[12:04] <pitti> night everybody
[12:05] <dholbach> good night pitti 
[12:12] <sabdfl> night all
[12:14] <jon1012> good night everybody :)
[12:15] <dholbach> i'm off too
[12:15] <dholbach> good night
[12:19] <doko> mithrandir: ping?
[12:26] <Kamion> HiddenWolf: shrug, I test i386 images on my amd64 at least half the time ;)
[12:27] <Kamion> they're quicker to burn than the amd64 images, and the amd64 is quicker at testing them ...
[12:27] <zenwhen> whats the name of the default theme used for KDE in Hoary?
[12:28] <Riddell> zenwhen: plastik style, crystal icons
[12:28] <zenwhen> I assume it is the one I see k3b using.
[12:28] <Riddell> zenwhen: if it's horrible then that's keramik from kde 3.3
[12:28] <zenwhen> it doesnt look half bad for a kde theme
[12:28] <Riddell> must be plastik then
[12:30] <HiddenWolf> kamion: what i ment was; intel will get those x64 celerons out, and after that, i'll be watching to see how long it takes for 64bit to mature
[12:32] <zenwhen> I am buying Intel's first dual core chip.
[12:32] <HiddenWolf> I want to see that amd64 4200 with sse3 et all. :)
[12:33] <HiddenWolf> btw, media is saying dual core won't be good on the desktop for the forseeable future, how is that under linux? it seems better suited to it.
[12:35] <shaya> has anyone ever ever configure a linux fiber san?
[12:42] <HiddenWolf> Anyone here?
[12:42] <mdz> Kamion: why quicker?  the i386 and amd64 ISOs are very nearly the same size
[12:43] <HiddenWolf> mdz, can you check something for me?
[12:43] <seb128> mdz: are we rolling the preview iso now ? 
[12:43] <seb128> mdz: or these are pre-preview isos ?
[12:43] <mdz> seb128: pre-preview
[12:43] <mdz> seb128: why, is there something you would like to upload?
[12:44] <HiddenWolf> seb128/mdz: if you guys select the applications menu, does the text turn white? does the same happen for places/system?
[12:44] <seb128> grab and include some "System" translations for the menu
[12:44] <mdz> HiddenWolf: we're preparing for the preview release and quite busy at the moment; please try to restrict conversation to development topics, and take other subjects to #ubuntu
[12:44] <seb128> and change the warty about to hoary
[12:44] <HiddenWolf> mdz: I want to know if it's a bug or something, sorry
[12:44] <seb128> ie: a gnome-panel upload
[12:44] <mdz> seb128: ok, but the translations will require new langpacks
[12:45] <seb128> hum
[12:45] <seb128> these are details, so that's as you want
[12:45] <mdz> it would be good to have the translations in the archive so that they will go into the next set of langpacks
[12:45] <seb128> k
[12:46] <Kamion> mdz: *shrug* if I can save 20MB, I see no reason not to do so
[12:46] <seb128> is that ok tu upload now ? or is there a freeze to get the same content for differents images ?
[12:47] <elmo> Kamion/mdz: you guys need anything from me for preview of doom before I run away?
[12:47] <mdz> elmo: not unless Kamion needs to do another d-i upload
[12:48] <Kamion> not as far as I know
[12:48] <mdz> seb128: that's fine
[12:48] <seb128> k, I'll upload in ~30min so
[12:48] <Kamion> and in the event of an emergency I can probably guess at what needs to be done / bug thom
[12:49] <lamont> mdz: fyi - off to fire training in about 40 minutes, gone until probably 2230 local
[12:49] <elmo> Kamion: not sure thom's anymore awake than me, but mdz has supah-fly katie powahs atm
[12:50] <mdz> get 'em while they last
[12:50] <Kamion> elmo: oh of course, thom's in the DC too, oh well
[12:50] <elmo> Kamion: ha ha.  no he's not, but he muttered about going to bed early when he left
[12:50] <elmo> SEVERAL HOURS AGO
[12:50] <Kamion> I can probably make a good guess at what d-i byhand processing requires, but I have no intention of trying right now :-)
[12:50] <elmo> </bitter ;>
[12:50] <Kamion> heh
[12:51] <elmo> I put my cheet sheeti n ~james/ on jackass I think
[12:51] <Kamion> elmo: speaking of, if I hacked up katie to do that automatically for raw-installer, would you take that patch?
[12:51] <ajmitch> elmo: sync for universe still? I've just uploaded gnue-common, gnue-forms, gnue-appserver to sid
[12:51] <elmo> Kamion: if it was sane, sure
[12:51] <ogra> ajmitch: we havent defined a real freeze for universe yet
[12:52] <Kamion> mkay
[12:52] <ajmitch> ogra: I know, it's on the agenda for next weeks meeting, iirc
[12:52] <Kamion> since that was kind of the point of the raw-installer change ;)
[12:52] <ogra> ajmitch: since there is a lot to do it will probably require newer versions more often....so i would call it a "soft freeze"
[12:53] <Kamion> gar, it would be nice if dak CVS were actually sanely usable by non-committers though
[12:53] <ajmitch> ogra: quite a slushy freeze then?
[12:53] <elmo> Kamion: how do you mean?
[12:53] <elmo> there's a package in the archive now :P
[12:53] <elmo> well not our archive, but stil
[12:53] <Kamion> cvs server: failed to create lock directory for `/cvs/dak/dak' (/cvs/dak/dak/#cvs.lock): Permission denied
[12:54] <lamont> hrm.. we used to call those 'chill's.
[12:54] <elmo> Kamion: as you or anoncvs?
[12:54] <elmo> if you, surely that's a problem for everything you don't have write privs to?
[12:54] <Kamion> elmo: hm, as me; is anoncvs liable to work better?
[12:54] <ogra> ajmitch: lets call it a freeze, behave like in a freeze, but keep in mind that we wont be able to really freeze, since i suspect we will fix things until the last day
[12:54] <ajmitch> ogra: alright
[12:55] <ogra> ajmitch: simply to many packages
[12:55] <elmo> Kamion: anoncvs definitely works
[12:56] <Kamion> elmo: ah, ok, I'm just being special then, thanks
[12:56] <Kamion> yeah, that works
[12:56] <lamont> ogra/ajmitch: speaking of universe... I occassionally give everything back in universe, just to see what else falls out...
[12:56] <dholbach> lamont: i dont understand
[12:56] <ajmitch> lamont: where do you keep the lists of unbuildables? and is there a list of uninstallables somewhere?
[12:57] <ajmitch> dholbach: push unbuilt packages back to the buildds
[12:57] <ogra> ajmitch: Kamion will assemble one if i understood it right
[12:57] <ajmitch> great
[12:57] <dholbach> oh yes, good
[12:57] <ajmitch> another MOTUTodo list :)
[12:57] <ogra> ajmitch: so lets be patient ;)
[12:57] <Kamion> er, well that's the plan, it's in the ever-growing pile I affectionately call my to-do list
[12:57] <lamont> http://people.ubuntu.com/~lamont/buildLogs/List/hoary.all.i386 (etc)
[12:58] <lamont> if it's marked Building, and someone didn't upload it recently, then it's really 'Failed'....
[12:58] <Kamion> use apt-cache unmet or something like that in the meantime
[12:58] <lamont> those are the ones I give back
[12:58] <ogra> Kamion: thats why i said patient :)
[12:58] <lamont> likewise, http://people.ubuntu.com/~lamont/buildLogs/List/hoary.building.i386
[12:58] <Kamion> although apt-cache unmet bitches about all kinds of stuff nobody cares about, like unsatisfied Suggests
[12:58] <ogra> we have enough to do to fill the time :)
[12:58] <lamont> is shorter. :-)
[12:58] <ajmitch> lamont: thanks
[12:59] <lamont> er, s/List/Lists/
[01:00] <ogra> lamont: what means "not ours" in a failed list ?
[01:01] <Kamion> mdz: cron.daily done
[01:04] <dholbach> now really *wave*
[01:04] <ogra> night dholbach
[01:04] <ajmitch> night dholbach 
[01:05] <zul> hey
[01:06] <lamont> ogra: architecture not specified in arch list
[01:06] <ogra> ah
[01:06] <lamont> but not in PaS
[01:06] <ogra> PaS ?
[01:06] <lamont> Packages-arch-specific
[01:07] <lamont> if a package is in PaS, then it won't even be in the list.
[01:07] <lamont> oh, and arch-all packags are only in the listings when they haven't been built - they're only built on i386
[01:09] <ogra> yup, thats oune i figuread already :)
[01:09] <ogra> lamont, thanks :)
[01:13] <zenwhen> wow
[01:13] <zenwhen> k3b realy slowed down a lot
[01:13] <zenwhen> my dvd burner wont burn over 2.4x now
[01:13] <zenwhen> even though its an 8x burner
[01:15] <Kamion> mdz: what time will preview be?
[01:16] <jdub> Kamion: some time after 10am UTC, at least :)
[01:17] <Kamion> what's at 10am UTC?
[01:17] <jdub> gnome 2.10
[01:17] <Kamion> heh, ok :)
[01:18] <Kamion> last time round was 4pm UTC IIRC, but I might not RC
[01:18] <jdub> yeah, it was
[01:18] <jdub> we should aim for earlier though
[01:18] <jdub> that kinda sucked for europe
[01:19] <Kamion> full set of CD builds takes around an hour nowadays
[01:19] <Kamion> need to take that into account
[01:25] <Kamion> jdub: if most stuff is done, we should probably freeze for all but emergency stuff and new GNOME tarballs now so that people can test the pre-preview candidate
[01:32] <jdub> Kamion: yeah; i have an ubuntu-artwork to go in
[01:32] <jdub> other than that, i can't think of anything important
[01:32] <jdub> (that isn't dangerous)
[01:35] <seb128> 'night
[01:35] <Kamion> ubuntu-artwork is always last :P
[01:40] <jdub> ;)
[01:40] <jdub> seb128: dude
[01:40] <jdub> seb128: thanks :-)
[01:40] <mdz> Kamion: 10am UTC works for me
[01:40] <jdub> seb128: and have a very good sleep :-)
[01:40] <lamont> seb128: good long hard work.
[01:40] <seb128> jdub: Thanks :)
[01:40] <Kamion> mdz: we will need to check tomorrow morning that we are still up-to-date with GNOME, and if not then wait for jdub/seb128's go-ahead
[01:41] <lamont> must apply pressure to the phone company.
[01:41] <ogra> seb128: night, great work :)
[01:41] <jdub> Kamion: unless there's anything critical, i don't think it's worth updating gnome stuff in general
[01:41] <HrdwrBoB> lamont: burn fown your exchange
[01:41] <HrdwrBoB> down
[01:41] <lamont> HrdwrBoB: no. that'd be bad
[01:41] <Kamion> in that event it'll be whenever-uploads-are-ready + one hour for builds + time for testing
[01:41] <lamont> besides, it's 10 miles away
[01:41] <HrdwrBoB> 10 miles and you want DSL?
[01:41] <seb128> mdz: panel need to some other hacks for the translations, I'll do that after the preview
[01:41] <jdub> Kamion: but i can track gnome stuff up to the minute of release, and do uploads if absolutely necessary
[01:41] <Kamion> jdub: ok
[01:41] <lamont> HrdwrBoB: the RT is only 3 miles away
[01:42] <HrdwrBoB> ah ok
[01:42] <seb128> jdub: we want all the modules uptodate ?
[01:42] <lamont> and then fiber back to the CO
[01:42] <Kamion> mdz: if there are no new GNOME uploads, are our current live CDs sufficiently final?
[01:42] <jdub> seb128: i'm find with what we have; will only upload if there's something super critical (unlikely)
[01:42] <lamont> well, 4.5
[01:42] <mdz> Kamion: I have a full set here and am starting a test cycle
[01:43] <Kamion> mdz: likewise (well, of install anyway, will have to catch up with the live-amd64 rootfs regen)
[01:43] <seb128> jdub: k, because there is a gdm 2.6.0.7 to 2.6.0.8, but with some conflict and changes, I prefer to upload it after preview
[01:44] <jdub> cool
[01:44] <mdz> I have install CDs, too; I'll do powerpc and amd64 after testing the live CDs
[01:44] <mdz> I can't clobber my i386  at the moment
[01:48] <Kamion> we still have that ugly scrollkeeper error at the end of the install; can we fix that?
[01:48] <Kamion> it's a wart
[01:49] <jdub> and we're not warty anymore :)
[01:49] <Kamion> http://people.ubuntu.com/~cjwatson/tmp/scrollkeeper-errors
[01:49] <jdub> what do you think the duplicate to damage ratio is?
[01:50] <Kamion> duplicate to damage?
[01:50] <jdub> duplicate bugs of that error vs. potential damage of fix
[01:51] <Kamion> it's a bad-first-impression thing; on slow machines that error message will hang around there for some time
[01:51] <Kamion> potential damage seems low to me, but I'm no expert
[01:51] <jdub> i'll find out now
[01:51] <Kamion> thanks
[01:52] <Kamion> other than that, install-i386 seems fine
[01:54] <jbailey> Kamion: Got a fresh instal handy? =)
[01:54] <Kamion> aha, yes
[01:54] <Kamion> hmm
[01:54] <Kamion> no RESUME=
[01:54] <jbailey> Yar.
[01:54] <Kamion> any debugging I can do?
[01:55] <jbailey> I need to know what /proc/swaps looks like at install time, and whether that exact same device exists in /dev
[01:55] <jbailey> mdz thought that it should be LSB naming in /proc/swaps, and the the lsb device should exist, just as a symlink.
[01:56] <Kamion> ok, will check for you
[01:56] <Kamion> LSB naming?
[01:56] <jbailey> So I weakened the test for the /dev file from wanting a block device to wanting any file.  Thought that should've done it.
[01:56] <jbailey> Kamion: /dev/hda, or whatever.
[01:56] <Kamion> right
[01:56] <mdz> Kamion: all 3 live CDs are releasable
[01:56] <Kamion> hmm, /dev might not be bind-mounted into /target
[01:56] <mdz> the login sound doesn't play, due to the known polypaudio breakage
[01:57] <mdz> and I noticed one odd thing
[01:57] <mdz> that the default sound effect for the terminal bell was different on powerpc than on i386
[01:57] <mdz> on i386 it was the "clack" and on powerpc the "bloomp"
[01:58] <jdub> *boggle*
[01:58] <jdub> load-sample x11-bell /usr/share/sounds/generic.wav
[01:59] <jdub> ^ default.pa
[01:59] <jdub> mdz: you chose not to revert to esound for preview?
[01:59] <mdz> jdub: right, it was too last-minute
[01:59] <jdub> ok
[02:00] <mdz> we'll talk on friday about what to do for final
[02:00] <zenwhen> so preview is tonight?
[02:00] <Kamion> zenwhen: what timezone are you in?
[02:00] <mdz> zenwhen: preview is on the day it was scheduled 6 months ago
[02:00] <zenwhen> lol
[02:00] <zenwhen> Eastern
[02:00] <jdub> i'd lean towards the last minute reversion instead of the preview bugs, though
[02:00] <Kamion> zenwhen: sometime early morning your time, then
[02:00] <zenwhen> oh cool
[02:00] <zenwhen> If I could, I would update.
[02:01] <schweeb> just in time for an early morning reinstall
[02:01] <mdz> jdub: I would lean toward not having rushed into this particular burning building in the first place
[02:01] <zenwhen> But I will do so after preview.
[02:01] <schweeb> should fill up the work day nicely
[02:01] <jdub> mdz: it had to be tested
[02:01] <Kamion> jbailey: /dev/scsi/host0/bus0/target0/lun0/part5 partition       497972  0       -1
[02:02] <zenwhen> The only bug I noticed is that k3b now hates my Lite-On ldw-811s
[02:02] <jdub> mdz: if you're that unhappy with it, pull it out now; that's been my recommendation
[02:02] <Kamion> jbailey: definitely not LSB naming in /proc/swaps, and that device probably won't exist in /target
[02:02] <mdz> jdub: not if upstream had already bugged out on us
[02:02] <Kamion> jbailey: a hard one, I'm afraid
[02:02] <Kamion> jbailey: I think it'll have to be kludged in base-installer
[02:02] <mdz> jdub: there is no way I am doing that hours before preview
[02:03] <jbailey> Kamion: Ugh.  Thanks for checking.
[02:03] <Kamion> jbailey: note that base-installer already does some mkinitrd.conf munging
[02:04] <Kamion> jbailey: I don't know the exact semantics of what it's doing, though
[02:04] <jbailey> Kamion: Ah?  I hadn't realised.  So it's not that ugly then.
[02:04] <Kamion> but it looks like it could be slotted in there somehow
[02:04] <jdub> mdz: right, i'm not really suggesting that, just expressing surprise.
[02:04] <Kamion> jbailey: but probably post-preview now, given that we'll need time to test it :( Sorry I didn't get round to checking this out earlier
[02:05] <Kamion> hm, that sucks, lots of people will install preview then upgrade
[02:05] <mdz> jdub: it doesn't seem like a disaster functionality-wise; the reason that I'm leaning toward reverting it for final is that we have no upstream support
[02:05] <Kamion> I could test by hacking base-installer on the fly ...
[02:05] <schweeb> mdz: I noticed in some recent mailing list posts that you were sort of interested in Xen for ubuntu... anyone else express interest in it?  I've been looking at it recently, and am willing to help, but am not sure I can devote a ton of time on it (plus, I need to gain a little experience too)
[02:05] <jbailey> Kamion: Is `mapdevfs` the tool for doing the conversion?
[02:06] <mdz> schweeb: yes, others have expressed interest, and with the same disclaimer
[02:06] <Kamion> jbailey: yep.
[02:06] <Kamion> jbailey: base-installer already depends on it
[02:06] <jbailey> Kamion: What's the best way to get the fix in without causing a race condition with you?
[02:06] <jbailey> (and with suitable timing and such)
[02:06] <jdub> mdz: yes, that's what i've said, in addition to certain regressions for stuff we use.
[02:07] <schweeb> mdz: heh, alright... regular users can create wiki pages, right?  possibly I'll post some results to a wiki page so at least interested people have a reference
[02:07] <Kamion> base-installer isn't in the d-i initrd, so it's just an upload; but we should both read over the diff
[02:07] <mdz> schweeb: yes
[02:07] <mdz> I'm surprised there isn't already a page (if indeed there isn't)
[02:07] <schweeb> there isn't that I could find
[02:08] <schweeb> just a few mentions of how it's a possible target of opportunity
[02:09] <schweeb> haven't been able to get it booting on ubuntu yet, but I'm guesing that's a possible udev issue or something... worked essentially oob w/ sarge
[02:09] <jbailey> Kamion: Do you know off hand if awk and wc are available at that point?
[02:09] <jbailey> Mm, and sort and head.
[02:09] <Kamion> jbailey: no awk
[02:09] <jbailey> I know that they're not available in a busybox environment.
[02:10] <Kamion> this is a busybox environment
[02:10] <Kamion> you have wc, sort, head
[02:10] <jbailey> It's not gnu sort, so my "sort by the third column to find the biggest swap" won't work. =)
[02:11] <Kamion> you could do:
[02:12] <Kamion> while read file a size b c; do if [ "$size" -ge "$maxsize" ] ; then maxsize="$size"; maxfile="$file"; fi; done
[02:12] <Kamion> or words to that effect
[02:12] <jordi> fabbione: ping
[02:12] <Kamion> obviously with initialisation and stuff
[02:13] <mdz> hmm
[02:13] <mdz> this sucks
[02:14] <mdz> I just realized that update-manager won't add/remove packages
[02:14] <mdz> so people upgrading from preview will need to use synaptic to get seed changes
[02:14] <jbailey> Kamion: Oh, is read posix?  Nice.
[02:14] <Kamion> yeah
[02:15] <mdz> oh, even better
[02:15] <Kamion> you don't have all of bash's read options I don't think, but you rarely need them
[02:15] <jdub> mdz: if we assume that most update-manager users will not be on devel branch, should we switch it to dist-ugprade?
[02:15] <mdke_laptop> does anyone know if the wiki's can have multiple parents? sorry to disturb
[02:16] <Kamion> WHOA
[02:16] <Kamion> locale generation fell over
[02:16] <mdz> jdub: until Hoary is released, every single update-manager user is on the development branch
[02:16] <mdz> mdke_laptop: no, they can't
[02:16] <Kamion> fuck, it tried to generate "en_GB.UTF-8en_US.UTF-8"
[02:16] <mdz> DOH
[02:16] <mdke_laptop> mdz, thanks
[02:17] <jdub> mdz: that's correct, but not useful :)
[02:17] <Kamion> and I seem to have old language packs on this CD
[02:17] <mdz> jdub: it is, in the same way as it's useful to change the dpkg force-overwrite flag before release
[02:17] <jdub> mdz: we're not going to be making dramatic seed or dependency changes
[02:17] <mdz> jdub: well, we're most likely going to polypaudio->esound
[02:17] <schweeb> mdz: hrm, interesting... there's no mention of xen that the search engine can find either... guess I'll make a wiki page
[02:18] <mdz> which is a seed and dependency change
[02:18] <mdz> and we made such changes the last time around too
[02:18] <jdub> mdz: and we know exactly what the effect of that is, it's not unmeasurable or dramatic
[02:18] <mdz> jdub: and it isn't handled by update-manager. that was my point.
[02:18] <jdub> also mine
[02:18] <jdub> given that we know dist-upgrade won't be dangerous for most update-manager users
[02:19] <jdub> should it use dist-upgrade?
[02:19] <mdke_laptop> mdz, is it a conscious choice not to allow wikis to have multiple parents, or just a limitation with the software?
