54] <cjwatson> I think it just prints to stdout
[06:54] <cjwatson> Adri2000: you need somebody in launchpad-buildd-admins for that (== "buildd admin"), not somebody in ubuntu-archive (== "archive admin")
[06:54] <iwj> It runs fine if I say `udevd&' from the initramfs shell.
[06:55] <Adri2000> cjwatson: ah, I didn't know
[06:55] <cjwatson> iwj: can you tell if it's running in the initramfs? (break=bottom and 'ps')
[06:55] <Adri2000> Mithrandir: please give back gaupol (i386, it's an architecture: all)
[06:55] <Kano> hi, did you patch mount to support UUID for ntfs?
[06:56] <iwj> cjwatson: It bombs out due to not finding the root fs and then in that shell it's not running, but starts fine when I run it by hand.
[06:57] <Kano> i dont get why this does not work on pure debian
[06:59] <iwj> cjwatson: Oh, I see it now.
[06:59] <iwj> cjwatson: BTW, YM debug=y, not BOOT_DEBUG=3
[06:59] <iwj> (I think)
[06:59] <cjwatson> iwj: in d-i it's BOOT_DEBUG=3
[07:00] <cjwatson> which is what I thought you were talking about
[07:00] <iwj> Ahhh.
[07:00] <cjwatson> perhaps we were at cross-purposes
[07:02] <Kano> is Scott James Remnant here?
[07:02] <Keybuk> yup
[07:02] <Kano> is it you?
[07:02] <Keybuk> it is
[07:02] <cjwatson> Kano: /whois
[07:03] <Kano> well do you know about that UUID thing for Ntfs
[07:03] <Keybuk> you'll have to be more specific, I'm afraid
[07:03] <somerville32> Mithrandir, Can we get herd 3 released for Xubuntu?
[07:03] <Kano> well on ubuntu UUID=... can be used for ntfs and on debian not
[07:04] <Keybuk> right... ?
[07:05] <Kano> what did you change
[07:05] <Kano> in the fstab
[07:05] <Kano> ubuntu usually uses UUID instead of /dev/XXX entries
[07:05] <Keybuk> yes, it does
[07:05] <Kano> but when i use the same entries on etch, i can not mount ntfs
[07:06] <Keybuk> ahh
[07:06] <Keybuk> see, if you'd started this with "Why can Ubuntu mount NTFS partitions by UUID, when Debian can't?" I would have been able to answer you instantly
[07:06] <Kano> so tell me why
[07:07] <Keybuk> the answer is that Ubuntu's util-linux package contains a patch from SuSE to use libvolumeid for UUID detection, instead of libblkid
[07:07] <Kano> well i found a similar entry in the mount changelog
[07:07] <Keybuk> libvolumeid comes with udev and is actively maintained, and supports far more filesystems than libblkid which comes with e2fsprogs and isn't actively maintained
[07:08] <Keybuk> the one with my name on it, I assume?
[07:08] <Kano> yes
[07:08] <Keybuk>   * Steal the libvolume_id-support patch from SuSE and apply here so that
[07:08] <Keybuk>     we use libvolume_id for accessing devices via UUID and/or LABEL instead
[07:08] <Keybuk>     of the less-capable blkid.
[07:08] <Keybuk>  -- Scott James Remnant <scott@ubuntu.com>  Mon, 21 Aug 2006 15:23:58 +0200
[07:08] <Keybuk> ironically
[07:08] <Keybuk> the changelog entry you refer to appears to answer your question for you :)
[07:08] <Kano> well why did n't you tell debian about it?
[07:09] <Treenaks> Debian is currently in a very deep freeze
[07:09] <Keybuk> because writing over 1,000 individual e-mails to each Debian developer would take a long time?