[02:19] <Kamion> mdz: ok, I will work around this in base-config rather than requiring a localechooser change (which => d-i initrd rbeuild)
[02:19] <Kamion> rebuild
[02:19] <mdz> mdke_laptop: I don't know
[02:19] <\sh> hola ogra
[02:19] <jdub> that does mean it will be slightly scarier if the users hugs the devel branch
[02:19] <ogra> whoo \sh
[02:19] <mdz> no hugging is required
[02:19] <mdz> this means that upgrades from preview->final can't be done with update-manager
[02:19] <jdub> but in the common case, not too bad
[02:19] <jdub> no
[02:20] <jdub> dude
[02:20] <ogra> nice to see you here \sh 
[02:20] <mdz> which is a shame
[02:20] <mdke_laptop> hi ogra
[02:20] <ogra> hi
[02:20] <jdub> currently, yes it does mean that
[02:20] <\sh> ogra: hehe...wanna see how you prepare your releases :)
[02:20] <jdub> however, i am asking if we should switch update-manager to use dist-upgrade
[02:20] <ogra> \sh, you old gentoo spy ;)
[02:20] <mdz> I haven't read the code
[02:20] <mdz> I don't know how update-manager works yet
[02:21] <Kamion> oh, crap, no, I have to change localechooser; /etc/locale.gen is broken
[02:21] <\sh> ogra: hehe...
[02:21] <jdub> perhaps i've confused you by mentioning the drawbacks while introducing this idea
[02:21] <mdz> omg
[02:21] <mdz> I shouldn't be reading this
[02:21] <Burgundavia> jdub: are we going to offer the users a choice to upgrade?
[02:21] <jdub> Burgundavia: ?
[02:22] <jdub> there is always a choice
[02:22] <ogra> heh
[02:22] <jdub> you have to very explicitly choose to upgrade
[02:22] <Burgundavia> I will just shutup now
[02:22] <Burgundavia> I should never jump into the middle of conversation while reading the beginning
[02:23] <\sh> choice? i thought this was the gentoo way ;)
[02:23] <ogra> \sh, i thought it was compiling in lonely nights
[02:24] <\sh> ogra: i'm not lonely ;)
[02:25] <Burgundavia> my hero, Johnny from Sun said that Linux was about vendor lock-in ;)
[02:25] <mdz> jdub: yes, I do think we should switch it over to dist-upgrade, but not for preview
[02:25] <mdke_laptop> ogra, its nice to have a compile going on those rainy days...
[02:25] <ogra> mdke_laptop: at least you got rain.....we got snow
[02:25] <mdke_laptop> lol
[02:26] <\sh> ogra: we got rain as wenn in kerpen ;)
[02:26] <mdke_laptop> ogra, its always raining in the uk
[02:26] <\sh> as well
[02:26] <Kamion> mdz: well, that allows upgrading preview->final in two steps. :-) first one upgrades update-manager ...
[02:26] <mdz> jdub: is there any mechanism to shut down polypaudio at all, or do we need to recommend a reboot/killall after the switch to esound?
[02:27] <ogra> \sh, i absoltely dont miss this place ;) (even i have to come around next week)
[02:27] <mdz> Kamion: right, so, how about some sleep between now and byhanding d-i?
[02:28] <mdz> Kamion: I don't think I'm really up for that right now
[02:28] <jdub> mdz: polypaudio -k
[02:28] <zul> heylog
[02:28] <\sh> ogra: but I'm missing the meetings and the discussions about linux in the cantine with u :)
[02:28] <mdz> jdub: I meant automatically
[02:28] <ogra> \sh, we can have them here ;)
[02:28] <mdz> jdub: but that's one of the bugs, isn't it. it doesn't get killed?
[02:28] <Kamion> mdz: it's going to be RSN
[02:28] <mdz> Kamion: what is?
[02:28] <jdub> mdz: oh, at logout? no, it should die
[02:29] <Kamion> mdz: d-i build
[02:29] <Kamion> mdz: well, depends how much sleep you need :)
[02:29] <mdz> Kamion: oh, are you going to byhand it onto little or something?
[02:29] <Kamion> no
[02:30] <mdz> you have a magic knob in the CD build to pull it from a directory outside of the archive?
[02:30] <Kamion> no ...
[02:30] <mdz> don't keep me in suspense
[02:30] <Kamion> just uploading now
[02:30] <Kamion> (localechooser)
[02:30] <mdz> localechooser is in the initrd, yes?
[02:30] <Kamion> ok, if you need sleep before the half an hour or whatever it is, go for it
[02:30] <Kamion> yes, sadly
[02:30] <mdz> so this change requires that somebody byhand d-i
[02:30] <Kamion> right
[02:31] <mdz> you said that you didn't want to do it
[02:31] <Kamion> ok, let's leave it until after you've slept
[02:31] <mdz> I'm significantly less confident about it
[02:31] <mdz> I'm happy for elmo to do it in the morning, and release in the afternoon
[02:32] <mdz> though I'm not sure what time elmo's morning will be, given that he just flew in
[02:32] <Kamion> sorry 'bout the breakage, I'd hoped not to have to do this
[02:32] <mdz> not a big deal
[02:33] <mdz> we've been ahead of schedule this time around
[02:33] <\sh> ogra: have to leave now..need to help thomas...cu here in kerpen...btw...branislav has his farewell drink 1) thursday, 5:30pm bowling center and 2) next week tuesday in cologne...
[02:33] <mdz> consider
[02:33] <mdz> this is the day BEFORE it's due :-)
[02:33] <Kamion> hah
[02:33] <Kamion> ok, time for coffee
[02:33] <mdz> if we're going to leave it overnight
[02:33] <ogra> \sh, i'll be there on thursday
[02:33] <mdz> jdub has nearly convinced me to throw caution to the wind and revert to esound
[02:33] <Clint> mdz: the baz completion function expression to grab the baz version is broken; which number from the baz --version output is safe to count on?
[02:33] <ogra> \sh, i'll have to do that as well i think
[02:34] <\sh> ogra: after this, i have to leave to bonn for a luusa meeting :) u can join for a beer as well there ;) 
[02:34] <mdz> Clint: which number?
[02:34] <ogra> \sh, yeah lets see ... :)
[02:35] <Clint> mdz: 1.1.1 or 1.2 is repeated a few times
[02:35] <\sh> ok cu bro :)
[02:35] <mdz> Clint: the current version is 1.2
[02:35] <mdz> Clint: I'm not sure how to answer your question more accurately than that
[02:35] <Clint> mdz: lemme rephrase.  unless there's a better way, I'm going to grab the value in between "baz Bazaar version" and "(thelove@canonical.com"
[02:36] <mdz> I think "Bazaar version (word)" is probably safest
[02:36] <mdz> but honestly I have no idea if the bazaar developers plan to change that output format
[02:36] <mdz> somebody in #bazaar might now
[02:36] <Clint> okay, I'll go there
[02:38] <jdub> mdz: overnight to which hour in UTC?
[02:38] <mdz> jdub: sometime after 1000
[02:38] <jdub> mdz: it really is just a matter of changing the seed for this one
[02:38] <Kamion> I'll be up from ~0900
[02:38] <jdub> mdz: heh
[02:38] <jdub> mdz: aha!
[02:38] <mdz> I have a choice between staying up to 1000, or sleeping and joining up later
[02:39] <mdz> jdub: the preview was looking too perfect and I didn't want to curse it by fiddling in the last hours
[02:39] <mdz> jdub: but now that a bug has turned up, I feel more relaxed :-)
[02:40] <jdub> :)
[02:40] <Kamion> so does that mean we can fix these scrollkeeper errors too? :)
[02:40] <mdz> Kamion: what scrollkeeper errors?
[02:41] <Kamion> mdz: http://people.ubuntu.com/~cjwatson/tmp/scrollkeeper-errors, on every install
[02:41] <jdub> Kamion: so i know how to fix the first two (easy, not dangerous)
[02:41] <Kamion> jdub: the first two are the ugliest
[02:41] <mdz> jdub: * committed ubuntu-devel@lists.ubuntu.com/seeds--hoary--0--patch-184
[02:41] <jdub> Kamion: gotta do some searching for the last one
[02:41] <mdz> and may god have mercy on my soul
[02:41] <Kamion> localechooser 0.04.0ubuntu7 fixes the locale generation bug
[02:41] <jdub> mdz: :-)
[02:41] <mdz> Kamion: can you kick the published seeds?
[02:42] <jdub> mdz: you kicked out alsa and x11 too?
[02:42] <Kamion> mdz: done
[02:42] <Burgundavia> jdub: who is working on the front page redesign?
[02:42] <mdz> jdub: zsh: exit 1     grep polypaudio desktop
[02:42] <jdub> bonus
[02:42] <Kamion> - * polypaudio-alsa
[02:42] <Kamion> - * polypaudio-x11
[02:42] <Kamion> + * esound
[02:42] <jdub> Burgundavia: henrik, ryan (forums dude) and the guys who won the comp
[02:43] <Burgundavia> jdub: thanks
[02:43] <mdz> uploaded ubuntu-meta 0.38
[02:43] <zenwhen> jdub, have you heard any complaints about dvd burn speeds dropping after having gone from warty to hoary?
[02:43] <Kamion> mdz: not leaving polypaudio in supported?
[02:43] <Kamion> of course polypaudio-clients is in supported still
[02:43] <mdz> Kamion: the reason we're ditching it is that it's unsupportable, so I see no reason to keep it in main
[02:43] <jdub> zenwhen: no
[02:43] <mdz> I'll fix that
[02:43] <zenwhen> jdub, Oh. Maybe it is just me then. :(
[02:44] <Kamion> mdz: 'k
[02:44] <mdz> dear baz
[02:44] <mdz> please make me type in my passphrase more
[02:44] <mdz> love, mdz
[02:44] <jdub> mdz: gnome-gpg :)
[02:44] <mdz> I do not trust that thing with my key
[02:44] <mdz> it doesn't even lock memory!
[02:45] <Kamion> but but but it's got "gnome-" on the front
[02:45] <Kamion> I'll get my coat
[02:45] <mdz> when ubuntu-desktop comes around on the cron.daily train, I'll build new cloop images
[02:46] <Kamion> jdub: teasing :)
[02:46] <mdz> that'll be 43 minutes
[02:46] <Kamion> quintuple-agent looked less bad last time I investigated
[02:46] <Kamion> IIRC it locks memory
[02:47] <jdub> there's some new gpg agent around too
[02:47] <jdub> gpg 4?
[02:48] <mdz> Kamion: I did not seem to encounter this locale problem; what should I be looking for?
[02:48] <mdz> jdub: gnupg2
[02:48] <mdz> KDE 3.4 uses it
[02:49] <mdz> looks a bit like ssh-agent
[02:49] <mdz> from the outside anyway
[02:52] <Kamion> mdz: look in /etc/locale.gen; I noticed it going past while language-pack-* was being installed
[02:53] <Kamion> mdz: but it won't happen if you did an en_US install
[02:53] <mdz> Kamion: my /etc/locale.gen looks pretty normal
[02:53] <mdz> ah
[02:53] <mdz> I did
[02:53] <Kamion> it appends en_US.UTF-8 to supported-locales if the locale was not en_US.UTF-8
[02:54] <Kamion> but I forgot a space, so localechooser/prebaseconfig and base-config/lib/menu/pkgsel got confused
[02:54] <mdz> gotcha
[02:54] <mdz> I just didn't realize US was magic
[02:54] <Kamion> yeah, it seemed silly to set it to "en_US.UTF-8, en_US.UTF-8" by default :)
[02:58] <schweeb> dammit, I'm a moron... is there any way to change your name in the wiki?  I managed to mistype my own last name
[02:58] <schweeb> s/the wiki/plone/
[02:59] <mdke_laptop> schweeb, preferences are pretty limited
[02:59] <schweeb> indeed
[03:06] <schweeb> bah, can only register an email once
[03:07] <mdke_laptop> also you can't change your email address i don't think
[03:07] <schweeb> nope
[03:07] <schweeb> is there any chance I could trouble our Ubuntu overlords to fix my error?
[03:08] <mdke_laptop> lol i doubt that
[03:08] <mdz> schweeb: mail webmaster@
[03:09] <schweeb> mdz comes through again, thx
[03:09] <mdke_laptop> :)
[03:09] <mdz> no problem, but please try to keep the conversation on-topic in here; we're under release pressure
[03:10] <schweeb> mdz: alright, sorry.  assumed the person responsible would be more likely to see my call for help here.  I'll go back to lurking and working on Xen
[03:28] <mdz> Kamion: does there exist a cheat sheet for d-i byhanding?
[03:30] <Kamion> mdz: elmo said he'd left one in jackass:~james/
[03:31] <mdz> Kamion: I see no instructions, but there is a script
[03:31] <Kamion> ICBW, but I'd expect it's mostly "tar xzvf" in the right place
[03:31] <Kamion> and maybe cleaning up old directories
[03:32] <mdz> the script has two sections, like the intent is for one of them to be cut and pasted
[03:32] <mdz> ah, one for daily builds and one for real uploads
[03:33] <Kamion> oh, daily builds would probably have installer-i386/ in their tarballs, but expand into daily-installer-i386/, which might be interesting
[03:34] <mdz> yeah, it mv's that stuff around
[03:34] <mdz> it isn't terribly frightening, but there's a bit more to it than I'd like
[03:34] <mdz> mailed you a copy
[03:35] <Kamion> thanks
[03:35] <mdz> it would be nice to be able to send out a request for overnight testing
[03:35] <Kamion> I started writing a katie patch to do it automatically; most of it will probably be sanity-checking the tarball
[03:37] <Kamion> localechooser 0.04.0ubuntu7 has not hit archive.u.c yet
[03:39] <Kamion> ah, there it is
[03:39] <mdz> should be there right now
[03:39] <mdz> as well as ubuntu-meta 0.38
[03:39] <Kamion> mdz: could you kick the d-i build? the machine with my script on it is currently running an install
[03:40] <mdz> oh fuck
[03:40] <mdz> polypaudio Provides: esound
[03:40] <mdz> Kamion: kicked
[03:41] <Kamion> tas
[03:41] <Kamion> ta
[03:41] <mdz> Kamion: what do you think we should do about esound?
[03:41] <mdz> what a mess
[03:41] <mdz> could upload polypaudio without the Provides, I suppose
[03:42] <jdub> that would break polypaudio unnicely
[03:42] <Kamion> no idea, I mean it'll work for new installs but not for upgrades
[03:42] <mdz> break?
[03:42] <mdz> Kamion: right
[03:42] <Kamion> do we care?
[03:42] <mdz> well, I'd really appreciate the testing
[03:43] <jdub> we could just add to release notes "those testing with polypaudio before the preview should install esound"
[03:43] <mdz> and at some point we need to upgrade people
[03:43] <Kamion> a versioned dep from ubuntu-desktop would do it
[03:43] <jdub> yeah
[03:43] <Kamion> bit weird though
[03:43] <mdz> and there's no infrastructure in ubuntu-meta for it
[03:43] <mdz> ubuntu-desktop could conflict with polypaudio
[03:44] <jdub> yucky
[03:44] <Kamion> that would be foul, and probably cause upgrade issues
[03:44] <mdz> that's the best idea I've got
[03:44] <jdub> do we need to forcibly upgrade those people who have polypaudio?
[03:44] <Kamion> since ubuntu-desktop would have to be deconfigured (at least) during the upgrade
[03:45] <Kamion> I suspect we would end up with some confused people without ubuntu-desktop
[03:45] <mdz> Kamion: nah, it should go fine, since ubuntu-desktop is being upgraded at the same time
[03:45] <Kamion> it will still have to be deconfigured
[03:45] <mdz> it will be unpacked and unconfigured anyway
[03:45] <mdz> the installed version won't be deconfigured
[03:46] <mdz> the new version just gets unpacked over it
[03:46] <jdub> hmm
[03:46] <jdub> so
[03:46] <Kamion> since polypaudio will have to be totally removed before the new ubuntu-desktop is unpacked
[03:46] <jdub> not a lot depends on esound
[03:46] <Kamion> a versioned dependency on esound is much, much cleaner than that
[03:46] <jdub> we could remove the provides
[03:46] <mdz> Kamion: the d-i builds all exited; I have no idea if they succeeded
[03:46] <jdub> and add it again later
[03:46] <Kamion> and doesn't break the people who are actually trying to test polypaudio
[03:46] <mdz> Kamion: ross failed due to an ssh host key mismatch; do you know which arch that is?
[03:46] <Kamion> mdz: powerpc I think
[03:46] <jdub> but it does break polypaudio in the mean time
[03:46] <Kamion> mdz: just run that one again by hand?
[03:47] <mdz> Kamion: retrying
[03:47] <mdz> jdub: how does removing the provides break polypaudio?
[03:47] <jdub> mdz: it means you can't install it without removing libesd...
[03:48] <Kamion> oh, is it another damn fam/gamin-style thing?
[03:48] <mdz> jdub: hmm?
[03:48] <Clint> mdz: s/star_merge/merge/g in _baz and your problem should be gone
[03:48] <jdub> mdz: which doesn't break polypaudio itself, but is pretty solidly broken
[03:48] <Kamion> library depends on daemon
[03:48] <jdub> Kamion: yes
[03:48] <mdz> Clint: I was sure I did that in the first place
[03:48] <Clint> oh
[03:48] <jdub> Kamion: libesd talks to esound *and* polypaudio
[03:48] <jdub> Kamion: fam was a little bit different
[03:48] <Kamion> how about libesd depends esound | polypaudio, then?
[03:48] <jdub> ok
[03:48] <jdub> will do that now
[03:49] <mdz> I have no qualms at all about breaking polypaudio if that's what it takes
[03:49] <jdub> much saner :-)
[03:49] <mdz> but I'd prefer not to mess with esound
[03:49] <mdz> at least not right now
[03:49] <Kamion> we could break polypaudio for preview, and fix it again right afterwards
[03:49] <Kamion> with the above method
[03:49] <jdub> Kamion: doesn't fix the upgrade issue though
[03:49] <jdub> unless we do the versioned esound
[03:50] <Kamion> jdub: if polypaudio stops providing esound, the ubuntu-desktop upgrade will install real esound
[03:50] <jdub> Kamion: (referring to libesd change only)
[03:50] <jdub> it doesn't depend anyway, just sugests
[03:51] <jdub> have we thrown out versioned esound depend entirely?
[03:51] <Kamion> whoa, billions of scrollkeeper errors on powerpc
[03:52] <Kamion> gcalctool apparently
[03:52] <jdub> Kamion: can you put output up again?
[03:52] <Kamion> yeah, just doing
[03:53] <Kamion> jdub: http://people.ubuntu.com/~cjwatson/tmp/scrollkeeper-errors-powerpc
[03:53] <mdz> ubuntu-desktop conflicts: polypaudio is perfectly reasonable to me
[03:53] <mdz> we actively do not want supported systems to have polypaudio
[03:54] <mdz> apt will prefer to upgrade ubuntu-desktop and install esound, rather than remove ubuntu-desktop and keep polypaudio
[03:54] <mdz> but I suppose we can punt on it for preview
[03:54] <mdz> kicking live fs builds
[03:55] <Kamion> jdub: locale is de_DE.UTF-8, btw
[03:59] <Kamion> but weirdly a scrollkeeper-rebuilddb after install doesn't show the gcalctool errors
[03:59] <mdz> Kamion: what do you want to do about the d-i build?
[03:59] <jdub> Kamion: it doesn't do anything for me after a recent rebuilddb
[03:59] <Kamion> jdub: yeah, maybe transient weirdness :(
[03:59] <Kamion> mdz: if you can byhand it, I think that would be best
[04:00] <jdub> Kamion: not getting any xmllint issues with any of those
[04:00] <jdub> Kamion: can you try on ppc?
[04:01] <mdz> Kamion: so I want the lower half, right?
[04:01] <Kamion> mdz: no, upper
[04:01] <mdz> oh, it's a daily
[04:01] <Kamion> yes
[04:02] <mdz> ARCHS="i386 amd64 powerpc ia64"
[04:02] <mdz> TAR_VERSION=20041227ubuntu20.0.200503090
[04:02] <mdz> VERSION=20041227ubuntu20.0.200503090
[04:02] <mdz> look correct?
[04:02] <Kamion> yes, if that matches the tarball version
[04:03] <mdz> Kamion: done
[04:04] <mdz> er, not yet
[04:04] <mdz> it drops me to a shell at one point
[04:04] <mdz> and I don't actually know what to do with it
[04:04] <mdz> it looks like everything is in order
[04:04] <Kamion> jdub: nor I, assuming it goes out on stderr
[04:05] <Kamion> mdz: I think it's for you to check it over, and also possibly to clean up old subdirectories of daily-installer-*
[04:05] <jdub> Kamion: oh well; the others will be fixed in a couple of minues
[04:05] <Kamion> jdub: cool
[04:05] <Kamion> thanks
[04:06] <Kamion> I wonder where jbailey went to; I wanted to discuss the base-installer RESUME= fix with him
[04:06] <mdz> hmm
[04:06] <mdz> Kamion: it drops me into lisa, but i'm not sure what I should tell lisa to do with these
[04:06] <mdz> so I'm going to leave them in queue/byhand for now
[04:07] <Kamion> lisa = ?
[04:07] <Kamion> oh, NEW processor
[04:09] <mdz> Kamion: stuff should be synching now
[04:09] <Kamion> mdz: I think you just tell it to accept
[04:10] <Kamion> it checks whether byhand components are still present, and does not offer [A] ccept if so
[04:10] <mdz> oh, ok
[04:10] <mdz> accepted, then
[04:11] <mdz> notified elmo
[04:24] <opi> jdub, ping? ;)
[04:25] <jdub> opi: pong
[04:28] <opi> jdub, query :-)
[04:35] <Kamion> http://people.ubuntu.com/~cjwatson/base-installer.resume.diff <- should make suspend-to-disk work out of the box, still testing
[04:35] <Kamion> tested with one swap partition and it's fine, I'll test with >1 and with none as well
[04:37] <bluefoxicy> is there some way i can tell if my cds were sent
[04:37] <bluefoxicy> I ordered some CDs for warty from shipit back in november
[04:37] <bluefoxicy> after noticing a message that it was "Reopened by popular request" or something
[04:38] <opi> Kamion, S-T-D is planed for Hoary, right?
[04:38] <bluefoxicy> opi:  protection lanned for Hoary :)
[04:38] <Kamion> opi: yes
[04:38] <opi> Kamion, yeah, super -- it's really missing
[04:39] <Kamion> I'm trying to make this work for preview even though it's very late, because it needs installer support and a lot of people will install with the preview and upgrade to final
[04:40] <Kamion> opi: it should work now if you put a suitable RESUME= in /etc/mkinitrd/mkinitrd.conf and regenerate your initrd
[04:40] <wasabi> haha so how long until the live cd is able to save data to itself? :)
[04:40] <opi> Kamion, remind me, resume should point to partiton that is swap type?
[04:41] <wasabi> It'd be a pretty easy interface edition, at log off check HAL to see if hte device the cd is in is a cd-rw, etc, then ask user if he wants to save. ;)
[04:41] <Kamion> opi: yes
[04:42] <opi> Kamion, is snapshoting to file impossible (or, hard to achive?)
[04:42] <Kamion> opi: no idea
[04:42] <Kamion> I'm no swsusp expert
[04:42] <opi> Kamion, because it would be hard to change my partiton table now
[04:42] <opi> or can I use my old swap
[04:42] <opi> because it's up before swapon /dev/xxx
[04:42] <opi> Kamion, ok, I'll google
[04:43] <Kamion> you can use the swap partition you're already using for swap
[04:43] <opi> ok, then I'm home
[04:43] <opi> :-)
[04:43] <Kamion> or whatever, really; it just has to be a scratch partition
[04:43] <Kamion> as far as I know
[04:43] <opi> thanks
[04:44] <opi> in that case, a hackinsh solution would be to mount file as a loop device and try to save it thre ;-)
[04:44] <Kamion> I doubt you'd be able to resume if you tried that
[04:44] <opi> well, what about initrd?
[04:45] <Kamion> if you feel like teaching it how to do that ...
[04:45] <opi> nah, I'm just doing a blind guesses
[04:45] <opi> I guess I'll better patch kernel myself and test it
[04:45] <Kamion> the initrd is what implements resuming
[04:46] <opi> but there's a part of the script that resumes the kernel state, no?
[04:46] <Kamion> however, /sys/power/resume only supports block devices
[04:46] <opi> ah, so that's seattled
[04:46] <Kamion> I suspect you'd be in for some fairly major hacking
[04:58] <mdz> Kamion: is [ "$foo" ]  in general a good idea, in contrast to [ -n "$foo" ]  ?
[04:58] <mdz> seems like it could cause a syntax error depending on the value
[05:00] <Kamion> it does seem that way at first glance, but leaving out -n is fine; test(1) has some very strange argument handling rules
[05:00] <Kamion>     1 argument:
[05:00] <Kamion>         Exit true (0) if $1 is not null; otherwise, exit false.
[05:00] <Kamion> http://www.opengroup.org/onlinepubs/009695399/utilities/test.html
[05:00] <mdz> weird
[05:01] <mdz> I just built a new set of live CDs with esound
[05:01] <Kamion> one reason why avoiding -a and -o is a good idea is so that you can use the normal && and || argument handling rules rather than the strange test(1) rules
[05:02] <mdz> Kamion: localechooser ubuntu7 is the good one?
[05:02] <Kamion> mdz: yes
[05:02] <mdz> ok
[05:02] <mdz> current live build has all the goodies, then
[05:04] <mdz>    589543424 100%    9.44MB/s    0:00:59  (1, 33.3% of 3)
[05:04] <mdz> rsyncs nicely
[05:07] <Amaranth> hey, take a look at http://www.euphorochrome.com.nyud.net:8090/photoblog/images/131.jpg
[05:11] <dilinger> whoa, nice pic
[05:11] <zul> how is the release coming?
[05:12] <mdz> zul: needs testing
[05:12] <zul> i could test it when i get in tomorrow
[05:18] <zul> night though
[05:21] <mdz> hmm, I thought seb128 was going to make an upload to fix About Ubuntu
[05:22] <mdz> Mar 08 15:52:20 <seb128>        k, I'll upload in ~30min so
[05:23] <jdub> mdz: which fix?
[05:23] <mdz> jdub: About Ubuntu points to the ubuntu-artwork Warty page
[05:23] <jdub> mdz: that's hard to fix
[05:23] <jdub> well
[05:23] <jdub> unless i fix it very hackily, non-i18n style
[05:24] <jdub> which i'm happy to do
[05:25] <mdz> well, you're going to upload ubuntu-artwork anyway
[05:26] <mdz> so we may as well fix it there, at least for now
[05:31] <Kamion> why does resizing a swap partition even take any time at all?
[05:35] <mdz> Kamion: fwiw, I'm not able to reproduce the boot-di-from-USB-cdrom problem at the moment
[05:35] <mdz> not that I was able to consistently reproduce it before
[05:37] <Kamion> mdz: you mean #5732?
[05:38] <Kamion> mdz: adding udevstart to hw-detect may have made a number of mysterious bugs go away, or at least manifest differently
[05:39] <Kamion> base-installer change works fine with three swap partitions, the middle one being the biggest
[05:39] <Kamion> now to try with no swap partitions
[05:41] <mdz> Kamion: yes, #5732
[05:41] <mdz> Kamion: current generation of live CDs are good (login/logout sounds and all)
[05:42] <mdz> Kamion: also that "setenv boot-device cd:" seems to have done the trick; that'll save me gobs of time, thanks
[05:42] <Kamion> cool
[05:44] <Kamion> hooray, yes, it is
[05:45] <mdz> neat
[05:46] <mdz> would be neater if you could "Connect to server..." to an NFS server
[05:46] <mdz> neater still would be if there were a network filesystem which didn't suck ass
[05:48] <Kamion> connect to server> yes, why don't we have that, anyway?
[05:48] <crimsun> mdz: RE: #6810, I've chatted with only one person who uses the modem module (snd-intel8x0m). I agree with Chuck RE: blacklisting it.
[05:49] <mdz> Kamion: would *you* want to implement an NFS client in libgnomevfs?
[05:49] <Kamion> mdz: it doesn't just use mount(8)?
[05:49] <mdz> Kamion: no, it does the protocol itself
[05:49] <Kamion> oh
[05:50] <mdz> Kamion: hence being able to access SMB shares without root, anonftp, etc.
[05:50] <mdz> crimsun: ok, can you remind me about it after preview is out?
[05:50] <crimsun> mdz: absolutely.
[05:50] <Kamion> being able to do mount(8) from the GUI would be good
[05:52] <Kamion> (for non-gnomevfs-integrated stuff)
[05:53] <mdz> that reminds me, I need to remember to implement mounting of hard disk partitions in casper sometime
[05:53] <mdz> we should work out a way to share that code between casper and the installer
[05:53] <Kamion> rescue mode does that too
[05:53] <mdz> oh? hmm, maybe I can squeeze it in for hoary hten
[05:53] <Kamion> although its use case is different
[05:54] <Kamion> well, rescue mode just mounts the partition you tell it too actually, that's quite different
[05:54] <Kamion> s/too/to/ # gah
[05:54] <mdz> oh
[05:54] <mdz> what I want is to poke around like os-prober does
[05:54] <mdz> make up some reasonable names for what it finds, and mount using those names
[05:54] <Kamion> yeah, sorry, realised that belatedly
[05:54] <mdz> maybe it should be an extension to os-prober
[05:54] <Kamion> "like os-prober does"> ITYM "using os-prober" :-)
[05:55] <Kamion> os-prober produces quite usable output
[05:55] <mdz> is os-prober already generic enough?
[05:55] <mdz> oh, good
[05:55] <Kamion> should be
[05:55] <Kamion> it's used in several wildly different applications
[05:56] <Kamion> duplicating the logic on top of that isn't great but wouldn't be too bad for hoary
[06:03] <mdz> jdub: cliff said he would come by tonight so that we can finalize stuff for preview; have you heard from him in the past 1-1.5 hours?
[06:07] <mdz> Kamion: anything on your list besides base-installer?
[06:07] <Kamion> mdz: no
[06:09] <Kamion> mdz: oh, those scrollkeeper fixes, jdub said he'd upload them in a few minutes two hours ago
[06:10] <mdz> Kamion: I'm happy with the live CD except for artwork, and am ready for install testing when you are
[06:10] <Kamion> jdub: ping :-)
[06:10] <Kamion> yeah, nearly done with base-installer testing now
[06:10] <Kamion> how do I un-suspend-from-disk, anyway? pressing the power button just seems to reboot from scratch
[06:10] <mdz> that means it didn't find your RESUME
[06:11] <mdz> it works for me when I set RESUME manually
[06:12] <Kamion> traditional device name, right?
[06:12] <mdz> that's what I use
[06:12] <mdz> of course it needs to be on a device which is visible in initrd-land
[06:12] <mdz> though it's pretty unusual for swap to be on a different physical device than root
[06:13] <mdz> different controller, that is
[06:15] <Kamion> it all looks fine
[06:15] <Kamion> could be that the laptop is just odd; it is a Via after all
[06:16] <mdz> unfortunately I don't think it provides any debugging at all
[06:16] <mdz> Kamion: do you have /sys/power/resume?
[06:17] <Kamion> yes, in the real system
[06:17] <Kamion> how come initrd-tools' initrd does not mount /sys?
[06:18] <Kamion> oh, never mind, it does
[06:18] <Kamion> shrug, dunno, I'm pretty sure the configuration is correct now
[06:19] <mdz> Kamion: anything in dmesg?
[06:20] <mdz> if there's some problem reading the resume data, I assume the kernel would log
[06:20] <fabbione> morning
[06:22] <Kamion> oh, hey, that time it worked
[06:23] <m_tthew> ever closer to the 'totally' in 'totally rad'
[06:24] <mdz> the amd64 test box will suspend
[06:24] <mdz> but it doesn't quite resume
[06:24] <Kamion> ok, base-installer 1.15ubuntu7 uploaded
[06:24] <mdz> I think my jaw would have dropped if it had
[06:27] <mdz> apparently it supports cpu frequency scaling, too, which I hadn't noticed before
[06:32] <jdub> Kamion: those aren't up?
[06:32] <jdub> i have .upload files here...
[06:35] <fabbione> guys, what is the status for preview cd?
[06:37] <lamont> morning fabbione 
[06:37] <fabbione> hey lamont
[06:42] <Kamion> jdub: nothing on hoary-changes
[06:51] <m_tthew> I like the 'determine my layout by pressing some keys'
[06:53] <bradg> Will 2.6.11 be the default in the final release of hoary or does feature freeze dictate that we have to stick with 2.6.10?  I ask because 2.6.11 adds support for my sound card that isn't in 2.6.10.
[06:56] <lamont> 2.6.10
[06:56] <lamont> although 2.6.11 will be in universe
[06:56] <Kamion> jdub: ah, there it is. just gnome-applets?
[06:57] <jdub> Kamion: also gnome-doc-utils
[06:57] <jdub> although i'm double checking my fix, because i can't believe it
[06:58] <fabbione> speaking of which....
[06:58] <fabbione> we could actually go 2.6.11.1
[07:02] <stub> If I suspend my laptop whilst playing music with Rythmbox, after I resume Rythmbox can no longer make a sound (could not pause playback, could not open resource). Should I file a bug against rythmbox, acpi-support or polypaudio ?
[07:02] <Kamion> elmo: probably best not to bother processing the next daily d-i build
[07:05] <bradg> jdub, Has there been any progress on the move to dbus-0.23.2?  Are there API incompatibilities that are sticking us to 0.23.0 for hoary?
[07:06] <jdub> Kamion: hrm, so the ugly ones are easy, but the last one i can't be sure of
[07:06] <jdub> bradg: there's an api change which we could stub, and may do so before final if it's worth it
[07:07] <Kamion> jdub: nod
[07:08] <bradg> jdub, The major benefit I see is just that it will allow for wider testing recent versions of beagle.  Anyway, good luck on the release tomorrow!  Ubuntu has been great (the software but especially the community)!  I'm glad to be involved (Web design contest - heh, it's something).
[07:10] <jdub> :)
[07:17] <fabbione> doko: ping
[07:23] <Kamion> jdub: I think we can live without the last one
[07:27] <Kamion> mdz: decided not to wait any longer, new install CDs building
[07:28] <Kamion> we have to rebuild later for ubuntu-artwork anyway
[07:28] <fabbione> Kamion: can you give me the green light for rsync and start testing?
[07:28] <Kamion> mdz: go ahead and test when ready; I'm going to get a little sleep
[07:29] <Kamion> fabbione: no, see above :)
[07:29] <Kamion> fabbione: now + 1 hour should be safe
[07:29] <fabbione> Kamion: roger that and good night houston. Apollo 13 out.
[07:29] <Kamion> erk, cancel that, I have to wait for the kubuntu dailies to finish
[07:30] <Kamion> oh, no I don't. whoops.
[07:31] <Kamion> really building
[07:35] <mdz> Kamion: you were waiting?
[07:58] <fabbione> night lamont
[08:01] <mako> fabbione: can you point sabdfl to that preview announcement when you see him?
[08:01] <fabbione> mako: sure
[08:01] <mako> fabbione: i think i told i'd put it there but i suspect he'll forget :)
[08:01] <mako> and i'm gonna crash
[08:01] <mako> but i'll be up soon
[08:01] <fabbione> mako: ehehe ok.. good night
[08:06] <jdub> mdz: do you want the (non-i18n-happy) About Ubuntu change done?
[08:06] <mdz> jdub: meaning we replace the existing .html with a new .html?
[08:07] <jdub> no
[08:07] <jdub> we change what it launches
[08:07] <jdub> to open the english-only docteam version
[08:07] <mako> mdz: ok, you saw the announcement url
[08:07] <mdz> mako: I didn't, but I have now
[08:07] <mako> mdz: good enough
[08:08] <mdz> mako: maybe swap the first two paragraphs?
[08:08] <mako> mdz: fine with me.. i'm not going to touch it for a bit.. i need to crash
[08:08] <jdub> mako: "automated install support with Kickstart compatibility"?
[08:08] <mdz> mako: ok, thanks
[08:11] <Burgundavia> mako: I had the same thought regarding the 1st two paras
[08:11] <mako> i'm really not partial to the order of the paragraphs :)
[08:12] <Burgundavia> mako: future announcements of this type? could that be changed to something more specific?
[08:13] <pitti> Morning
[08:13] <mako> Burgundavia: could just be future announcements
[08:17] <Burgundavia> mako: ok, why not say future release announcments?
[08:17] <jdub_> mdz: about ubuntu change? i have to do it soon
[08:18] <mdz> jdub_: why do you want to change the panel to launch a different program, rather than swapping in a new HTML file?
[08:18] <jdub_> because the new about ubuntu doc is docbook, to be viewed in yelp
[08:19] <jdub_> i could render it to html and swap it in
[08:19] <mdz> it's trivial to convert it into HTML
[08:19] <jdub_> but it will look a bit arsey
[08:19] <mdz> if the panel change is absolutely simple and safe, go for it
[08:19] <mdz> I just don't want to destabilize things there
[08:21] <dholbach> good morning
[08:23] <pitti> Hi dholbach 
[08:24] <dholbach> morning pitti
[08:31] <m_tthew> fabbione : looks like the build is up
[08:31] <fabbione> m_tthew: thanks.
[08:31] <mdz> yep
[08:32] <mdz> rsyncing it now
[08:32] <mdz> we're still going to need a gnome-panel and ubuntu-artwork from jdub before this is over
[08:32] <mdz> jdub_: did you find out what happened to those two lost uploads?
[08:32] <fabbione> well
[08:32] <m_tthew> this will be the last build I can test before sleep
[08:33] <fabbione> jdub_: wrong target for the upload?
[08:33] <mdz> the key for this build is working hibernate
[08:34] <fabbione> sparc is catching up pretty fast
[08:35] <fabbione> but gcc-4 has a wrong build-dep
[08:35] <fabbione> that means rebuilding it again after all the libs are done
[08:35] <mdz> jdub_: ah, I see them now. gnome-applets and gnome-doc-utils
[08:36] <mdz> hmm, no, I'm seeing seb128's gnome-doc-utils upload
[08:36] <mdz> jdub_: do you have a gnome-doc-utils upload to make, or no?
[08:36] <jdub_> no
[08:36] <jdub_> decided not to
[08:36] <mdz> ok
[08:36] <mdz> so only ubuntu-artwork and gnome-panel?
[08:36] <jdub> yes
[08:38] <mdz> once you've uploaded those, I'll need to do livefs builds, and then a full set of install+live CDs
[08:39] <mdz> so I'm going to try to take a nap until then
[08:39] <mdz> please SMS me if there are any issues
[08:39] <jdub> going to be > 1hr
[08:45] <doko> fabbione: pong
[08:45] <fabbione> doko: gcc-4 FTBFS on sparc... interested in the log?
[08:45] <fabbione> it looks to me it needs a newer version of libcairo... 
[08:45] <fabbione> if so it needs a versioned build-dep
[08:45] <fabbione> but i might be 100% wrong here
[08:47] <doko> it builds ok with the version in hoary on the other archs ... 0.3.0-1
[08:47] <fabbione> fuck irc!
[08:48] <fabbione> i hate this protocol
[08:48] <fabbione> yeah i had an older version of cairo
[08:48] <fabbione> i will kick it back as soon as libcairo will build
[08:48] <fabbione> but it should still have a better version build-dep imho
[08:48] <fabbione> if that's the problem
[08:49] <doko> ok, added for the next version
[08:49] <fabbione> i will let you know to be 100% sure
[08:49] <Treenaks> anyone with basic knowledge of US legalese here? :)
[08:49] <fabbione> it will take sometime before i can actually kick back gcc
[08:52] <Burgundavia> Treenaks: I might be able to help
[08:55] <Treenaks> Burgundavia: can I /msg you? :)
[09:15] <pitti> Moin mvo
[09:16] <mvo> hey pitti! good morning
[09:19] <dholbach> hai mvo
[09:20] <mvo> hey dholbach 
[09:25] <amu> moins 
[09:29] <pitti> erm, why does cdrecord ignore my speed attribute (speed=4) ???
[10:04] <carlos> pitti: when is the base language pack released? with the Ubuntu preview or with final release?
[10:04] <pitti> carlos: I uploaded up-to-date packs yesterday night
[10:04] <carlos> the *final* base language pack
[10:05] <carlos> pitti: I know, that's why I'm asking
[10:05] <pitti> carlos: right before the final Hoary version, I will update them again, of course
[10:05] <carlos> ok, that was the question
[10:05] <carlos> thanks
[10:05] <pitti> carlos: btw, the tarballs of yesterday's rebuilds are in ~lamont/translations/20050306
[10:05] <pitti> carlos: that's a "fake" directory :-)
[10:05] <pitti> carlos: I also generated an index file, so you should be able to import it normally
[10:06] <carlos> perfect, thanks
[10:06] <pitti> carlos: btw, yesterday I defined a lot of domain overrides, so that I now have a complete set for hoary/main
[10:06] <pitti> carlos: do you have any use for it?
[10:06] <pitti> s/it/them/
[10:07] <carlos> pitti: please, send me them, it will help while cleaning up the first main import
[10:07] <pitti> carlos: you can always grab the latest ones from rookery
[10:07] <pitti> carlos: /home/pitti/strip2lparc/domain-overrides
[10:08] <pitti> carlos: there are some 0 byte files, that means that this source package is completely ignored
[10:08] <pitti> carlos: this is necessary sometimes
[10:09] <carlos> ok
[10:09] <pitti> carlos: like, e. g. flex-old and flex have the same domain (flex), so I cannot install translations for both of them
[10:10] <carlos> pitti: does it means we have two packages with the same domain but with different .pot and .po files?
[10:11] <pitti> carlos: yes, flex-old is an -- erm -- older version
[10:11] <pitti> carlos: I only use the translations from the newer flex
[10:12] <carlos> and how is that we are supporting flex-old if it's not being developed anymore?
[10:12] <pitti> carlos: it's a build-dep of some pacakges
[10:13] <carlos> pitti: what about patching it so it uses flex-old domain name?
[10:14] <pitti> carlos: that would be possible; but I didn't do this last night, too much other stuff
[10:14] <pitti> carlos: besides, nobody will care about translations for flex-old anyway
[10:15] <carlos> pitti: I know, but that makes us handle an special case so doing the rename could be better than just ignore it
[10:16] <pitti> carlos: you need ignore-overrides anyway for some odd KDE packages :-)
[10:17] <carlos> pitti: well, in this case I prefer if you don't create the tarballs for those packages than ignore them from my side :-)
[10:17] <pitti> carlos: hmm, that's more difficult
[10:17] <pitti> carlos: I can't know which domains collide before I actually build the tarballs
[10:17] <pitti> argh
[10:17] <carlos> don't worry, I need to talk with Mark about it today
[10:18] <carlos> I suppose we will find an easy solution 
[10:18] <pitti> Kamion, mdz: ^ can we fix this for preview?
[10:22] <Mithrandir> doko: pong
[10:25] <doko> Mithrandir: built a binutils-i486 for ia64, needs now something like kernel-headers-i486 and glibc-headers-i486, not sure if that is less ugly than including the gcc source in the ia32-libs tarball.
[10:26] <Mithrandir> doko: that's horrible, it's way better to just include the gcc source in the ia32-libs tarball.
[10:27] <Mithrandir> imho, at least
[10:28] <Keybuk> hmm, oops; anyone using aptitude will now have broken sound
[10:28] <doko> so where was the problem building lib32gcc1 from ia32-libs?
[10:29] <Keybuk> (because polypaudio provides esound, so that satisifies the dependency of ubuntu-desktop ... so you don't get esound, just a polypaudio without the alsa or x11 drivers)
[10:30] <Mithrandir> doko: it's not really a problem, I think I just asked whether gcc could generate lib32gcc1 on ia64 :)
[10:33] <trukulo> jdub, hei, you awake?