[07:09] <Kano> thats an important change
[07:09] <jamesh> Kano: you could ask SUSE why they didn't tell Debian about it either
[07:09] <Keybuk> the Debian maintainer of util-linux is well aware of the patch
[07:09] <Keybuk> aren't you lamont ? :)
[07:10] <Keybuk> you'd be better off asking the Debian maintainer why he hasn't applied the patch to Ubuntu
[07:10] <Keybuk> err
[07:10] <Keybuk> s/Ubuntu/Debian/
[07:11] <Keybuk> I suspect you'll get the entirely sensible and legitimate answer that udev isn't essential in Debian
[07:11] <Keybuk> so an essential thing like mount can't depend on it
[07:11] <Keybuk> (udev is part of a minimal Ubuntu installation, so we *can* depend on it)
[07:12] <Treenaks> Keybuk: how.. Debian..
[07:12] <Kano> Keybuk: why could the patch not use volid first and if not there use old code?
[07:12] <Keybuk> it could
[07:13] <Keybuk> go write it like that :p
[07:13] <Keybuk> there's no point Ubuntu spending time doing that, because udev is in our base install, so we don't need that patch
[07:14] <Keybuk> (where "that patch" = "the clever use volid-or-blkid one"
[07:14] <Keybuk> of course
[07:14] <Keybuk> using dlopen() inside mount might make people run away
[07:14] <Keybuk> (unless you know of a clever ELF hack to load alternate .so files?)
[07:14] <kylem> LD_PRELOAD mount.
[07:14] <kylem> muahahahahahah.
[07:14] <kylem> that's not a root hole waiting to happen.
[07:15] <Keybuk> please don't suggest that to Lamont
[07:15] <Keybuk> he might get ... carried away
[07:16] <Kano> Keybuk: the funny thing is: suse is not even using UUIDs in 10.2's fstab
[07:17] <Keybuk> they're going to
[07:17] <Keybuk> pretty much everyone is, as it's the easiest way to deal with /dev/hda1 suddenly becoming /dev/sda3
[07:18] <keescook> mornin'
[07:18] <Kano> Keybuk: i want to use it too...
[07:18] <kylem> morning kees.
[07:18] <keescook> hiya kylem 
[07:19] <iwj> OMG
[07:19] <iwj> + mknod /dev/root b 253 0
[07:19] <iwj> + ROOT=/dev/root
[07:19] <iwj> (253 is devmapper)
[07:19] <Keybuk> iwj: no 50p in the meter?
[07:19] <pkl_> i.e. when the ide drivers move over to the new serial ide infrastructure.
[07:20] <Keybuk> Kano: grab the util-linux source package from Ubuntu, copy the appropriate patch (I think it's just libvolumeid-support or something) out of the debian directory into the Kanotix package, profit
[07:20] <Keybuk> Share And Enjoy
[07:21] <iwj> Time for one last attempt before I go to dinner.
[07:21] <Kano> will just recompile yours in case the version is higher
[07:21] <Keybuk> our util-linux package also carries a different hwclock script
[07:21] <Keybuk> the feisty util-linux package is the same base version as the unstable one
[07:21] <Kano> is that essential?
[07:21] <Keybuk> no, it's just because of our different boot process in edgy
[07:22] <Kano> so better fetch only that patch from it?
[07:22] <Keybuk> yeah
[07:22] <Kano> ok
[07:24] <Keybuk> those are the only two, afaik
[07:24] <Keybuk> lamont touched it last
[07:25] <Kano> libvolume-id-dev 
[07:25] <Kano> that to use for debian
[07:26] <Kano> hmm looks that is exists too..
[07:26] <Kano> just notthing with ubuntu...
[07:27] <Keybuk> hmm?
[07:28] <Kano> libvolume-id-d 0.103-2 
[07:28] <Kano> libvolume-id-dev (>= 093-0ubuntu11),
[07:30] <Kano> >=0.93 then?
[07:34] <Keybuk> whatever matches
[07:39] <somerville32> sladen, ping
[07:44] <Kano> Keybuk: do you know ntfs-3g
[07:44] <Kano> Keybuk: the patch works for ntfs
[07:44] <Kano> but not for ntfs-3g
[07:46] <Kano> Keybuk: i guess, because /proc/mount shows /dev/disk/by-uuid/XXX
[07:46] <Kano> but output of mount is "fixed" to show the real device
[07:46] <Kano> just not for ntfs-3g
[07:47] <Kano> (type fuse)
[07:51] <infu> hello
[07:51] <pochu> hi infu!
[07:53] <infu> excuse me...this channel is only for ubuntu's developers?
[07:53] <pochu> infu: not only, but mostly :)
[07:53] <infu> ah ok
[07:53] <infu> =)
[07:54] <infu> are there some employee of canonical?
[07:54] <somerville32> I'm sure there are
[07:55] <infu> 'cause after university
[07:55] <infu> i would like to work as developers for free software
[07:56] <infu> and canonical or red hat i think are a great point
[07:57] <cjwatson> Keybuk: our patch doesn't require udev, of course, just libvolumeid - I'm sure that could be borged into Debian's base system without a problem, post-etch
[07:57] <pochu> infu: you can start helping now without being employed ;)
[07:57] <infu> sure
[07:57] <cjwatson> infu: you're welcome to watch out for positions that interest you on http://www.ubuntu.com/employment
[07:58] <infu> yes i'd read that document
[07:59] <infu> it's "simple" can work for canonical?
[07:59] <cjwatson> the majority of our contributors aren't employed by Canonical, though, which is a good thing :)
[07:59] <infu> became an employee
[08:00] <cjwatson> our job interview processes are much like those of many other companies, except for being generally conducted by phone out of necessity
[08:00] <infu> yes i know but there are also some employee on canonical or red hat or novell or oracle
[08:00] <infu> oh ok
[08:03] <sladen> somerville32: yo, !justask
[08:04] <somerville32> sladen: How is the writeup coming for the Oslo weekend?
[08:05] <sladen> somerville32: haven't started, when's it needed for
[08:05] <somerville32> sladen: Today
[08:05] <sladen> somerville32: sounds good
[08:33] <crimsun> cjwatson: BenC has merged it into ubuntu-2.6.git, should be in the next kernel
[08:36] <cjwatson> crimsun: what *is* it, though? :)
[08:36] <crimsun> cjwatson: oh, right. Patches to make your audio work.
[08:36] <cjwatson> aha
[08:36] <cjwatson> sure, I can test that
[09:20] <lamont> jamesh: the libvolumeid patch is going in right after etch
[09:21] <lamont> Keybuk: we'll have to chat sometime about udev and hwclock and all that fun stuff so that I can understand what ubuntu did to my poor little package
[09:27] <LaserJock> mdke: is it ok if we make it clearer on SyncRequestProcess who non-MOTU people should subscribe (ubuntu-universe-sponsors in case of Universe packages, ubuntu-sponsors for Main?)
[09:27] <LaserJock> mdke: sorry that was for mdz
[09:27] <LaserJock> but he's not here, darnit
[09:29] <Kano> why is gspca not in the kernel
[09:31] <mez> hmm - in main, something isnt meant to link against something in universe - right ?
[09:32] <mjg59> Kano: Because it's crack?
[09:32] <kylem> society for the prevention of cruelty to animals?
[09:32] <Kano> it works
[09:32] <mez> for some reason, installing xorg-driver-fglrx just pulled in gfortran from universe
[09:33] <LaserJock> mez: because Fortran is awesome ;-)
[09:33] <Kano> mez: too funny
[09:33] <Kano> mez: for fglrx i would prefer using my script ;)
[09:33] <mjg59> Kano: Reading it makes me vigerously unhappy
[09:34] <mez> it just fails on the modprobe :(
[10:10] <bluefox_> o.o buki
[10:18] <geser> !seen infinity 
[10:19] <geser> has somebody seen infinity here lately?
[10:44] <davmor2> Query:  In ubiquity you were told in advance of partitioning that there was windows or whatever installed, use remaining space for install?  At what point does it do it in the new partitioner?
[10:50] <bddebian> cjwatson: I know you are swamped.  Are you ever going to touch kbd-chooser?