[10:33] <trukulo> morning to everybody
[10:33] <Treenaks> hi trukulo 
[10:33] <jdub> yes
[10:33] <jdub> very busy atm
[10:34] <jdub> leave it a couple of hours :)
[10:34] <trukulo> k jdub, just to say bittorrent 4 has multiple downloads and gtk interface
[10:35] <doko> Mithrandir: no
[10:36] <pitti> jdub: why does the live cd (ppc at least) use esd, while the install cd uses polypaudio?
[10:36] <jdub> pitti: install cds probably haven't finished building yet
[10:36] <jdub> or there's a desync
[10:36] <pitti> jdub: oh, did we switch back to esd finally?
[10:37] <carlos> jdub: do you know that with polypaudio the welcome and session end sounds are not played anymore?
[10:37] <mpt_london> carlos: That's not a bug, it's a feature
[10:37] <Keybuk> jdub: read my comment above
[10:37] <Keybuk> polypaudio provides esound
[10:37] <carlos> mpt_london: ;-)
[10:37] <Keybuk> so ubuntu-desktop still depends on polypaudio
[10:38] <jdub> carlos: yes
[10:38] <jdub> Keybuk: mdz put in a fix for that
[10:38] <jdub> mdz will be back again soon
[10:38] <Keybuk> jdub: where?  it wasn't in his 01:45 upload
[10:39] <trukulo> anyone but jdub who knows anything about torrent support in hoary? i mean, wouldn't be good to include bt 4.0.0 ?
[10:39] <Treenaks> trukulo: I think this'll be the same as the firefox 0.93/1.0 thing in warty :(
[10:39] <fabbione> hmmm no
[10:39] <fabbione> there is also a licence change with bt that needs to be investigated afaik
[10:39] <Treenaks> fabbione: no?
[10:40] <fabbione> + UVF = NO NO NO
[10:40] <fabbione> it will stay as it is
[10:40] <Treenaks> fabbione: that's what I said..
[10:40] <fabbione> i was answering to trukulo :)
[10:41] <mdz> Keybuk: ubuntu-desktop doesn't depend on polypaudio; polypaudio provides: esound
[10:41] <trukulo> fabbione, ok, i understand, only say that new bt have multiple downloads, and gtk interface
[10:41] <fabbione> so does the actual one :-)
[10:41] <trukulo> exactly what you was looking for in u-d-l
[10:41] <trukulo> no, the actual one use wxwidgets
[10:41] <trukulo> as i know
[10:42] <trukulo> and no multiple downloads in the same window
[10:42] <fabbione> just goodies.. 
[10:42] <fabbione> it's nothing critical for release
[10:43] <Keybuk> mdz: yes, that's what I was just saying
[10:43] <trukulo> fabbione, i agree, i was only questioning ;)
[10:43] <Keybuk> mdz: so you end up with polypaudio instead of esound sometimes
[10:43] <pitti> mdz: right now install uses polypaudio, live uses esd
[10:43] <Keybuk> apt (at least aptitude) doesn't appear to favour real packages instead of providers
[10:44] <pitti> mdz: at least on the ppc CDs from this morning
[10:44] <pitti> mdz: but polypaudio is broken again for ppc/install on the current CDs :-(
[10:45] <mdz> Keybuk: apt and aptitude shouldn't enter into it; only esound should be on the CD
[10:45] <mdz> looking at it now
[10:45] <Mithrandir> mdz: make ubuntu-desktop have a versioned depends (>= 0) on esound?
[10:45] <mdz> pitti: there is no polypaudio on the daily/current CDs
[10:46] <mdz> Mithrandir: scroll back a few hours; we talked about this
[10:46] <pitti> mdz: I just installed this morning's ppc/install without network (no chance of downloading), and this uses polypaudio
[10:46] <mdz> pitti: md5sum?
[10:46] <Keybuk> mdz: yeah, but at the point the cd stuff gets installed, it's already added the network to sources.list hasn't it
[10:46] <mdz> mdz@little /srv/cdimage.no-name-yet.com/www/full/daily/current $ grep polypaudio *.list
[10:46] <mdz> mdz@little /srv/cdimage.no-name-yet.com/www/full/daily/current $
[10:47] <mdz> Keybuk: nope
[10:47] <mdz> it explicitly does the desktop install with only the CD in sources.list
[10:47] <Keybuk> there's also the fact that ubuntu-desktop breaks polypaudio without forcing esound
[10:47] <pitti> mdz: 642877440 2005-03-09 08:43 hoary-install-powerpc.iso
[10:47] <pitti> mdz: 54baddb324353badb9be9bf7ce9434b3  hoary-install-powerpc.iso
[10:47] <Keybuk> so anyone tracking needs to make sure they replace polypaudio with esound manually
[10:48] <mdz> 2e33c1f0cccb268a4605399d3b2ac5b9  hoary-install-powerpc.iso
[10:48] <mdz> pitti: ^^ that's current
[10:48] <mdz> Keybuk: we discussed all of this hours ago; it's not important for the preview release
[10:48] <mdz> we'll deal with the upgrade issue later
[10:48] <mdz> pitti: how long ago?
[10:48] <pitti> mdz: downloaded 2 hours ago (7:43 UTC)
[10:49] <mdz> mdz@little /srv/cdimage.no-name-yet.com/www/full/daily $ grep 54baddb324353badb9be9bf7ce9434b3 */MD5SUMS
[10:49] <mdz> 20050308.1/MD5SUMS:54baddb324353badb9be9bf7ce9434b3  hoary-install-powerpc.iso
[10:49] <mdz> mdz@little /srv/cdimage.no-name-yet.com/www/full/daily $ ls -l current
[10:49] <mdz> lrwxrwxr-x    1 cjwatson cdimage         8 Mar  9 07:24 current -> 20050309
[10:49] <mdz> you have 20050308.1
[10:49] <mdz> 20050309 is current
[10:49] <pitti> bah
[10:49] <pitti> maybe the current symlink was wrong then?
[10:50] <Keybuk> mdz: I wasn't awake hours ago :p
[10:50] <mdz> Keybuk: we're going to remove the provides from polypaudio
[10:50] <Keybuk> yeah, that's pretty much what I'd do
[10:50] <mdz> Keybuk: I wasn't particularly awake myself :-P
[10:51] <Keybuk> "Hoary Hedgehog Preview Release ... brought to you by CAFFIENE!"
[10:52] <mdz> jdub: ETA for ubuntu-artwork?
[10:53] <jdub> mdz: conferring with mark/jane atm
[10:53] <jdub> er, in a moment
[10:54] <dholbach> hi seb128 
[10:54] <Treenaks> Good morning France! :)
[10:54] <jdub> yoseb
[10:54] <jdub> seb128: see gnome-panel upload
[10:54] <pitti> Hi seb128 
[10:54] <seb128> hey hey
[10:54] <seb128> jdub: thanks for the applets & panel love :)
[10:54] <jdub> seb128: we'll fix that before preview :)
[10:54] <jdub> erer
[10:54] <jdub> er
[10:54] <jdub> eeek
[10:54] <jdub> final
[10:54] <seb128> fix what ?
[10:55] <jdub> the panel hack
[10:55] <jdub> (it is not i18n aware, due to doc package)
[10:55] <seb128> that's not a hack
[10:55] <jdub> but ok for preview ;)
[10:55] <seb128> oh, right
[10:55] <jdub> it so is dude ;)
[10:55] <seb128> but there is no translation of that atm
[10:55] <jdub> we should do yelp ghelp:about-ubuntu or whatever
[10:55] <jdub> yeah
[10:55] <seb128> bah, a .desktop file doesn't seems to be a hack
[10:55] <jdub> hopefully taht'll be fxied for final too ;)
[10:56] <seb128> do we already have the preview images ?
[10:56] <mdz> seb128: we are waiting for ubuntu-artwork from jdub
[10:57] <mdz> and then we should be able to test the next (and possibly last) candidate
[10:57] <seb128> bah, that's only a preview
[10:57] <seb128> (this panel menu not localized bother me)
[10:57] <pitti> mdz: is it possible to s/Warty Warthog/Hoary Hedgehog/ on the "About Ubuntu" html page?
[10:57] <pitti> mdz: which package provides this?
[10:57] <jdub> pitti: we don't need to
[10:57] <mdz> pitti: jdub's gnome-panel upload points it to a new file
[10:57] <jdub> pitti: it's going to be removed anyway
[10:57] <pitti> ah, ok
[10:57] <pitti> cool
[10:58] <seb128> pitti: you have not planned any language-pack update before preview ?
[10:59] <pitti> seb128: I uploaded all the crack about midnight
[10:59] <pitti> seb128: please don't tell me you uploaded packages with heaps of new strings afterwards ....
[10:59] <seb128> no
[10:59] <mdz> pitti: are you still having problems with your keyboard layout on the live CD?
[10:59] <pitti> seb128: I can always publish new ones, but they need to be built and put into the CDs
[10:59] <seb128> I'm just bothered to have a "System" menu not localized in the middle of my french GNOME
[11:00] <pitti> mdz: yes, same thing
[11:00] <pitti> mdz: I replied to the bug
[11:00] <seb128> and that's probably the same for all the different languages :p
[11:00] <mdz> pitti: but it works on install?  that is very strange if so
[11:00] <Treenaks> seb128: same here in Dutch
[11:00] <seb128> that's a really bad effect
[11:00] <pitti> mdz: maybe the reason is that xorg.conf is readonly on the live CD?
[11:00] <pitti> mdz: yes, it works (almost) fine on install; for install it only configures pc104 instead of pc105, but this is a minor thing
[11:00] <mdz> pitti: if that were so, X would not work at all
[11:01] <pitti> hmm, right
[11:01] <Treenaks> seb128: someone added a comment about this here: https://bugzilla.ubuntu.com/show_bug.cgi?id=7285                                
[11:01] <mdz> pitti: can you send me /var/cache/debconf/config.dat from the live CD?
[11:01] <pitti> mdz: yes, let me boot it
[11:01] <seb128> Treenaks: I know, I don't understand anything of this "Ad ..:" and I've read the comment several times
[11:02] <Treenaks> seb128: it's just backreferences.. "Ad 1" refers to [1] 
[11:02] <pitti> mdz: btw, keyboard on text consoles is fine
[11:02] <seb128> Treenaks: that's no sense
[11:03] <mdz> pitti: yes, that's handled by a less hairy mechanism :-)
[11:03] <seb128> Treenaks: "- Add/Remove Programs[2] " == "Ubuntu-package: not translated at all (isolate i18n-strings and use pygettext to create po-template)"
[11:03] <seb128> Treenaks: what does pygettext does in that ?
[11:03] <Treenaks> seb128: he's a gnome person.. not a lot of ubuntu experience
[11:08] <mdz> jdub: how is the GNOME release?
[11:10] <pitti> mdz: http://people.ubuntu.com/~pitti/config.dat-ppc-live-german
[11:11] <mdz> Name: debian-installer/keymap
[11:11] <mdz> Template: debian-installer/keymap
[11:11] <mdz> Value: mac-usb-de-latin1-nodeadkeys
[11:11] <mdz> Owners: base-config, d-i
[11:11] <mdz> that looks reasonable
[11:11] <pitti> ack
[11:11] <pitti> mdz: btw, daniels' detection script works fine, too
[11:12] <mdz> pitti: that's so strange
[11:12] <pitti> $ sudo ./detect-keyboard.sh
[11:12] <pitti> layout is de
[11:12] <pitti> rules are xorg
[11:12] <pitti> model is pc104
[11:12] <pitti> options are nodeadkeys
[11:12] <mdz> pitti: look at /var/lib/dpkg/info/xserver-xorg.config, line 1715
[11:12] <mdz> this should be the same code as detect-keyboard.sh
[11:13] <mdz> I don't see an entry in the case statement which corresponds to mac-usb-de-latin1-nodeadkeys
[11:13] <mdz> oh, I see, this fallback:
[11:13] <mdz>     *--de*) XMAP="de"; OPTIONS="nodeadkeys";;
[11:13] <mdz> that is based purely on LANG
[11:14] <mdz> so perhaps $LANG is not set properly at that point
[11:14] <mdz> mvo: do you have this problem as well?
[11:14] <mdz> it seems like it may be specific to the mac-* layouts
[11:14] <pitti> mdz: maybe, locale in the running system is correct; maybe it is read too late
[11:14] <mvo> mdz: I had it, but I can't test right now, my test-systems cdrom seems to be b0rked today
[11:14] <pitti> mdz: however, the real fix seems to be to add the mac-usb keymaps then
[11:15] <pitti> mdz: since we already saw that locale-based choice is wrong
[11:15] <mdz> pitti: the question is whether $LANG is set in d-i
[11:15] <mdz> I would expect so, but perhaps not
[11:15] <pitti> mdz: ah, that's the reason why the detection script works in the running system
[11:15] <pitti> because LANG is correct then
[11:15] <mdz> pitti: could you paste our conversation into the bug so that we do not forget?
[11:15] <pitti> yes
[11:15] <mdz> pitti: yes, it seems very likely that this is the source of the problem
[11:17] <mdz> mvo: are you on powerpc as well?
[11:18] <mdz> pitti: what is the bug#?
[11:18] <pitti> mdz: 7138
[11:18] <pitti> mdz: just posted the followup
[11:18] <mdz> thanks
[11:18] <mvo> mdz: no, i386. I could get access to a  ppc if it is required/urgent
[11:19] <mdz> mvo: I'm very interested to know if it happens for you on i386
[11:22] <mvo> mdz: live-cd image is downloaded in ~2 minutes, then I'll burn it on a CD-R (my testsystem has either problems with it's cdrom or my cdrws are all faulty now). I'll keep you updated
[11:22] <mdz> mvo: thanks
[11:23] <fabbione> mdz: do we have also the last gnome stuff?
[11:23] <mdz> fabbione: we have everything except ubuntu-artwork from jdub
[11:23] <fabbione> ok
[11:23] <Mithrandir> elmo: http://sources.redhat.com/ml/binutils/2004-09/msg00299.html ; do you have any thoughts on the patch or should I ask doko?
[11:23] <fabbione> hmm no
[11:24] <fabbione> they are not propagate to the CD yet, are they?
[11:24] <mdz> did Kamion mention at what time he intended to return?
[11:24] <mdz> fabbione: there are a few small things which are not on the CD yet, because I am waiting for ubuntu-artwork to do a full set of builds (live fs, install CDs, live CDs)
[11:24] <elmo> Mithrandir: h j lu patches are to be avoided unless they're a) approved, b) have had a two week cooling off period to ensure they haven't broken random shit
[11:24] <elmo> IMO
[11:25] <doko> elmo: approved by amodra
[11:25] <elmo> then it works for me
[11:25] <elmo> however, I hope you're not thinking we need this for hoary?
[11:27] <Mithrandir> it would be nice to have so ld would barf instead of falling over with an assertion error.
[11:27] <Mithrandir> collect2: ld terminated with signal 11 [Segmentation fault] 
[11:27] <Mithrandir> /usr/bin/ld: BFD 2.15 assertion fail ../../bfd/elflink.c:6081
[11:27] <elmo> Mithrandir: we're at preview - is it really worth changing the way ld dies now?
[11:27] <Mithrandir> elmo: ok, hoary+1, then.
[11:27] <fabbione> mdz: ok..
[11:27] <Mithrandir> you know binutils better than I do. :)
[11:28] <doko> first class funeral are important ;)
[11:28] <elmo> probably not, but I do bear H.J.Lu-scars
[11:30] <seb128> mdz, jdub: I've a gnome-panel with "System" localized in all the locale (the string comes from yelp 2.6) ... any change to get that in the preview ?
[11:31] <mdz> seb128: it would require that pitti do new langpacks
[11:31] <jordi> seb128: are the translations in hoary the official ones in the GNOME tarballs?
[11:31] <mdz> jordi: yes
[11:31] <jordi> cool
[11:31] <pitti> mdz: I can do update packages, they are relatively cheap
[11:31] <jordi> so everything but "System" will be translated?
[11:31] <mdz> pitti: they worry me, though
[11:31] <jordi> that's a very visible string
[11:31] <seb128> mdz: I know, but that's the most visible string on screen
[11:31] <pitti> mdz: let's get it perfect for final
[11:31] <mdz> pitti: because we have not fixed bug #164595
[11:31] <seb128> s/that's/that's like/
[11:32] <seb128> I'm a bit annoyed to ship a preview with a panel not localized correctly
[11:32] <pitti> mdz: we worked around this by a Pre-Dependency
[11:32] <seb128> you see 3 labels on the desktop and one of them is "System" in english
[11:32] <mdz> pitti: oh, ok
[11:32] <mdz> seb128: go ahead, pitti can update the langpacks
[11:32] <jordi> I curse jdub, I gotta sort a few things with sabdfl and he's  totally centered on him
[11:32] <pitti> mdz: so the -base package will be installed first in any case
[11:33] <seb128> mdz: thanks
[11:33] <pitti> seb128: ping me if the package is built
[11:33] <pitti> seb128: because I need the translation tarball
[11:33] <jordi> seb128: can you confirm ca does have that translation?
[11:33] <seb128> pitti: I know, don't worry
[11:33] <jordi> should be. It's "Sistema" in any case
[11:33] <mdz> seb128: so you were able to get translations for System in many languages from yelp?
[11:33] <seb128> jordi: 
[11:33] <seb128> msgid "System"
[11:33] <seb128> msgstr "Sistema"
[11:33] <jordi> thanks
[11:34] <jordi> I hope about ubuntu and stuff is preserved since warty
[11:34] <jordi> but anyway
[11:34] <seb128> mdz: 76 languages
[11:35] <mdz> jordi: you could help us by testing the live CD :-)
[11:36] <jordi> mdz: which doesn't have ca translations :(
[11:36] <pitti> mdz: am I right with my assumption that now is not a good time to upload a perl security update to Hoary?
[11:36] <seb128> jordi: nop, About Ubuntu is a temporary .desktop hack atm
[11:36] <jordi> I've tried, I tell you
[11:36] <mdz> pitti: that is correct
[11:36] <mdz> jordi: it doesn't?  which one did you try?
[11:36] <mdz> oh, right
[11:36] <jordi> mdz: but sure, point me to an image I can use and I will rush to test it
[11:36] <jordi> mdz: hmm array6
[11:37] <mdz> ca is not in the list for the live CD
[11:37] <mdz> we may have some space left to add more languages, but not much
[11:38] <jordi> mdz: oh dude YOU WOULD MAKE ME HAPPY
[11:38] <seb128> mdz: panel uploaded
[11:38] <seb128> pitti: I'll let you know when the build is ok
[11:38] <pitti> ok
[11:38] <mdz> pitti: after preview, can you check to see how many more langpacks we could fit on the live CD?
[11:38] <pitti> seb128: luckily I just improved langpack-o-matic to spit out minimal update packages without much effort
[11:38] <seb128> thanks guys for that, better to have the menu in user's locale :)
[11:38] <jordi> mdz: and I would start promoting those live CD's against "Catix" (knoppix based crappy Catalan distro) egemony in Catalonia
[11:39] <pitti> mdz: yes, of course; after preview I also wanted to do some -0ubuntu0 uploads for stripping the handful of big packages you selected
[11:39] <jordi> mdz: I mistakenly suggested people to download array6 and then we discoevered ca wasn't n
[11:39] <mdz> jordi: unfortunately I don't think we can change this for preview, but we will try to support more languages after
[11:39] <jordi> in
[11:39] <jordi> mdz: hrm. The other day I thought about having some extra cd images with some other langs
[11:40] <jordi> but if ca can go in final, hey that rocks :)
[11:42] <mdz> jordi: it would be very easy to build a customized live CD with ca support
[11:42] <mdz> jordi: http://www.ubuntulinux.org/wiki/LiveCDCustomizationHowTo
[11:42] <mdz> just a little bit of time and disk space
[11:43] <mdz> but hopefully we can support it by default for final
[11:44] <jordi> mdz: can I do something so you remember to have a look?
[11:44] <mdz> jordi: you can remind pitti in a few days :-)
[11:44] <pitti> mdz: so the battle for langpack CD space officially begins :-)
[11:45] <jordi> pitti: I am going to be like your shadow
[11:46] <mdz> jdub: I need an answer on ubuntu-artwork
[11:46] <mdz> jdub: it will take 2+ hours to get a candidate built once we say go
[11:47] <seb128> jdub: I've a patch ready for the submenus switches in System
[11:47] <pitti> Hi sivang
[11:47] <pitti> hrm
[11:47] <pitti> Morning sivang 
[11:49] <sivang> pitti: Morning :)
[11:51] <Amaranth> anyone know why gnome.org is down?
[11:51] <mdz> 2.10 release overload?
[11:51] <Treenaks> Amaranth: because they're releasing?
[11:52] <Amaranth> wow, i don't even think mozilla.org when down when firefox came out
[11:52] <Amaranth> well, not for this long
[11:52] <Amaranth> err, went
[11:52] <Amaranth> it's 5am
[11:53] <pitti> mvo: successful?
[11:53] <mdz> perhaps they were h4x0red
[11:53] <mvo> pitti, mdz: yes, booting with german as language gives me a pc105 german keyboard in X
[11:53] <mdz> at any rate, that'd be a question for #gnome or such
[11:53] <pitti> mvo: okay, it's the lack of detection support for mac-usb keyboards for sure
[11:54] <Amaranth> yeah, they're pretty idle, figures someone here would know
[11:54] <mdz> mvo: ok, so I think we are confident abotu the nature of pitti's problem
[11:54] <pitti> mvo: good to hear that i386 works fine :-)
[11:54] <mdz> and it affects relatively few systems
[11:54] <pitti> yes, that's relieving
[11:54] <pitti> it would be a shame if the live CD was broken for i386
[11:54] <pitti> because it is so much better than warty's :-)
[11:55] <mvo> yeah, the new live cd really rocks!
[11:56] <pitti> sivang: connection problems today?
[11:56] <mdz> mvo: it has this great new feature where it actually boots most of the time ;-)
[11:57] <pitti> mdz: indeed, the live CD failed for 75% of the people I gave it to
[11:57] <pitti> mdz: (warty's)
[11:57] <mdz> it failed for many people I know, too :-//
[11:57] <pitti> pretty embarassing...
[11:57] <mvo> mdz: yes, that's one of it's strong points :) I like the design too
[11:57] <sivang> pitti: since yesterday :-/ had a nice time of no problems at all, then suddenly 3 days ago things b0rk3d, cable infrastructure seems fine, so maybe this is only a problem with the ISP's routing etc..
[11:58] <sivang> pitti: so since about 3 days I having accesss that comes and goes with no apparent reason
[12:11] <mdz> I am unable to reach jdub or sabdfl
[12:11] <mdz> does anyone know what is going on?
[12:13] <jdub> mdz: we're on the phone
[12:13] <jdub> mdz: you're not going to like the results
[12:13] <jdub> mdz: brb
[12:14] <mdz> jdub: I'm OK with whatever as long as we don't delay preview further
[12:15] <jdub> is elmo around?
[12:15] <mdz> yes
[12:15] <pitti> seb128: gnome-panel_2.10.0-0ubuntu3_translations.tar.gz is present
[12:15] <pitti> seb128: that should be the right one?
[12:16] <mdz> jdub: whenever you guys would like to bring me into the loop regarding the release, I'm standing by.  running on little sleep over here
[12:16] <elmo> jdub: yes
[12:17] <jdub> mdz: will get off in a sec
[12:18] <seb128> pitti: correct
[12:19] <pitti> seb128: okay, I can build new update packs if everything is in
[12:19] <pitti> I mean _really_ :-)
[12:19] <seb128> as far as I'm concerned go for it
[12:20] <jdub> ok
[12:20] <jdub> mdz: we need to get clearlooks in main
[12:20] <jdub> gtk2-engines-clearlooks
[12:20] <mdz> beg your pardon?
[12:20] <jdub> yeah, serious
[12:21] <mdz> I am not sure that I am prepared to stay up an extra hour for this
[12:21] <jdub> can Kamion take over?
[12:21] <mdz> Kamion is dead asleep
[12:21] <seb128> jdub: you want to switch the default theme for preview ?
[12:21] <mdz> he was up until 6am or osmething
[12:21] <jdub> seb128: sabdfl has asked us to, yes
[12:21] <fabbione> that means seeding -> archive propagation -> cd rebuild
[12:22] <fabbione> + testing
[12:22] <fabbione> = 2 hours if we are all here
[12:22] <fabbione> (at least!)
[12:22] <seb128> jdub: we have the fix for the label colors ?
[12:22] <Amaranth> oh sure, now you work on clearlooks, after i install from source :P
[12:22] <mdz> fabbione: it is much more than that
[12:22] <jdub> well, gnome 2.10 is not out yet anyway
[12:22] <mdz> the live fs build alone is 20-30 minutes
[12:22] <jdub> Amaranth: it's in universe
[12:22] <seb128> Amaranth: it's in universe for a week or so
[12:22] <Amaranth> d'oh
[12:22] <jdub> seb128: we have a different gtk theme anyway
[12:22] <Amaranth> i installed from source 4 days ago
[12:22] <seb128> jdub: different of what ?
[12:22] <mdz> jdub: clearlooks is new code
[12:23] <jdub> mdz: it's well tested
[12:23] <seb128> jdub: you want to use clearlooks for what ? not gtk ?
[12:23] <mdz> jdub: not in Ubuntu it isn't
[12:23] <jdub> seb128: yes
[12:23] <Amaranth> isn't there already a clearlooks-human theme?
[12:23] <mdz> this is madness
[12:23] <jdub> mdz: you need to call sabdfl 
[12:23] <mdz> I SMSed him
[12:23] <jdub> i'm just the messenger
[12:23] <jdub> and i told him no one would like the message
[12:25] <pitti> jdub: does clearlooks involve any translations?
[12:25] <mdz> jdub: I would have liked it fine a week ago
[12:25] <jdub> pitti: no, it's a widget theme
[12:25] <jdub> mdz: i understand
[12:25] <jdub> mdz: i boringly and rigidly said no.
[12:26] <fabbione> jdub: is that why you are telling us to do it? ;)
[12:26] <Amaranth> sabdfl == Mark?
[12:26] <jdub> i also told mark that my testicles would be blue at UDU because of this
[12:26] <jdub> i'm fundamentally aware of the problem
[12:26] <mpt_london> Amaranth: yes
[12:26] <Amaranth> ah, that explains a lot
[12:27] <theine> Hi, are there any plans for including gnome-bluetooth into main?
[12:27] <pitti> theine: not for hoary
[12:28] <theine> ok
[12:28] <pitti> theine: can you please ask this again in some days?
[12:28] <pitti> theine: the folks are currently very busy with the preview release
[12:28] <theine> sure, no problem
[12:36] <ogra> morning
[12:36] <pitti> Hi ogra
[12:36] <mdz> jdub: spoke with Mark, I'm crashing, please do what needs to be done, see you in 7 hours
[12:37] <ogra> clearlooks hasnt a human colorset yet, i'm using it since about a week and it seems stable....
[12:37] <pitti> mdz: sleep well, mdz
[12:37] <ogra> g'night mdz
[12:37] <jdub> mdz: thanks; sleep very well
[12:37] <jdub> ogra: that's coming in ubuntu-artwork
[12:38] <dholbach> good night mdz
[12:38] <ogra> jdub, i just read up.....(just wated to take mdz's fear abour stability a bit) :)
[12:38] <seb128> 'night mdz 
[12:38] <ogra> s/abor/about
[12:39] <dholbach> hi ogra
[12:39] <jdub> ok, making seed change now
[12:41] <elmo> jdub: please tell me what to promote too - waiting for seeds to propogate will take time
[12:41] <jdub> gtk2-engines-clearlooks
[12:42] <elmo> ok, promoted
[12:43] <sivang> hi ogra , dholbach 
[12:43] <dholbach> hi sivan!
[12:43] <ogra> morning sivang 
[12:44] <jdub> so what we're doing
[12:44] <jdub> so everyone groks what's going on
[12:44] <jdub> is adding the clearlooks theme to the desktop (and thus the CD)
[12:44] <jdub> and using it as the default theme
[12:44] <jdub> to make maximum impact
[12:45] <jdub> though it does mean we delay
[12:45] <jdub> meanwhile, if anyone hadn't noticed, GNOME 2.10 hasn't been released yet because the servers are down
[12:45] <pitti> *sigh*
[12:45] <seb128> jdub: anybody knows what's going on with the servers ?
[12:45] <sivang> jdub: yeah, I thought it was becasue of the load on the servers?
[12:45] <jdub> so everyone's having fun tonight it seems :)
[12:46] <jdub> seb128: haven't got a reasonable answer yet
[12:46] <seb128> k
[12:46] <jdub> seb128: going to ring owen soon
[12:46] <fabbione> jdub: and what are the 2.10 pkgs we are shipping now?
[12:46] <jdub> fabbione: the latest seb did hours ago
[12:46] <jdub> fabbione: the ones that will be in the release
[12:46] <ogra> fabbione: sneak preview ;)
[12:46] <jdub> (the tarballs are gathered a couple of days before the release)
[12:46] <pitti> seb128: I need your help
[12:46] <seb128> fabbione: he's speaking about official announce for 2.10
[12:47] <jdub> ah, fuck it, going to call owen now
[12:47] <fabbione> seb128: ahhhhhh
[12:47] <seb128> fabbione: the tarballs/packages are the 2.10 ones
[12:47] <pitti> seb128: tarball for gnome-panel is broken (no mo files)
[12:47] <maswan> jdub: not the ftp server though, right?
[12:47] <fabbione> seb128: you rock!
[12:47] <pitti> seb128: and there is only one pot file "yelp.pot"
[12:47] <fabbione> seb128: even if you are a gtk bug!
[12:47] <seb128> pitti: are you kidding me ?
[12:47] <jdub> maswan: no
[12:47] <pitti> seb128: is the domain really "yelp"?
[12:47] <jdub> ah
[12:47] <jdub> fuck
[12:47] <jdub> man
[12:48] <maswan> Good, had it been the ftp server, I should have heard of it (and been working hard on fixing it) by now.
[12:48] <pitti> seb128: is the domain "gnome-panel"?
[12:48] <seb128> pitti: correct
[12:49] <seb128> pitti: the yelp.pot has been copied with the System strings from yelp
[12:49] <seb128> pitti: but I don't get the "there is no .mo"
[12:49] <seb128> I've not changed anything out of adding a string to the .po files
[12:50] <pitti> seb128: erm, I just found an older tarball, there it is "gnome-panel-2.0"
[12:50] <seb128> pitti: arg
[12:50] <seb128> lemme check
[12:50] <pitti> seb128: no, really, in the latest upload there is a "yelp.pot"
[12:50] <seb128> pitti: <seb128> pitti: the yelp.pot has been copied with the System strings from yelp
[12:50] <Kamion> fuck, sorry, overslept; anything I need to do now?
[12:50] <fabbione> Kamion: no.. we released already
[12:50] <seb128> pitti: I can fix it since we are delayed for the gtk2-engines stuff
[12:51] <ogra> Kamion: waiting
[12:51] <pitti> seb128: find -name "*.mo" -exec basename '{}' \; | sort -u
[12:51] <ogra> Kamion: for inclusion of clearlooks in main
[12:51] <Kamion> fabbione: I've read enough scrollback not to be fooled by that :)
[12:51] <pitti> seb128: no need to fix it, I just need the correct domain
[12:51] <Kamion> 11:42 < elmo> ok, promoted
[12:51] <fabbione> Kamion: :P
[12:51] <pitti> seb128: I think it is gnome-panel-2.0
[12:51] <seb128> pitti: gnome-panel-2.0
[12:51] <seb128> pitti: let me fix it
[12:51] <seb128> pitti: there is no hurry due to clearlooks
[12:51] <pitti> seb128: no worries, I add an override
[12:51] <elmo> Kamion: if we're going to need another d-i for this (can't see why), lemme know, as we still have this morning's cron.daily in q/byhand
[12:52] <Kamion> elmo: nope, best not
[12:52] <pitti> lamont: ping?
[12:52] <Kamion> jdub: have you done ubuntu-meta, or do you want me to?
[12:53] <jdub> Kamion: please do
[12:56] <Kamion> will have to wait until cron.daily
[12:59] <jdub> when's that?
[12:59] <Kamion> +4mins
[01:01] <jdub> making related u-a changes
[01:01] <Kamion> jdub: what needs to change to make clearlooks the default theme, rather than just present? just u-a?
[01:01] <jdub> yes
[01:01] <jdub> it will be called Human
[01:01] <jdub> so no default changes or anything
[01:01] <sivang> what's with gamin_server , everytime I do an upgrade it eats CPU cycles and apparently scans all my drives, bringing the system to a painful slowdown..
[01:02] <jdub> (*phew*)
[01:02] <pitti> seb128: "there is no .mo" -> not your fault
[01:02] <seb128> pitti: oh ?
[01:02] <pitti> seb128: either one buildd is outdated, or there is another bug in pkgstriptranslations
[01:02] <seb128> pitti: BTW I'm doing an upload now to fix that
[01:03] <pitti> seb128: what was wrong? how did the yelp.pot file came into the g-panel source package?
[01:03] <Kamion> man, I had dreams last night that I had to fix installer bugs for preview
[01:03] <seb128> pitti: the 76 "System" translations come from yelp
[01:04] <fabbione> Kamion: and did you write down the patches before waking up?
[01:04] <seb128> pitti: I've made a quick python script to grab "System" in all the po files and copy it with the translation in the panel po files
[01:04] <seb128> pitti: and it has grabbed the .pot with them
[01:04] <pitti> ah, ok
[01:04] <seb128> look on the .pot :p
[01:04] <pitti> seb128: because older tarballs do not ship a pot file at all (which is another bug, because the Rosetta guys need it)
[01:05] <seb128> k
[01:05] <Kamion> fabbione: nope :)
[01:05] <pitti> seb128: but I don't need the pot for langpack-o-matic, so this has time
[01:05] <Kamion> are there more langpack changes needed for preview?
[01:07] <pitti> Kamion: yes, I was just asked to upload new update packages for the gnome-panel changes
[01:07] <seb128> Kamion: I've added the "System" translations in 76 languages to the panel, better to have the menu title translated
[01:07] <Kamion> ok, tell me when those are done since I won't see them on hoary-changes
[01:08] <Kamion> jdub: intentional that indubstrial's still in desktop?
[01:09] <jdub> industrial? yes
[01:09] <Kamion> er, yeah, that
[01:09] <Kamion> ok
[01:09] <pitti> Kamion: ok
[01:18] <Kamion> ubuntu-meta done, too
[01:18] <Kamion> jdub: will ubuntu-artwork make this cron.daily?
[01:19] <jdub> could do
[01:19] <jdub> no
[01:25] <jdub> Kamion: uploading now
[01:25] <fabbione> Kamion: what about desktop-sparc?
[01:25] <fabbione> (ubuntu-meta)
[01:26] <koke> seb128: the System panel menu keeps appearing to me as "Desktop" (and untranslated)
[01:26] <koke> seb128: using 2.10.0-0ubuntu3
[01:26] <Kamion> fabbione: it didn't want to auto-update; I have no intention of waiting for it
[01:26] <fabbione> ah ok
[01:26] <seb128> koke: should be "System", you have probably and old translation somewhere
[01:26] <jdub> Kamion: you've got gnome-panel 2.10.0-0ubuntu2 ?
[01:26] <fabbione> works for me
[01:27] <seb128> jdub: upload what ?
[01:27] <fabbione> jdub: there is gnome-panel ubuntu4 from seb....
[01:28] <Kamion> jdub: gnome-panel | 2.10.0-0ubuntu3 |         hoary | amd64, i386, ia64, powerpc, source
[01:28] <jdub> oh man
[01:28] <Kamion> I guess ubuntu4 will make it next cron.daily
[01:28] <seb128> .daily ?
[01:29] <Kamion> oh, actually, next but one cron.daily since it has to build
[01:30] <elmo> jdub: try again - you'd managhed to upload just the .dsc
[01:31] <Kamion> pitti: what version of language-pack-*-update am I watching out for?
[01:31] <jdub> seb128: ubuntu-artwork
[01:31] <jdub> my net's fucked
[01:31] <jdub> sec
[01:31] <jdub> oh
[01:31] <jdub> right
[01:31] <jdub> great
[01:31] <jdub> n/m
[01:31] <jdub> am i back?
[01:31] <Kamion> yes
[01:31] <pitti> Kamion: no -update; language-pack-foo 20050309
[01:31] <Kamion> all at once
[01:31] <Kamion> pitti: oh, duh, yeah
[01:31] <Kamion> thanks
[01:31] <pitti> Kamion: the -base packages remain at 20050308
[01:31] <jdub> yo?
[01:32] <jdub> fuck
[01:32] <Kamion> jdub: PING
[01:32] <Kamion> jdub: PONG
[01:32] <Kamion> 12:31 < jdub> yo?
[01:35] <thom> Kamion: grub install on i386 just bombed with "The file /boot/grub/stage1 not read correctly", current daily built from jigdo
[01:35] <thom> could be jfs related (testing it for kicks), trying ext3 now
[01:36] <Kamion> I doubt jfs /boot will work, it's just a missing check
[01:36] <thom> ok
[01:38] <Kamion> you should be able to use ext3 /boot and jfs / though
[01:38] <thom> nod; i don'tmuch care, was just curious and figured i should report it
[01:40] <Kamion> yep, thanks
[01:46] <ogra> thom: philosophical english ?
[01:46] <Treenaks> thom: PH is Philippines
[01:47] <thom> Treenaks: i was kinda assuming that was the case, ut even so
[01:48] <Kamion> the installer already puts en_* locales into locale.gen; perhaps it could check to see whether relevant locales are already present, and not be too stressed about regenerating if they are
[01:48] <elmo> thom: see u-d and my previous whinings
[01:48] <thom> ah, you already whined? cool cool
[01:48] <elmo> (or more to the point, mdz's reply)
[01:51] <koke> seb128: it's ok, my System menu appears now as system after cleaning /usr/local :)
[01:52] <jdub> back on air?
[01:52] <jdub> aha
[01:52] <jdub> i am
[01:53] <tseng> you're live, jdub 
[01:53] <seb128> koke: nice
[01:54] <seb128> jdub: is there somebody working on the translations updates for the desktop/hoary ?
[01:54] <jdub> seb128: rosetta people?
[01:54] <seb128> jdub: ie: is that worth than I make a list of the hoary specific strings all over the desktop and mail the lists ?
[01:55] <Kamion> jdub: oh, if you're changing Human, what happens to the old Human theme?
[01:55] <seb128> jdub: dunno but nobody has sent any string atm, the current system obviously doesn't work. I guess that rosetta people are busy on the tool and doesn't bother about what should be put in and pointed to people :)
[01:56] <jdub> Kamion: gone.
[01:56] <jdub> seb128: yeah
[01:56] <Kamion> ok
[01:57] <jdub> Kamion: it's not a hugely major change, and you can always click back to industrial
[01:57] <carlos> seb128: dude, we are busy because we are finishing the Ubuntu's l10n infrastructure
[01:57] <seb128> jdub: same about the hoary xml files pointed by the about ubuntu  ... is there anything running for that ? If not I'll tackle that this afternoon
[01:57] <carlos> seb128: working full time on it
[01:58] <seb128> carlos: I don't blame anybody, I just note than we don't have any localization for the ubuntu changes atm
[01:58] <Kamion> hm, the blue-on-brown thing with clearlooks is a bit weird
[01:58] <jdub> seb128: i don't think the doc i18n stuff is sorted with rosetta yet
[01:58] <jdub> Kamion: blue on brown?
[01:58] <carlos> seb128: I know, and you can blame us, we are out of schedule
[01:58] <Kamion> blue window title bars, brown background
[01:58] <jdub> erm
[01:58] <Kamion> desktop background
[01:58] <jdub> Kamion: choose Human
[01:58] <carlos> jdub: right, no documentation translations with Rosetta for Hoary
[01:58] <Kamion> oh, it'll munge the colours? ok.
[01:58] <seb128> carlos: don't worry, I'll push on that, I just don't want to dup work with you guys
[01:59] <jdub> Kamion: yeah, gtk+ theme defines the colours
[01:59] <seb128> jdub: we can use the same system as GNOME 2.10 /xml-po stuff ?
[01:59] <jdub> Kamion: phheeew, you had me worried for a sec
[01:59] <jdub> seb128: yes
[01:59] <jdub> seb128: well, probably
[01:59] <carlos> seb128: if you put the translations inside the source package, it will be added automatically to Rosetta, so you will not duplicate any work
[01:59] <Kamion> jdub: well it hasn't been uploaded yet, so I just have to guess :P
[01:59] <jdub> it has
[01:59] <Kamion> jdub: 12:30 < elmo> jdub: try again - you'd managhed to upload just the .dsc
[01:59] <jdub> it's already up
[01:59] <seb128> jdub: k, I'll do that today if you don't mind (ie: no other stuff for me to do first)
[01:59] <jdub> i did again
[01:59] <Kamion> no it isn't
[01:59] <Kamion> hoary-changes does not have mail
[02:00] <jdub> argh
[02:00] <jdub> -rw-r--r--  1 jdub jdub  457 2005-03-09 23:36 ubuntu-artwork_0.2.19-1_source.upload
[02:00] <seb128> jdub: I'm sure than some locale teams can translate that for hoary if we provide the po files
[02:00] <carlos> seb128: too late to move into xml-po for Hoary
[02:00] <jdub> i'll do again
[02:00] <seb128> carlos: ?
[02:00] <carlos> seb128: you asked: " we can use the same system as GNOME 2.10 /xml-po stuff ?"
[02:00] <seb128> carlos: not speaking about you, I'm working that around 
[02:01] <seb128> carlos: yeah, GNOME 2.10 notes are translated using that, I want to copy the system 
[02:01] <sabdfl> hi all
[02:01] <carlos> seb128: if you handle it by hand to create the .pot file, we are able to use Rosetta, yes
[02:01] <Kamion> afternoon Mark
[02:01] <seb128> hey sabdfl 
[02:01] <sabdfl> Kamion: thanks for the marathon session last night
[02:01] <jordi> Hmm, it's nearly time to pack up. I don't think I'll be back at the appartment.
[02:01] <seb128> carlos: that's the idea
[02:01] <sabdfl> seb128, Kamion, jdub: can you guys give me a quick status please
[02:01] <Kamion> sabdfl: just a shame I didn't quite make it up again before mdz went to sleep :)
[02:02] <pitti> Hi sabdfl 
[02:02] <sabdfl> Kamion: i had to force mdz to rest
[02:02] <ogra> hi sabdfl 
[02:02] <sabdfl> hey ogra
[02:02] <Kamion> sabdfl: as far as I know we're now just waiting for ubuntu-artwork, then I'll kick install CDs, live cloops, and live CDs in turn
[02:02] <sabdfl> ok
[02:02] <Kamion> since we'll likely be ready before mdz gets up again, who's doing the announcement?
[02:02] <sabdfl> we'll wait for mdz
[02:02] <Kamion> ok
[02:03] <sabdfl> jdub: hard deadline of 3pm UTC for the artwork uploads
[02:03] <sabdfl> Kamion: please start the rebuild no later than 3pm
[02:03] <sabdfl> do we have any mirrors we can trigger?
[02:03] <Kamion> sabdfl: jdub is trying to get it uploaded right now, he just ran into upload glitches
[02:03] <sabdfl> Kamion: in case there are ongoing issues, i want to have a hard cutoff
[02:03] <sabdfl> agreed with mdz
[02:04] <Kamion> fine
[02:04] <sabdfl> Kamion: how long does the build process take, usually?