[10:59] <Mez> is there an easy way to find out what packages B-D on any package in a source package
[10:59] <Kano> dpkg -S file
[11:00] <Kano> apt-get source result
[11:01] <Mez> Kano, thats not helpful ;)
[11:01] <slomo> apt-cache showsrc $package
[11:02] <Mez> slomo, I know of that, but I want to find out if ANY package B-Ds on a package
[11:02] <Mez> like apt-cache rdepends
[11:02] <Kano> build-deps
[11:02] <Kano> thats easy
[11:02] <Kano> apt-get build-dep package
[11:02] <slomo> Mez: ah... that's not that easy... grep-dctrl on the Sources files ;)
[11:03] <Mez> Kano, reverse build-deps
[11:03] <slomo> Mez: grep-dctrl -FBuild-Depends -FBuild-Depends-Indep libglib2.0-dev /var/lib/apt/lists/*_Sources | grep ^Package
[11:03] <slomo> for example
[11:04] <Kano> Mez: i just use pbulder
[11:04] <Kano> but ok...
[11:05] <Mez> Kano, I'm trying to basically code in regression testing for prevu
[11:05] <Mez> so, it'll build something, install it, then go in and build everything that B-D's on it against it 
[11:06] <Mez> see if they break
[11:06] <slomo> then hope that nobody uploads dpkg-dev ;)
[11:07] <Kano> Mez: i use a reverse pbuilder for that
[11:07] <Mez> slomo, ?? 
[11:07] <Kano> that uses the already build debs in higher priority
[11:07] <Mez> Kano, reverse pbuilder ?
[11:07] <Kano> basically i call pbuilder
[11:08] <Kano> but while running i use apt-ftparchive on the results dir
[11:08] <Kano> and then alll later runs will use that dir
[11:08] <Mez> Kano, yeah, we have that in my branch of prevu
[11:08] <Kano> well i have it in a script
[11:08] <Kano> adopted from a website
[11:09] <Kano> http://kanotix.com/files/fix/pbuilder-adv.sh
[11:09] <Kano> just use the commented feisty settings
[11:09] <Kano> defaults to etch
[11:09] <Kano> then pbuilder-adv.sh clean
[11:09] <Kano> first
[11:10] <Kano> pbuilder-adv.sh build file.dsc
[11:10] <Kano> later
[11:10] <Kano> thats cleaner als compiling in your standard environment
[11:11] <Mez> Kano, you dont know of prevu do you
[11:11] <Kano> no i dont
[11:12] <Kano> but that works for me, when i want to build vdr or something 
[11:15] <LaserJock> Mez: where is your branch of prevu?
[11:15] <Mez> LaserJock /scratch/development/prevu/hooksnstuff/
[11:21] <LaserJock> Mez: is https://code.launchpad.net/~mez/+branch/prevu/prevu a good branch?
[11:21] <Mez> it's not the best ;)
[11:21] <Mez> probably a bit old
[11:22] <Mez> https://code.launchpad.net/~ubuntu-backporters/+branch/prevu/dev
[11:42] <cjwatson> bddebian: sometime this week, I guess
[11:43] <bddebian> cjwatson: I wouldn't mind doing it for ya but man it's messy :-(
[11:43] <cjwatson> there's no sense wasting your time on it
[11:44] <cjwatson> the only possible reason to merge it is to improve the statistics
[11:44] <cjwatson> and so it might as well be left to somebody who already understands it :)
[11:47] <bddebian> cjwatson: Just trying to clean up the list. :-)
[11:48] <vdepizzol> what composite-by-default spec with delivery "Deferred" means exactly?
[11:52] <mdke> vdepizzol: the question on everybody's lips
[11:52] <mdke> otherwise known as the "how the heck can you get to see spec change history with launchpad" question
[11:53] <cjwatson> there'll be an announcement soon; I don't really want to pre-empt it
[11:54] <cjwatson> but basically, means "won't be done in this form for this release"
[11:54] <Mez> ne1 for a game of frozen-bubble2?