[02:04] <jdub> gnome isn't even out yet ;)
[02:05] <sabdfl> is it going to change from what we have?
[02:05] <Kamion> sabdfl: about an hour to build
[02:05] <Kamion> everything
[02:05] <jordi> jdub: are all the tarballs released though?
[02:05] <Kamion> then time for people to download and test
[02:05] <jdub> sabdfl: it's already done
[02:05] <jdub> but i'm having network issues, evidently
[02:06] <jdub> jordi: of course
[02:06] <jdub> jordi: due date is monday
[02:06] <Kamion> can you scp to somewhere more reliable and upload from there?
[02:07] <jdub> attempting to upload 4MB with this packet loss is ridiculous
[02:07] <Kamion> sabdfl: two hours from beginning of build to being able to say go is a comfortable timeframe; we can pull it lower than that but I'd rather not
[02:07] <sabdfl> aiming for 19h00 UTC
[02:07] <sabdfl> mdz will be up 18h00 UTC
[02:07] <sabdfl> would like to have it rolled, mirrored as widely as possible, and tested by then
[02:07] <jdub> yeah, trying that
[02:07] <sabdfl> do we have any mirrors we can trigger
[02:07] <sabdfl> ?
[02:07] <Kamion> elmo: can we give mirrors some notice?
[02:08] <sabdfl> jdub: will the gnome code change at all before release? or are we good to go
[02:08] <jdub> sabdfl: we're fine
[02:09] <sabdfl> eta on the gnome release?
[02:10] <sabdfl> jdub: ^
[02:10] <sabdfl> i have to step out for half an hour
[02:10] <sabdfl> jdub, kamion, i think it would be poor form to release before gnome
[02:10] <ogra> hmm, gnome.org is still down....
[02:11] <sabdfl> can you guys discuss and come up with a plan if gnome.org is still down at 18h00 UTC
[02:11] <sabdfl> brb
[02:11] <mvo> hi froud 
[02:11] <HiddenWolf> gnome.org down? whoops
[02:12] <jdub> why does everyone keep asking that? :)
[02:12] <jdub> gnome tarballs were due two days ago
[02:12] <jdub> we don't need any updates at all
[02:14] <Kamion> elmo: can we kill boot-floppies from universe? :-) There seems little point in trying to build it
[02:15] <jdub|piractical> ahar!
[02:15] <ogra> heh
[02:15] <jdub|piractical> uploading via alternate route :-)
[02:15] <Kamion> "piratical" :-)
[02:15] <jdub|piractical> it's up!
[02:15] <thom> jdub|piractical: next doors dsl?
[02:15] <thom> :P
[02:15] <jdub|piractical> fie on my stinkin' upstream bandwidth, me hearties!
[02:16] <jdub|piractical> i be shippin' with me neighbour!
[02:16] <jdub|piractical> thom: ahaye
[02:16] <Kamion> elmo: could you hurry this one through?
[02:16] <jdub> and sshing back home, where the connection seems fine :|
[02:17] <thom> ok, crrent dailies are fine on amd64/i386/powerpc for me; only minor issue is the scrollkeeper spew which i know colin has already mentioned
[02:18] <Kamion> yeah, current daily still has old gnome-applets
[02:19] <fabbione> Kamion: do you have the new gnome-applets?
[02:19] <fabbione> perhaps i can test the DVD, if you feel like updating it
[02:20] <Kamion> I'm not going to update the DVD just yet, 'cos it'll take ages and block the CD build
[02:20] <fabbione> ah ok
[02:20] <fabbione> i didn't know it was a blocker
[02:20] <Kamion> I'll do it after the CD builds are all done, though
[02:20] <Kamion> well, it's just insufficient locking/separation on cdimage
[02:20] <Kamion> I think I can do Ubuntu and Kubuntu builds in parallel
[02:21] <Kamion> but I cannot do multiple Ubuntu builds at once, or multiple Kubuntu builds at once
[02:21] <Kamion> gnome-applets | 2.10.0-0ubuntu2 |         hoary | amd64, i386, ia64, powerpc, source
[02:30] <lamont> Kamion: which version of postfix is on the install CD these days, I wonder
[02:30] <Kamion> lamont: 2.1.5-9ubuntu1
[02:31] <lamont> GRRR
[02:33] <sabdfl> back
[02:33] <Kamion> sabdfl: we have ubuntu-artwork source now, waiting for buildds
[02:38] <dholbach> {ftp,mail}.gnome.org are online
[02:38] <ogra> dholbach: havent been of afauk
[02:38] <ogra> s/u/i
[02:38] <ogra> off even
[02:38] <dholbach> ah ok
[02:38] <dholbach> i see
[02:39] <HiddenWolf> why is gnome.org off?
[02:40] <jdub> no answer yet.
[02:42] <ogra> probably the webmaster slipped on a banana peel.....
[02:42] <ogra> ...there were a lot around these days
[02:42] <sabdfl> Kamion: thanks
[02:42] <sabdfl> while  we're waiting, Kamion, could you take a turn backing on:
[02:43] <sabdfl> http://www.ubuntulinux.org/wiki/DraftHoaryPreviewAnnouncement
[02:43] <Kamion> I'm keeping little warmed up building kubuntu CDs
[02:43] <sabdfl> let's add in as much technical stuff as possible to the list
[02:43] <Hannes_> Gnome.org is changing looks for 2.10 release :P
[02:43] <Hannes_> or not
[02:44] <dholbach> ubuntu-artwork built on i386
[02:44] <jdub> Hannes_: that's mourning for europe.
[02:44] <Treenaks> has the final name for hoary+1 been chosen yet?
[02:44] <sabdfl> Treenaks: yes
[02:44] <thom> Treenaks: cold turkey
[02:44] <sabdfl> :-)
[02:45] <jordi> woa, if it's a secret, I was veery close to talk too much :)
[02:46] <lamont> dholbach: and since it's arch: all, that's the only one that matters.
[02:46] <Treenaks> thom: :P
[02:46] <dholbach> lamont: i imagined that, but didnt have a closer look :-)
[02:47] <sabdfl> jdub: when's the best time to announce the name for hoary+1?
[02:47] <jdub> sabdfl: preview release announcement would be awesome
[02:48] <jdub> sabdfl: certainly better there than in the final release announcement
[02:48] <sivang> jdub: shouldn't we mention a word or two about gnome-app-install on the technical list?  (A pointy clicky way to install/remove new applications)
[02:48] <jdub> sabdfl: or do you want to do a separate one?
[02:48] <jdub> sivang: hrm, last i read we had a general install/update tools thing
[02:48] <sabdfl> jdub: yes, i'm happy with preview release, would also go in final release
[02:49] <seb128_> jdub: do we gave updated desktop files ?
[02:49] <sabdfl> Kamion: done with the announcement?
[02:49] <jdub> sabdfl: seems a bit "don't look now but..." for final
[02:49] <jdub> seb128_: not for preview
[02:49] <jdub> sivang: yeah, what's in there is fine
[02:49] <seb128_> k, so no need to point g-i-a
[02:50] <sivang> jdub: ok cool
[02:50] <Kamion> sabdfl: just made a couple more tweaks
[02:50] <Kamion> sabdfl: done for now
[02:50] <sabdfl> ok
[02:50] <sabdfl> seb128: could you take a turn on it?
[02:51] <sabdfl> http://www.ubuntulinux.org/wiki/DraftHoaryPreviewAnnouncement
[02:51] <sabdfl> i'm playing "plone locking system" today
[02:51] <sabdfl> call me the BPL
[02:51] <Kamion> BPL?
[02:51] <Kamion> oh :-)
[02:51] <jdub> sabdfl: hrm, would be good to switch from 'their' to 'our'
[02:52] <ogra> busy plone locker ?
[02:52] <jdub> sabdfl: seems more personal
[02:52] <pitti> ogra: s/busy/big/ ?
[02:52] <Kamion> ogra: analogy with BKL
[02:52] <ogra> K ?
[02:52] <sabdfl> that's a trouble entendre if i ever saw one
[02:53] <thom> big kernel lock
[02:53] <ogra> ahh
[02:53] <zul> gday
[02:54] <sivang> when are we expecting to release? I want to know when to spread over the local portals etc.
[02:54] <Kamion> sivang: currently aiming for 1900 UTC
[02:54] <Treenaks> sivang: not to mention release parties :P
[02:54] <sivang> Treenaks: right :)
[02:55] <sivang> btw, how would you explain "shrinkwrapped" cds?
[02:55] <koke> where is the source for release-notes??
[02:55] <koke> are they going to be translated?
[02:57] <sivang> "..includes all of Debian as well as most of the packages of apt-get.org.." => is this referring to universe?
[02:58] <Treenaks> sivang: yes
[02:58] <sivang> Treenaks: did the motus take many packages from there? :)
[02:58] <Treenaks> sivang: unknown. :)
[02:59] <sivang> Treenaks: this mention of apt-get.org was a surprising bit in the announcment draft
[02:59] <abelli> what package is ssh.h in?
[02:59] <ogra> sivang: nope
[02:59] <sivang> abelli: apt-file search ssh.h
[02:59] <Burgundavia> To be more specific with this line - "To sign up for future announcements of this type" what about changing of this type to "about Ubuntu releases" or something similar
[02:59] <sivang> abelli: but you have to install apt-file first.
[02:59] <Hannes_> abelli: openssl-dev?
[03:00] <Kamion> ssh != ssl
[03:00] <Hannes_> abelli: openssh-dev?
[03:00] <Kamion> no such package
[03:00] <abelli> dont have it
[03:00] <abelli> Kamion, hello my friend.. -:)
[03:00] <Kamion> openssh does not ship an ssh.h; there's an ssh.h buried in kdelibs4-dev
[03:00] <Kamion> usr/include/kde/kdesu/ssh.h                                 libdevel/kdelibs4-dev
[03:00] <sivang> abelli: kdelibs4-dev: usr/include/kde/kdesu/ssh.h
[03:00] <sivang> Kamion: ;-)
[03:01] <Kamion> but that's all
[03:01] <abelli> ok so no Pure package avaible..
[03:01] <abelli> thank you all
[03:01] <Kamion> "Pure package"?
[03:01] <Treenaks> Kamion: "libssh-dev"
[03:01] <abelli> not regarding KDE...
[03:01] <Kamion> Treenaks: huh?
[03:01] <Treenaks> Kamion: that'd be a clean package with ssh.h in it.. libssh-dev
[03:02] <Kamion> Treenaks: but there is no openssh -dev package at all
[03:02] <Treenaks> Kamion: but I think that's what he's looking for?
[03:02] <Kamion> AFAIK no SSH implementation exposes its internals like that
[03:03] <Kamion> all ssh.h would give you, hypothetically, would be SSH protocol constants
[03:03] <Kamion> if you're competent enough to do anything with those, you're also competent enough to write the header file yourself :)
[03:03] <abelli> that's why i am asking it here: i'm not competent at all.
[03:03] <Treenaks> abelli: may we quote you on that one?
[03:03] <abelli> :) but i think that i'll just change the configure
[03:03] <Kamion> abelli: perhaps it would help if you said what you're trying to do
[03:04] <abelli> Kamion, yeah right, i'm sorry, i'm trying to compile paranoy, which is
[03:04] <abelli> an email client with crypto mailboxes and config files..
[03:05] <abelli> really nothing important, but before trying to package it i need to build it.
[03:05] <abelli> Kamion, thank you
[03:07] <koke> mako: is there a release-notes source .xml anywhere on the net?
[03:07] <dholbach> www.gnome.org is back
[03:07] <Kamion> sivang: needs rather more technical detail
[03:07] <Kamion> I feel
[03:07] <ogra> dholbach: yay
[03:07] <dholbach> they only have some "banana republic" warning on their front page
[03:07] <Kamion> many people would say we are not multimedia-ready, because we don't support MP3 or whatever out of the box
[03:08] <Kamion> so maybe more backup for the buzzwords :-)
[03:08] <sivang> Kamion: I thought of being breif and summing .
[03:08] <Kamion> sivang: see the top of the page
[03:08] <Kamion> MarkShuttleworth -- I think this preview announcement should be aimed at the technical audience and include a much longer list of technical enhancements since Hoary. Go into the bling and the plumbing.
[03:09] <Kamion> abelli: ssh.h is a typo in configure, it's really looking for openssl/ssl.h
[03:09] <sivang> Kamion: ok, I'll break those up to a list. :)
[03:09] <Kamion> abelli: so install libssl-dev
[03:09] <Kamion> sivang: and really I'd be inclined to leave out the multimedia bit
[03:09] <abelli> Kamion, thank you
[03:09] <abelli> ;)
[03:09] <sivang> Kamion: sure, whatever you say , Sir :)
[03:09] <Kamion> it's not an order, just a suggestion :-)
[03:09] <fabbione> "Dear Kamion the allmighty, i have some issues with ...."
[03:10] <fabbione> this how we should address Kamion ;)
[03:10] <sivang> fabbione: hehe
[03:10] <jdub> or "duderino" for short :-)
[03:10] <sivang> fabbione: I tell him this in /msg , all the time :)
[03:11] <fabbione> Kamion: kernel-wedge || true doesn't exist anymore..
[03:11] <fabbione> Kamion: tested and verified
[03:12] <Kamion> fabbione: hooray
[03:12] <ogra> added even
[03:13] <Hannes_> 1601.10 < abelli> ok so no Pure package avaible..
[03:13] <Hannes_> sorry
[03:13] <abelli> Hannes_, huh?
[03:14] <ogra> Kamion, thanks :)
[03:15] <pitti> sabdfl: would you mind if I add a small note about derooting to DraftHoaryPreviewAnnouncement?
[03:15] <Kamion> damnit, after all that waiting I got distracted
[03:15] <Kamion> probably-preview CD images building now
[03:16] <Kamion> and live cloops
[03:16] <jdub> gnome servers are up
[03:16] <jdub> should be out soon
[03:16] <sabdfl> pitti: i would *love* you to add a note about derooting
[03:17] <pitti> sabdfl: I would *love* to do it :-)
[03:17] <abelli> pitti, did you build the new -hardened?
[03:17] <pitti> abelli: sorry, not yet
[03:17] <pitti> abelli: I have to convert to 2.6.11 and review all Ubuntu patches for this
[03:17] <abelli> pitti, YOU don't HAVE to be sorry
[03:17] <pitti> abelli: so this is non-trivial
[03:18] <abelli> pitti, right.
[03:18] <tseng> abelli: when 2.6.11 is in, grsecurity should go on clean besides 2 patches
[03:18] <tseng> spender merged the bits fron 2.6.11.1, ppc64 and input fixes
[03:18] <tseng> im imagining ubuntu will merge them as well
[03:18] <abelli> yeah what about the rootly hole they found?
[03:18] <tseng> its fixed.
[03:19] <abelli> no suicides? beautiful.
[03:19] <tseng> if pitti doesnt find the time I'll try it once 2.6.11 is in universe
[03:20] <tseng> it should be much easier than .10, no fighting with the elf security fixes
[03:20] <jordi> jdub: it's been great to hear you over the phone :P
[03:20] <jordi> I'm leaving London now, ttyl
[03:20] <sabdfl> pitti: tell me when i can edit the preview announcement again please
[03:20] <jdub> jordi: ;-)
[03:20] <pitti> sabdfl: just finished
[03:21] <Kamion> heh, language pack installation is amusingly O(n^2) on the live CD cloop builds
[03:21] <pitti> tseng: once our kernel team finishes 2.6.11 final, it's a piece of cake to build the -hardened kernels
[03:21] <tseng> yeah it should be very easy
[03:22] <Kamion> maybe locale-gen should grow an --only-new option or something
[03:22] <pitti> tseng: the hard thing is to build a 2.6.11 kernel in the first place
[03:22] <lamont> Kamion: heh
[03:22] <lamont> livecd would love you
[03:22] <tseng> pitti: thats why we have fabbione :)
[03:22] <Treenaks> tseng: and his Crack Team of Kernel Commandos
[03:22] <zul> tseng: and the rest of us is chopped liver? :)
[03:23] <fabbione> 2.6.11 is very very low priority for hoary guys
[03:23] <tseng> zul: yes.
[03:23] <zul> tseng: oh ok
[03:23] <sabdfl> elmo: "as well as most of the packages of apt-get.org"... are my pants on fire?
[03:23] <jdub> jdubtv! -> http://node.waugh.id.au/
[03:23] <tseng> :8800
[03:23] <sivang> sabdfl: yeah, we wondered about it
[03:23] <sivang> sabdfl: :)
[03:24] <jdub> tseng: ugh, thanks
[03:24] <jdub> jdubtv! -> http://node.waugh.id.au:8800/
[03:24] <sabdfl> well, i did ask for it to be made so
[03:24] <sabdfl> elmo: ?
[03:24] <pitti> tseng: what do you use for displaying the stream?
[03:24] <tseng> pitti: totem
[03:25] <tseng> just append the address to the command line
[03:25] <tseng> or Open Location
[03:25] <Kamion> firefox wants to use rhythmbox by default
[03:25] <fabbione> jdub: omg.. i couldn't remember you were so.. hairy
[03:25] <jdub> :-)
[03:25] <sivang> jdub: hey jdub :)
[03:25] <jdub> not like a penguin
[03:26] <fabbione> i am gonna buy a webcam tomorrow
[03:26] <carlos> jdub: do you have screenshots of the new Ubuntu look?
[03:26] <fabbione> any good suggestion?
[03:26] <theine> Are there any plans for updating the ipw2100 driver in the Ubuntu Kernel to the newest version (1.0.5)?
[03:26] <tseng> not very good here jdub 
[03:26] <fabbione> theine: not for hoary
[03:26] <zul> yeah not quickcam
[03:26] <jdub> carlos: one sec
[03:26] <theine> ok
[03:26] <tseng> on the sync
[03:27] <sivang> jdub: bah, I got the voice after you talk :)
[03:27] <sivang> jdub: what's the other display for ? :) (I see you looking over it all the time)
[03:27] <sabdfl> theine: what version do we currently have?
[03:27] <theine> savdfl: 1.0.2
[03:27] <jdub> sivang: big screen, close webcam
[03:29] <sivang> s/pats/pets/
[03:29] <pitti> sivang: isn't "pat" corrent?
[03:29] <Treenaks> pitti: both are :)
[03:29] <Kamion> either would work with somewhat different meanings, "pat" would be more usual I think
[03:29] <sivang> pitti: hm, not sure :)
[03:30] <jdub> pat is correct in uk/au
[03:30] <jdub> pet is correct in USA
[03:30] <pitti> isn't "correct" correct...
[03:30] <sabdfl> fabbione: could you look into theine's version issue for ipw2100 please?
[03:30] <fabbione> sabdfl: sure
[03:31] <maswan> jdub: Done at Wed Mar  9 15:30:15 MET 2005
[03:31] <sabdfl> also ipw2200
[03:31] <jdub> maswan: rocking :-)
[03:31] <sabdfl> thom: are you near elmo?
[03:31] <fabbione> theine: what problems do you have with 1.0.2?
[03:31] <jdub> maswan: has luis pinged you about the livecd?
[03:31] <maswan> jdub: I haven't heard anything. Currently, ftp.gnome.org will be happy to handle the load tohugh. :)
[03:32] <jdub> maswan: http://torrent.linux.duke.edu/gnome-livecd-2.10.torrent
[03:32] <theine> fabbione: not too severe ones actually, it's just that network-admin frequently hangs and the signal strength is not displayed correctly
[03:32] <sivang> jdub: we want to see some quality programming! I'd vote for a lugradio show over your place :)
[03:32] <maswan> jdub: ah, want an http source for that too on ftp.gnome.org?
[03:32] <theine> fabbione: Of course I'm not quite sure if that's an driver issue or not
[03:32] <jdub> maswan: if you think it can handle it
[03:33] <sivang> jdub: *with* live video :)
[03:33] <jdub> sivang: live gnome release not good enough for you? :)
[03:33] <sivang> jdub: heheh, sure it does, the best :-)
[03:33] <fabbione> theine: you can try building the 1.0.5 externally and see if it helps.. changing upstream version so close to release is quite dangerous because it might introduce more bugs than it fixes
[03:33] <maswan> jdub: Will take a while to get it from the torrent though
[03:33] <Treenaks> jdub: we want to have your terminal output scrolling as an overlay over jdub-tv!
[03:33] <Treenaks> jdub: WHILE you release gnome
[03:34] <theine> fabbione: I had some problems building 1.0.5 against the ubuntu kernel headers but I'll try again
[03:34] <sivang> Treenaks: wow that would be COOOOOL
[03:34] <maswan> jdub: current estimate is 2 hours.
[03:34] <fabbione> theine: thanks
[03:34] <Treenaks> sivang: exactly
[03:34] <jdub> maswan: ok
[03:34] <jdub> maswan: ping me when you have it, i'll add it to the mirror list
[03:34] <maswan> jdub: thanks, I will
[03:36] <maswan> jdub: Hm. got a bit more speed now, so it might get here sooner. we'll see. :)
[03:37] <koke> jdub: do you need more mirrors??
[03:37] <koke> downloading at http://pulsar.unizar.es/~koke/mirror/gnome-livecd-2.10/
[03:37] <maswan> jdub: hmm.. will this be a one-off thing or will it be something more permanent that we could mirror?
[03:38] <koke> jdub: forget that, not enough free space :(
[03:39] <sivang> jdub: what was that? beer? :)
[03:40] <jdub> carlos: http://people.ubuntu.com/~jdub/screenshots/ubuntu-hoary-preview.jpg
[03:40] <thom> sabdfl: nowhere close
[03:40] <jdub> maswan: possibly more permanent
[03:40] <carlos> jdub: thanks
[03:41] <jdub> maswan: but it will most likely be part of the ubuntu derivatives set later on
[03:41] <sivang> jdub: nice tune, and cool artwork!
[03:41] <kent> jdub, still defaults to the gnome icons for folders etc?
[03:41] <Kamion> jdub: the inner one seems to be the old theme
[03:41] <maswan> jdub: Ok, I'll just throw this into the temp/ folder then instead of creating a proper gnome-livecd folder etc.
[03:41] <jdub> Kamion: hrm, nup
[03:41] <jdub> kent: yes
[03:42] <maswan> jdub: "temp" meaning that they do not have a proper updating mechanism etc
[03:42] <jdub> yeah
[03:44] <fabbione> b/wind goto 1
[03:44] <fabbione> ops
[03:47] <jdub> ah bum
[03:47] <jdub> missed one file in ubuntu-artwork
[03:47] <jdub> :)
[03:47] <Kamion> ok, I have to go help out ill fiancee
[03:47] <Kamion> jdub: no not say stuff like that
[03:47] <fabbione> jdub: you kidding?
[03:48] <Kamion> s/no/do/
[03:48] <jdub> Kamion: not critical :-)
[03:48] <Kamion> what did you forget?
[03:48] <jdub> cursor theme definition
[03:48] <jdub> can do update straight after preview though
[03:48] <jdub> no biggie
[03:48] <sivang> heheh
[03:48] <Kamion> ok
[03:48] <Kamion> install CD's building
[03:49] <Kamion> please download and test when it's done
[03:49] <Kamion> ETA 20 minutes or so
[03:49] <fabbione> Kamion: roger
[03:50] <HiddenWolf> jdub: is the old artwork still available? this looks rough
[03:51] <jdub> switch to the industrial theme
[03:51] <koke> HiddenWolf: http://archive.ubuntu.com/ubuntu/pool/main/u/ubuntu-artwork/
[03:51] <koke> HiddenWolf: three versions there
[03:53] <sivang> jdub: yeah, what was it?
[03:53] <jdub> madonna
[03:53] <jdub> never mind
[03:53] <jdub> (yeeesh)
[03:53] <jdub> you can tell i'm concentrating or something
[03:53] <HiddenWolf> jdub: I can't find the industrial theme in the selector
[03:53] <mpt_londo1> HiddenWolf: That's not a bug, it's a feature!
[03:54] <mpt_londo1> wah
[03:56] <HiddenWolf> Is this the final theme for ubuntu hoary?
[03:57] <koke> well, have to eat smth. bye, and happy hacking :)
[03:58] <sivang> jdub: eh, what a classic
[04:02] <jbailey> I discovered last night when trying to fix my wife's machine to boot without Internet access that single user mode is pretty useless when root doesn't have a password. =)
[04:02] <pitti> jbailey: init=/bin/bash helps a lot :-)
[04:03] <jbailey> pitti: Yeah, I'm thinking from the perspective of possibly having to support people on these systems.  I was wondering if the script should be hacked to just drop to a rootshell without prompting for a password.
[04:03] <jbailey> If you've got that kind of physical access to the machine all bets are off for security anyway.
[04:04] <jdub> that's what it did for warty
[04:04] <maswan> jdub: http://ftp.acc.umu.se/mirror/temp/gnome-livecd-2.10/
[04:04] <thom> jbailey: it did for warty; has that been dropped?
[04:04] <jbailey> Her machine is running pretty-current Hoary, but it was an upgrade from Debian.  I'll check to make sure the changed init script isn't sitting as a .dpkg-new file.
[04:05] <jdub> maswan: ta!
[04:05] <thom> jbailey: it's not a changed init script; it's a fix to sulogin
[04:05] <schweeb_> jbailey: I'm pretty sure that single user still asks for password in hoary
[04:06] <pitti> jbailey: probably a good idea, yes
[04:06] <schweeb_> I could check, if you like
[04:06] <jbailey> thom: Ah?  Hmm, I see that in my script here.  I had always assumed it just fired up bash =)
[04:07] <thom> jbailey: nup; does passwd checking then starts the shell
[04:09] <lamont> jbailey: that's what init=/bin/sh is for
[04:09] <jbailey> lamont: I'll make sure I refer the first support call where we don't have a password to the system and some 80 year old doing a first time linux install to you then. ;)
[04:09] <thom> jbailey: works for me
[04:09] <theine> fabbione: I have the same issues with ipw2100 1.0.5
[04:09] <lamont> jbailey: heh/
[04:10] <lamont> yeah, phone support with that kinda sucks
[04:10] <fabbione> theine: such as?
[04:10] <thom> jbailey: what is root's password in shadow set to?
[04:10] <theine> fabbione: such as network-admin hanging and wrongly displayed signal strength
[04:10] <thom> jbailey: ( i just rebooted to single user on a clean install and wasn't prompted for a password)
[04:11] <fabbione> theine: so it doesn't solve the problem....
[04:11] <jbailey> thom: I'm not near the machine atm, I have no net connection at home.  I think I may have set it to * though, judging by what I have on this laptop.
[04:11] <theine> fabbione: no
[04:11] <jbailey> (which is also an upgrade for me)
[04:11] <fabbione> theine: ok. thanks for testing tho
[04:11] <theine> fabbione: np
[04:11] <thom> jbailey: * and ! both work fine, *LOCK* will cause problems
[04:11] <jbailey> thom: Thanks.  now that I know that it's suppoed to work, I'll do some troubleshooting. =)
[04:12] <fabbione> theine: i think we will stay with .2 for hoary. at least it is known to work as is
[04:12] <Kamion> lamont: can you check if all the cloops are done?
[04:12] <theine> fabbione: sure, that's probably a good idea
[04:12] <lamont> Kamion: checking
[04:12] <jbailey> Mithrandir: Did doko talk to you about an ia64->i486 cross compiler?  He tried to get the three of us together last night, but it didn't work out.
[04:13] <fabbione> yup
[04:13] <lamont> Kamion: no
[04:13] <Mithrandir> jbailey: yes, and I'll just do it another way which is way less pain.
[04:13] <Mithrandir> jbailey: I'm telling myself that ia32-libs is temporary and multiarch is the right solution, so.. :)
[04:14] <lamont> Kamion: you didn't happen to try to fire them off in parallel (on the same machine) did you?
[04:14] <Kamion> lamont: I did all four in parallel, yeah
[04:14] <jbailey> Mithrandir: 'kay.  Is the goal to eventually build cross compilers for the various target arch's?  With multilibs it should be dead easy, and gcc already likes having the TRIPLE-program setup.
[04:15] <Mithrandir> jbailey: no, you need more than just a compiler.  I think multiarch runtime is a good start -- most apps won't need multiarch build-time as well
[04:15] <lamont> Failed to fetch http://archive.ubuntu.com/ubuntu/pool/main/k/kdepim/ktnef_3.3.2-1ubuntu3_amd64.deb  Bad header line
[04:15] <lamont> ew
[04:15] <lamont> all 4 architectures in parallel is fine
[04:16] <lamont> but BuildLiveCD kubuntu & BuildLiveCD ubuntu & will verifiy the locking code...  they build in the same chroot
[04:16] <lamont> so one will wait/fail
[04:16] <lamont> Kamion: I'll get both builds done on all 3 (that'll succeed) and let you know
[04:18] <jbailey> Mithrandir: True.  Years of Hurd hacking have left me with a strong desire to make getting a packaged cross-compiler alot easier.
[04:19] <jbailey> Mithrandir: The idea where you just install the libs and pass a parameter to debuild for gcc is kinda fun. =)
[04:21] <ronalde> saving a translation in rosetta fails with "A system error occurred." any clues?
[04:21] <maswan> jdub: oh, oops. http://ftp.gnome.org/mirror/temp/gnome-livecd-2.10/
[04:21] <lamont> Kamion: i386 is in pass 2/2 in partimage, then will need to compress
[04:21] <maswan> jdub: same hosts, but still, that's more on-topic. :)
[04:22] <jdub> maswan: ;)
[04:22] <sabdfl> maswan: what's that?
[04:23] <HiddenWolf> jdub: is this artwork the final version for hoary?
[04:23] <maswan> sabdfl: oh, it's the gnome livecd for the coming gnome release, perhaps we should be off in a more gnome-related channel for this. :)
[04:24] <maswan> sabdfl: ftp.acc.umu.se, which I admin, is also known as ftp.gnome.org
[04:24] <jdub> luis's ubuntu-based gnome branded livecd
[04:24] <sabdfl> cool!
[04:25] <jdub> HiddenWolf: not entirely
[04:26] <HiddenWolf> jdub: i've got some usability issues with it, can I file bugs?
[04:26] <Mithrandir> jbailey: sure, and it should work.  You need support in the whole toolchain for it to work, though
[04:26] <sivang> btw, does anybody know if anything advanced in the direction of having d-i translatable through rosetta?
[04:26] <pitti> yay, new daily CDs
[04:26] <ronalde> saving a translation in rosetta fails with "A system error occurred." any clues?
[04:27] <jdub> of course
[04:27] <lamont> Kamion: _ubuntu_ cloops are done everywhere
[04:27] <Simira> carlos: a lot of people are complaining on that error ronalde is talking about
[04:27] <pitti> Kamion: are the 20050309.1 CDs the final candidates?
[04:28] <jbailey> Mithrandir: Yes, but a cross binutils is generally trivial to produce.
[04:28] <lamont> Kamion: kubuntu: i386 compressing, amd64 installing kubuntu-desktop et al, ppc in debootstrap
[04:30] <Mithrandir> jbailey: you need dpkg-cross too and if you want to do funky stuff, pkg-config with cross support and so on.
[04:30] <carlos> Simira: I know, and I have a patch already, mailing list and Rosetta's bts knows that already
[04:30] <Simira> carlos: ok, I saw your answer on the thread, just seemed to still be a problem
[04:30] <ronalde> carlos: is there an intermediate procedure to get translated stuff uploaded?
[04:31] <sivang> carlos: any knowledge about when the installer will be translatable through rosetta?
[04:31] <jbailey> Mithrandir: Yeah.  I've worked with dpkg-cross a bunch, and in practice most of things I want a cross compiler for are bootstrapping toolchains and the bare essentials on another arch.    I can see it being a godsend for embedded targets, though.
[04:31] <Mithrandir> jbailey: absolutely.. would be nice for my WRT54GS, for instance.
[04:31] <carlos> Simira: it's just that there is a delay since I prepare a patch and it's applied to the server, that's all
[04:31] <jbailey> Mithrandir: Someone the other day was asking me for advice on porting Ubuntu to sh3 - He was running into problems building glibc in scratchbox.  Little things like that could be made so much easier.
[04:32] <Mithrandir> jbailey: I'm also quite excited about the stuff we can do with linux support in the BSD kernels, for Debian.
[04:32] <carlos> ronalde: not sure, it depends on pitti accepting translations directly to the language packs or sending it to the maintainer of the package you are translating, none of them easy to maintain later
[04:33] <mako> sabdfl: changes to the announcement look great
[04:33] <fabbione> Mar  9 15:27 hoary-install-i386.iso
[04:33] <Mithrandir> (port everything which really wants BSD, multiarch the rest)
[04:33] <fabbione> is this the last one?
[04:33] <pitti> carlos, ronalde: if I get po files with translation domains, I can put them directly into the language packs
[04:33] <carlos> sivang: no idea, when we finish the Rosetta changes we have been designed  recently, it should be soon, but I have said it too much times already...
[04:34] <ronalde> pitti: i'll send you the Dutch translation of gksu
[04:34] <carlos> pitti: but they are going to be lost when we start using Rosetta, unless you upload them later
[04:34] <carlos> pitti: please, remember that
[04:34] <pitti> carlos: is gksu already in Rosetta?
[04:35] <ronalde> pitti, carlos: yes
[04:35] <pitti> carlos: is there an easier way than the web interface to upload a po?
[04:35] <ronalde> pitti, carlos: https://launchpad.ubuntu.com/rosetta/products/gksu/gksu-1.2.2/
[04:35] <pitti> ronalde: please upload them there
[04:35] <pitti> ronalde: I can download them from Rosetta
[04:35] <carlos> pitti: current modules are not going to be reused for hoary, those are for upstream releases not Hoary ones
[04:35] <ronalde> pitti: i can' t thats the problem ...
[04:36] <carlos> pitti: we have a bug atm that prevents the use of Rosetta, it's pending to get the patch applied
[04:37] <pitti> ronalde: please upload them later then
[04:37] <pitti> ronalde: I cannot upload new packs today anyway
[04:37] <pitti> ronalde: preview freeze
[04:37] <ronalde> pitti: i see, i wished to get it included in hoary
[04:38] <pitti> ronalde: no problem for the final
[04:38] <carlos> ronalde: in theory, Rosetta will be on time to translate hoary
[04:38] <ronalde> carlos: great!
[04:39] <metallikop> Could someone please explain why pbuilder wants to use older versions of the -dev packages?
[04:39] <tseng> metallikop: is your pbuilder up to date?
[04:39] <metallikop> afaik
[04:39] <tseng> it has its own apt cache
[04:39] <metallikop> oh oh , that'd be my problem then, haven't had any issues with it thus far
[04:40] <tseng> sudo pbuilder update
[04:40] <tseng> and have another go :)
[04:40] <metallikop> hehehe, that'll fix it, i'm sure.
[04:40] <tseng> if you really need the newer depend.. you'll want to be sure to mark it as such in debian/control
[04:41] <metallikop> thanks, all fixed now
[04:41] <Kamion> lamont: oh, sorry if I confused you, I only actually care about Ubuntu right now
[04:41] <fabbione> lamont: what's the status with LiveCD?
[04:41] <Kamion> lamont: I just said BuildLiveCD, no 'ubuntu'
[04:41] <lamont> Kamion: ubuntu is all done
[04:41] <Kamion> yes, to those who asked, 20050309.1 is candidate
[04:41] <lamont> Kamion: no args == 'ubuntu kubuntu'
[04:42] <Kamion> ok, live CDs building
[04:42] <Kamion> lamont: oh, oops, will adapt
[04:42] <lamont> and kubuntu/ppc is generating locales
[04:42] <lamont> and if you try enough, then you run the ppc and amd64 machine out of disk space, and I have to slap you
[04:44] <fabbione> hell! this cdrom doesn't read CD-RW
[04:45] <lamont> fabbione: I hate it when I run into that...
[04:47] <Kamion> my bug reporters have many :P
[04:47] <Mithrandir> you all live in the stone age. :P
[04:48] <fabbione> ok.. i can't find the other CD-RW.. i will test netinstall from cd :(
[04:48] <mpt_london> fabbione: Is the error message nice and helpful?
[04:48] <fabbione> that's all i can do atm
[04:48] <fabbione> mpt_london: there is no error message.. when you boot from cd the bios says.. no bootable cd found
[04:49] <fabbione> because i can hear that is not reading it properly
[04:50] <mpt_london> Oh, and you can't file bugs against the bios
[04:50] <mpt_london> dang :-)
[04:51] <Kamion> live CD candidates up, 20050309.1
[04:51] <zul> you can they will be ignored though :)
[04:54] <tseng> mako: ping?
[04:56] <mako> tseng: yes
[04:56] <tseng> may I msg for just a moment?
[05:02] <fabbione> there 2309209302932 CD and only 2 RW in there that i can use
[05:02] <fabbione> of course i can't find them!
[05:03] <zul> heh the irony
[05:03] <fabbione> and the funny thing is that i have 590 DVD that i could use....
[05:07] <jbailey> daniels: ping?
[05:07] <thom> jbailey: asleep, i think
[05:07] <jbailey> Yeah, probably, but you can never tell with him ;)
[05:08] <thom> 14:03 < daniels> night dudes ;-)
[05:08] <thom> 2 hours ago
[05:08] <jbailey> Thanks.
[05:09] <fabbione> ahh i might have found some other media.. 
[05:09] <thom> jigdo is a work of art
[05:10] <fabbione> thom: my problem is not to rsyunc the cd
[05:10] <Mithrandir> for some value of art including "huge amounts of crack"?
[05:10] <fabbione> is to find proper media
[05:10] <thom> Mithrandir: possibly
[05:11] <thom> but since i have a local mirror rsync seems like a huge waste of time
[05:11] <jbailey> Mithrandir: Modern art.  The kind you find in government funded museums. =)
[05:11] <fabbione> thom: i have both mirror of the pool and the cd images
[05:11] <fabbione> thom: a daily sync takes no more than few minutes
[05:11] <thom> fabbione: i used to, but it seems pointless
[05:11] <Mithrandir> jbailey: guess so. :)
[05:11] <fabbione> i am pretty sure less than jidgo
[05:12] <Mithrandir> jigdo can use the old .iso
[05:12] <fabbione> can it?
[05:12] <Mithrandir> iirc, yes
[05:12] <fabbione> there must have been a lot of water flowing under the bridges since last time i tested it
[05:12] <thom> top
[05:13] <Mithrandir> fabbione: been there for ages, iirc
[05:13] <thom> uh, wrong keyboar
[05:13] <thom> d
[05:13] <fabbione> thom: time to install Xmdx ?
[05:13] <fabbione> meh Xdmx
[05:13] <thom> no, just time to stop reinstalling my desktop
[05:15] <thom> oh well, it just took jigdo 8 minutes to build 3 cds, so i'm not sure nightly rsyncs would win much if anything over that
[05:15] <fabbione>  /mirrors/ubuntu-cd/hoary-live-i386.iso
[05:15] <fabbione> sent 162829 bytes  received 22426870 bytes  79401.40 bytes/sec
[05:15] <fabbione> total size is 540127232  speedup is 23.91
[05:17] <fabbione> of course it depends a lot of bw you have
[05:17] <fabbione> i am ok with rsync
[05:17] <fabbione> and offload my machine
[05:17] <fabbione> i am not one of these lucky bastards
[05:17] <fabbione> with amd64 and 64GB of ram
[05:17] <Kamion> thom: hey, jigdo actually works still? cool. somebody complained it didn't.
[05:17] <Kamion> I'd been meaning to test
[05:17] <seb128> jdub: is there a public archive for the devel version of g-a-i ?
[05:18] <jdub> in ross's arch repo
[05:19] <fabbione> Kamion: grub-install fail
[05:19] <seb128> jdub: k
[05:19] <fabbione> Kamion: grub-install /dev/hda
[05:19] <fabbione> bum 
[05:19] <jdub> seb128: burtonini.com/arch
[05:19] <seb128> thanks
[05:20] <maswan> Mithrandir: Well, it _can_ use the old iso, but that involves non-obvious but somehwat documented trickery
[05:20] <fabbione> nope.. it doesn't recover
[05:21] <thom> it worked just peachily for last two dailies
[05:21] <fabbione> oh this is WEIRD!
[05:22] <fabbione> it's an IDE only machine.. partitioner saw the disk properly
[05:22] <fabbione> now cat /proc/partitions
[05:22] <fabbione> and it's all SCSI?
[05:22] <fabbione> since when we can update IDE to SCSI via software?
[05:22] <zul> sata?
[05:22] <fabbione> that's a really cool feature
[05:22] <fabbione> zul: not that i am aware of...
[05:23] <fabbione> it still makes people think that they have IDE and at end it is seen as SCSI
[05:23] <zul> thats weird
[05:23] <thom> the new background is pretty sexy
[05:23] <tseng> thom: the new gdm is pretty not
[05:23] <lamont> Kamion: installing on a multi-disk machine the other day, it wanted me to type in where to put the boot record...  the cute part was it suggesting hda, when the machine is SCSI...
[05:24] <thom> hrm, not sure i noticed or saw
[05:24] <tseng> it just makes the ubuntu logo 2x bigger
[05:24] <Kamion> lamont: grub-installer makes my head hurt, in general
[05:24] <jdub> tseng: gdm is not? it's the same apart from the logo
[05:24] <thom> yeah, the multidisk grub message is pretty darn awful :(
[05:24] <tseng> jdub: "not sexy"
[05:24] <Kamion> fabbione: we turned on LIBATA_ENABLE_ATAPI, or whatever it's called
[05:24] <jdub> tseng: you don't like it?
[05:25] <tseng> no =/
[05:25] <jdub> man
[05:25] <jdub> you're like
[05:25] <fabbione> Kamion: ah
[05:25] <jdub> the only person in the entire world
[05:25] <tseng> i liked the nice subtle size
[05:25] <Mithrandir> maswan: I thought it asked?
[05:25] <fabbione> Kamion: that can mostlikely break....
[05:25] <Kamion> fabbione: for #1440
[05:25] <Kamion> fabbione: sabdfl override for #1440, we were running out of options
[05:25] <tseng> jdub: ill get over it :P
[05:25] <maswan> Mithrandir: Oh, it did? Well, I've never looked that hard, since I've always had a package mirror close enough.
[05:26] <thom> yeah, it asks for an oldiso
[05:26] <Mithrandir> maswan: how strange. ;)
[05:26] <ogra> tseng: didnt you know ? it woll grow with each 1mio new users ;-P
[05:26] <ogra> will even
[05:26] <tseng> ah
[05:27] <tseng> 10000000 served
[05:27] <lamont> Kamion: heh
[05:28] <fabbione> Kamion: ok....
[05:28] <thom> woah!
[05:28] <fabbione> anyway netistall wasn't too happy
[05:28] <thom> that is one big ubuntu logo
[05:28] <fabbione> Kamion: i got aptitude at base-config
[05:28] <fabbione> and running manually the aptitude install ubuntu-desktop was ok
[05:28] <fabbione> i will have to check again
[05:29] <thom> jdub: two people in the entire world, now
[05:31] <jdub> thom: hrm
[05:31] <jdub> thom: too big?
[05:31] <seb128> jdub: are we going to have a new g-i-a in the archive soon ? Just wondering for the translations mail I'm writting. 
[05:31] <thom> yep
[05:31] <jdub> thom: can you sshot?
[05:31] <seb128> jdub: the translations should be sent to ross ?
[05:31] <jdub> seb128: unlikely
[05:31] <thom> sure
[05:31] <jdub> seb128: no, we should do them
[05:31] <seb128> grumpf
[05:31] <thom> when this install run finishes
[05:31] <jdub> seb128: ross is away for two weeks
[05:32] <seb128> oh
[05:32] <seb128> and we are not updating to a translated g-i-a soon ?
[05:32] <jdub> seb128: just focusing on gnome release atm
[05:32] <mvo> jdub, seb128 : I already branched a archive, I can deal with it 
[05:33] <seb128> jdub: right, sorry
[05:33] <seb128> mvo: I put your mail for update-manager/update-notifier/gnome-app-install translations ?
[05:34] <mvo> seb128: yes
[05:34] <seb128> thanks
[05:35] <Kamion> fabbione: base-config does 'aptitude install ~tubuntu-desktop', not just ubuntu-desktop
[05:35] <mvo> jdub: I removed the "services" menu from my gnome-app-install branch (#7311). IIRC you said you don't want to use it for -final?
[05:35] <Kamion> fabbione: the Task headers in the archive are probably out of date; bug elmo to update them
[05:35] <jdub> mvo: later please :)
[05:35] <mvo> jdub: we can talk later about it, I guess you are busy, sorry
[05:35] <mvo> sure :)
[05:35] <Kamion> fabbione: so what should it have been, instead of /dev/hda?
[05:36] <fabbione> Kamion: i had to put /dev/sda but now i have a doubt that i will need to check again later
[05:37] <fabbione> i am not 100% sure it was hda in partitioner 
[05:37] <fabbione> or sda
[05:37] <Kamion> ok
[05:37] <fabbione> i just read the partition numbers without too much care
[05:37] <fabbione> if the 2 matches is ok
[05:37] <fabbione> but if they don't than it's an issue for the user
[05:39] <fabbione> YES this media can boot
[05:39] <fabbione> Kamion: testing cd install now
[05:40] <thom> looks good from here so far
[05:41] <Kamion> that ide/scsi thing is kinda bad, not sure it should delay preview since we have no clue what it is, but we need to look at it for final
[05:41] <Kamion> I have no scsi systems though
[05:41] <fabbione> Kamion: this is IDE
[05:41] <Kamion> or any that think they're scsi :) only an SATA system that works fine
[05:41] <fabbione> the cdrom fails here
[05:42] <fabbione> it fails loading the installer components
[05:42] <Kamion> has it successfully mounted /cdrom?
[05:42] <fabbione> Kamion: can you give me the md5sum for the install iso?
[05:42] <fabbione> Kamion: checking
[05:42] <fabbione> Kamion: yes it is mounted
[05:42] <Kamion> fabbione: there's an MD5SUMS file next to the ISOs ...
[05:43] <Kamion> 90fae3a5bc65db83d8fa8fe4aa8c0d8a  hoary-install-i386.iso
[05:43] <thom> this is a mixed pata/sata system and it's great
[05:43] <fabbione> 90fae3a5bc65db83d8fa8fe4aa8c0d8a  hoary-install-i386.iso
[05:43] <fabbione> looks ok
[05:44] <fabbione> weird.. this is all weird
[05:44] <fabbione> at the second read is ok
[05:46] <pitti> mdz, Kamion: 20050309.1 ppc/install on iBook G4 works fine
[05:48] <thom> amd64 is locale-gen-of-justice'ing
[05:48] <Mithrandir> (:
[05:49] <thom> Kamion: amd64 looks good for me
[05:51] <fabbione> Kamion: forget about the IDE/SCSI thingy
[05:51] <fabbione> it is ok
[05:52] <Kamion> fabbione: what went wrong the first time?
[05:52] <fabbione> me reading partitioner
[05:52] <fabbione> the hda/sda is on top
[05:52] <zul> hah!
[05:52] <fabbione> and i didn't even check it since there is only one disk
[05:52] <fabbione> and i looked only at the partition numbers
[05:53] <fabbione> but i knew that it was hda from the last installs i did
[05:53] <Kamion> but did grub-installer not do the right thing by default?
[05:53] <fabbione> (it's the multiarse setup i am testing on)
[05:53] <fabbione> Kamion: grub did prompt to know to which partition or disk i wanted him to install the MBR
[05:54] <Kamion> oh, and you typed in the wrong thing by hand?
[05:54] <fabbione> yes.. since i knew it was hda
[05:55] <Kamion> heh, ok
[05:55] <Kamion> right, that's a relief
[05:55] <fabbione> yes
[05:55] <fabbione> definetely
[05:55] <jdub> gnome's shipped
[05:55] <sivang> jdub: yay!
[05:55] <fabbione> i am installing again via net
[05:55] <fabbione> to see the second stage problem with ubuntu-desktop
[05:57] <Kamion> elmo: you around? could you regenerate hoary's Task headers?
[05:57] <fabbione> that would be helpful
[05:57] <fabbione> i can put the install in sleep for a little while :-)
[05:57] <fabbione> while i test the live CD
[05:57] <fabbione> the first install cd doesn't go 
[05:57] <fabbione> it must be a media problem
[05:58] <ogra> yeah, Celebrating the release of GNOME 2.10!
[06:02] <sivang> ogra: wheeee
[06:03] <fabbione> Kamion: i am sorry but i get the same error from the installer on the LiveCD
[06:03] <fabbione> i doubt it is only a cd burning issue
[06:03] <fabbione> "Unable to load installer components"
[06:03] <fabbione> and i am not doing anything weird.. just booting
[06:04] <pitti> mdz, Kamion: 20050309.1 ppc/live on iBook G4 works great, too
[06:04] <fabbione> and also on this machine (NON-SATA) my disk is now scsi
[06:04] <pitti> now I really have to run
[06:04] <pitti> cu tomorrow
[06:04] <fabbione> weird enough.. they are all intel chipsets
[06:04] <pitti> jdub: congrats for 2.10 :-)
[06:04] <jdub> pitti: :)
[06:05] <thom> yaaargh
[06:05] <thom> auto key guesser doesn't get a UK powerpc usb keyboard at all correct
[06:06] <thom> it ended up asking me for a set of keys none of which i had
[06:08] <Kamion> fabbione: please look through syslog and try to figure out why it's doing that
[06:08] <fabbione> Kamion: i am working on it. don't worry :-)
[06:09] <fabbione> ah i found one machine where it works
[06:09] <fabbione> the other 2 have the same problem
[06:10] <Kamion> both install-i386 and install-amd64 are going well for me so far
[06:10] <fabbione> Kamion: ok.. 2 things.. machine 1 & 2: boot from net -> ok. boot from cd -> nok. machine 3 is ok.
[06:10] <fabbione> it's not really a GO for me
[06:10] <fabbione> also because this is pretty common hardware
[06:10] <Kamion> i386 on Averatec test laptop, amd64 on SATA disk / IDE CD
[06:10] <Kamion> fabbione: dude I can't do anything until you give me log details :)
[06:11] <fabbione> Kamion: yes.. i am working on it :-)
[06:11] <Kamion> we're also well past sabdfl's deadline for starting the final build for preview
[06:11] <fabbione> well.. it's a preview
[06:11] <Kamion> so we may just have to diagnose issues as best we can and do Array 7 ASAP
[06:11] <fabbione> right
[06:12] <thom> ppc is installing core packages
[06:12] <thom> amd64 and x86 (same box) are go from here
[06:12] <fabbione> thom: do you have i386 to test?
[06:12] <fabbione> a plain i386?
[06:12] <thom> fabbione: amd64 in x86 mode
[06:12] <thom> no difference
[06:12] <Kamion> i386 unpacking desktop, amd64 downloading language-support-de deps
[06:13] <thom> i can do a netboot on my laptop now i guess
[06:13] <fablive> live seems ok
[06:13] <fablive> let's go back in debugging
[06:21] <fabbione> Kamion: i found 3 possible reasons of failure
[06:21] <fabbione> Kamion: just a sec that i need to copy and paste manually
[06:22] <fabbione> DEBUG: unable to find (libc6) ignored
[06:22] <fabbione> main-menu: sed: Unsupported command T
[06:23] <fabbione> main-menu: cat /proc/ide/*/media : No such file or dir
[06:23] <fabbione> (possibly due to the change to scsi)
[06:23] <Kamion> jdub: hm, no rounded window corners in Human?
[06:24] <fabbione> anna[<pid>] ; WARNING **: parser_rfc822: lek! Don't find end of field, it seems to be after the end of line!
[06:24] <Kamion> fabbione: the sed T thing should be harmless; need to implement that in busybox, will probably do it pre-final
[06:24] <jdub> Kamion: mark wanted the old wm theme, not the clearlooks one
[06:24] <fabbione> anna[<pid>] ; ERROR: can't find packages file
[06:24] <Kamion> fabbione: hm, ok, does /cdrom/dists/hoary/main/debian-installer/binary-i386/Packages* look sane?
[06:25] <jdub> Kamion: i expect that a bug will be opened about that, lots of people *love* the clearlooks wm theme
[06:25] <fabbione> main-menu: Configuring 'load-cdrom' failed without error code 1
[06:25] <Kamion> jdub: the wm thing seemed to be the biggest change in clearlooks; don't really get the point of switching otherwise :-)
[06:25] <jdub> Kamion: all the lovely widgets!
[06:26] <fabbione> Kamion: no. it's corrupted
[06:27] <fabbione> approx at 53%
[06:27] <fabbione> Packages: input-modules-blabla
[06:27] <fabbione> in the description
[06:27] <fabbione> and later too
[06:28] <fabbione> around 60%
[06:28] <lamont_r> how goes the RC?
[06:28] <Kamion> fabbione: that would be it then ...
[06:28] <fabbione> Kamion: does it look corrupted on your machine too?
[06:29] <fabbione> because the md5sum matches with the iso's
[06:29] <fabbione> and both Live and install?
[06:29] <Kamion> the /proc/ide/*/media thing is from a udev script, either ide-devfs.sh or scsi-devfs.sh
[06:29] <Kamion> fabbione: it's fine on mine or it wouldn't have installed successfully
[06:30] <fabbione> zcat Packages.gz | more
[06:30] <fabbione> zcat: inflate error
[06:30] <fabbione> 1
[06:31] <Kamion> looks fine here
[06:31] <Kamion> live-i386
[06:31] <Kamion> md5sum 24323b5da4533d415d1821cb7a3ded24
[06:32] <fabbione> checking with a 3rd reader
[06:32] <Kamion> (of the Packages.gz file)
[06:32] <fabbione> yes the md5sums of the iso's are ok
[06:32] <mvo|live> live-cd looks fine here (boots, comes up with correct keyboard, X, sound, network). anything particular you want me to test?
[06:32] <trukulo> fabbione, 2.6.11.2 out (if you don't know)
[06:32] <Kamion> fabbione: check md5sum /cdrom/dists/hoary/main/debian-installer/binary-i386/Packages.gz
[06:32] <fabbione> Kamion: yes. doing that now
[06:33] <fabbione> 24323b5da4533d415d1821cb7a3ded24  Packages.gz
[06:33] <fabbione> looks ok!
[06:33] <jdub> OK: Checksums match, image is good!
[06:34] <Kamion> what was the gdm change people were complaining about?
[06:34] <fabbione> this is so weird....
[06:34] <Kamion> it looks about the same to me
[06:34] <thom> Kamion: the UBUNTU logo is ginormous
[06:34] <lamont_r> Kamion: does the preview get a weekly-dvd?
[06:34] <Kamion> lamont_r: no
[06:34] <fabbione> i suspect some memory corruption due to the last changes
[06:34] <lamont_r> partypooper
[06:34] <fabbione> since on the machine where the file is ok
[06:34] <fabbione> the cdrom is recognized as IDE
[06:35] <fabbione> and not as SCSI
[06:35] <fabbione> elmo: sorry dude, did you update the task list?
[06:35] <Kamion> fabbione: live-i386 boots fine here, sorry I can't help more
[06:35] <fabbione> Kamion: don't worry...
[06:35] <fabbione> that's the only thing i can think of
[06:36] <fabbione> and it would definetely matches the only changes that have been done to the kernel
[06:36] <fabbione> and that scares the hell out of me
[06:36] <Kamion> fabbione: yeah, could be libata screwage
[06:36] <Kamion> fabbione: IIRC zul mentioned libata had had some more work done to it in 2.6.11
[06:36] <ogra> lamont_r, is that the semi official name for the preview ? (perky partypooper ?)
[06:36] <Kamion> lamont_r: mdz's decision :)
[06:37] <lamont_r> Kamion: OK, then.  mdz is a party pooper.  I'll quote you on that. :-))
[06:37] <lamont_r> just lack of testing time?
[06:38] <fabbione> does busybox implements md5sum?
[06:38] <lamont_r> fabbione: trying to decide which answer would confuse me more...
[06:38] <Kamion> fabbione: yes
[06:38] <Kamion> thom: seems ok here
[06:39] <fabbione> lamont_r ?
[06:39] <lamont_r> fabbione: on busybox/md5sum
[06:40] <fabbione> ok
[06:40] <fabbione> libata is screwed
[06:40] <fabbione> confirmed
[06:40] <lamont_r> neat
[06:40] <fabbione> md5sum of Packages.gz via busybox is totally different
[06:40] <Kamion> yay :-/
[06:41] <Kamion> so I guess that's an errata item
[06:41] <fabbione> unless somebody wants a fix on the fly
[06:41] <fabbione> no correction
[06:41] <fabbione> unless somebody wants to find and produce a fix on the fly
[06:42] <zul> but it works on some and not on others?
[06:42] <sivang> have we released yet? 
[06:42] <Kamion> sivang: no
[06:42] <lamont_r> fabbione: the "only change" was defining two config vars
[06:42] <fabbione> i am afraid that the others didn't even see the problem
[06:42] <Kamion> sivang: I told you we were aiming for 1900 UTC, it is now 1742
[06:43] <Kamion> speaking of kernel stuff, can we please please please revert that powerpc LEDs thing post-preview?
[06:43] <sivang> Kamion: right sorry.
[06:43] <Kamion> s/LEDs/LED/
[06:43] <jdub> Kamion: ugh, yeah
[06:43] <jdub> Kamion: already getting complaints about it
[06:43] <Kamion> or make it switchable default off, but preferably just nuke it for hoary
[06:43] <sivang> when can we agree on the preview release announcment being final? I want to start transalte it :)
[06:44] <thom> ppc up to ttf-malayalam-fonts
[06:44] <lamont_r> wrt LED
[06:44] <sivang> (althought it can probably be left to about 30 minutes before release)
[06:45] <fabbione> lamont_r: i am checking what these 2 vars do
[06:45] <fabbione> it can be a driver <-> scsi remapping thing
[06:46] <fabbione> not necessarely a bug in libata
[06:46] <Nafallo> fabbione: got time for a question about 2.6?
[06:47] <fabbione> no
[06:47] <Nafallo> dang
[06:47] <fabbione> Nafallo: sorry but i am very busy atm
[06:47] <lamont_r> about 5 min, and I must go fetch kid.
[06:47] <Nafallo> fabbione: that's okey :-).
[06:48] <lamont_r> Kamion: the fact that she's sick seemed (to her) insufficient reason to miss the nationa latin exam...
[06:48] <Kei> hey google www.otomotivshow.com 
[06:49] <jdub> hrm, my cd burners are all unhappy today
[06:51] <thom> PPC is a GO!
[06:52] <Kamion> lamont_r: heh
[06:54] <thom> lamont_r: new install:
[06:54] <thom> Mar  9 17:53:37 localhost postfix/local[9097] : fatal: open database /etc/aliases.db: No such file or directory
[06:54] <thom> Mar  9 17:53:38 localhost postfix/master[7122] : warning: process /usr/lib/postfix/local pid 9097 exit status 1
[06:54] <thom> Mar  9 17:53:38 localhost postfix/master[7122] : warning: /usr/lib/postfix/local: bad command startup -- throttling
[06:54] <lamont_r> thom: see 6232, iirc
[06:55] <lamont_r> fixed in -9ubuntu2, scheduled to upload tonight, after the preview ships
[06:55] <fabbione> who has a piix chipset handy? meaning that you can try stuff on it?
[06:55] <lamont_r> thom: very interesting interactions between base-installer (et.al.) and postfix's postinst
[06:56] <lamont_r>   * On initial install, make sure that /etc/aliases.db gets created.
[06:56] <lamont_r>     Closes: Ubuntu#2683, Ubuntu#6232 (debian #293889)
[07:01] <mdz> good morning
[07:01] <thom> hey matt
[07:02] <jdub> morning
[07:02] <mdz> what's on the menu?
[07:02] <fabbione> Kamion: we might have a fix...
[07:02] <Kamion> looking pretty good so far, apart from a libata regression reported by fabbione
[07:02] <fabbione> it's only on piix chipset
[07:02] <Kamion> ok
[07:02] <fabbione> we are isolating and taking a decision on what to do
[07:02] <mdz> oh good, new kernel, we hadn't changed that one yet :-P
[07:02] <Kamion> haha
[07:03] <Kamion> I doubt we'll try to fix it for preview
[07:03] <Kamion> mdz: everything else is in, built, tested though
[07:03] <lamont_r> back in about 45 minutes
[07:03] <fabbione> Kamion: we actually can
[07:03] <fabbione> it's a 2 line revert a patch for sabdfl 1440 override
[07:03] <fabbione> or we can sabdfl what he prefer :-)
[07:04] <Kamion> that breaks other systems
[07:04] <fabbione> silent data corruption..
[07:04] <mpt_london> here comes the sab
[07:04] <Kamion> fabbione: we could, but any rebuild now takes us way past the deadline
[07:04] <sabdfl> hi all
[07:04] <fabbione> Kamion, mdz: your call
[07:04] <sabdfl> where are we?
[07:04] <mdz> fabbione: you can reproduce the bug?
[07:05] <fabbione> mdz: yes, on 2 boxes
[07:05] <fabbione> i found it
[07:05] <T-Bone> hey sabdfl
[07:05] <mdz> that patch fixed unable-to-install problems for a lot of users
[07:05] <sabdfl> mdz, Kamion, fabbione: what's the status please
[07:05] <mdz> sabdfl: I just got here, catching up myself
[07:05] <fabbione> mdz: the other introduces silent data corruption and mostlike another bunch of "i am unable to install" users
[07:05] <mdz> sabdfl: fabbione has found a kernel problem
[07:06] <sabdfl> ok. AIUI the kernel we are about to ship still has 1440?
[07:06] <mdz> sabdfl: no, it has 1440 fixed
[07:06] <Kamion> a kernel rebuild now pushes us way into tomorrow UTC, and simply exchanges bugs for bugs
[07:06] <T-Bone> mdz: as i said on #u-kernel: i think it's not a good thing to introduce yet another untested "fix". Reverting to the previous situation were some users couldn't install but others were safe seems preferable, for 'preview' timeframe
[07:06] <mdz> sabdfl: and fabbione says that the fix introduced a bug
[07:06] <fabbione> sabdfl: no. it has the fix for 1440 that introduces another bug
[07:06] <fabbione> sabdfl: like I/O corruption. for what i can see only on CDROM
[07:06] <T-Bone> mdz: after preview we'll have more time to find a better solution
[07:06] <fabbione> sabdfl: but it means another set of people won't be able to install
[07:07] <fabbione> not even to see the LiveCD
[07:07] <Kamion> my feeling is that it is too late to revert for preview
[07:07] <mdz> how are you guys certain that this is caused by that patch?
[07:07] <fabbione> mdz: 99,9%
[07:07] <mdz> how, not how much
[07:07] <fabbione> because both these machines have piix controllers
[07:07] <fabbione> and the change to ATA_ENABLE_PATA
[07:08] <sabdfl> fabbione: do you have a fix for the new problem? has this been discussed and solved upstream or elsewhere?
[07:08] <fabbione> adds 4 pci ids to the list of ata_piix.c
[07:08] <fabbione> that matches the 2 machines i have
[07:08] <fabbione> as a result my ide disks are recognized as scsi
[07:08] <Kamion> would using some other module instead work around the problem?
[07:08] <fabbione> booting from the normal kernel (as IDE)
[07:08] <fabbione> everything works fine
[07:08] <mdz> this patch has been in our kernel for over a month
[07:09] <Kamion> as in, could we monkey-patch modules.pcimap or something? :-)
[07:09] <fabbione> sabdfl: i discovered it 1 hour ago Mark.. working with the others..
[07:09] <Kamion> right now anything is preferable to a kernel rebuild
[07:09] <T-Bone> mdz: the libpata enable? No. 2 weeks at max
[07:09] <mdz>   * Enable ATAPI support in libata [Chuck Short <zulcss@gmail.com>] 
[07:09] <mdz>     Patch: enable_atapi_ata.dpatch
[07:09] <mdz>     Closes: #1440
[07:09] <mdz>  -- Thibaut VARENE <varenet@debian.org>  Tue, 01 Feb 2005 19:56:33 +0100
[07:09] <fabbione> mdz: no it's very recent..
[07:09] <mdz> T-Bone: is that not when you uploaded that kernel?
[07:09] <fabbione> it was introduced while i was away
[07:09] <T-Bone> hmmm
[07:10] <fabbione> that means between the 17th of Feb and no
[07:10] <fabbione> now
[07:10] <jbailey> T-Bone: The changelog entry says 1 Feb, which has got to be way wrong.
[07:10] <T-Bone> definitely
[07:10] <fabbione> sabdfl: a possible fix would be to revert the change for preview and reopen 1440
[07:10] <mdz> hmm, yes, the changelog is not right
[07:10] <T-Bone> mdz: if it's -25, it's one or two weeks old
[07:10] <mdz> it was built on 2005-03-02
[07:10] <mdz> T-Bone: why is the date wrong in the changelog?
[07:10] <fabbione> so a week ago
[07:10] <T-Bone> mdz: no clue
[07:11] <mdz> sabdfl: reverting the patch means missing our date
[07:11] <Kamion> mdz: bad clock on build machine probably
[07:11] <T-Bone> mdz: in any case, if taking action on the kernel means missing our date. we're in hell
[07:11] <sabdfl> this is a bug that has bitten us again and again
[07:12] <fabbione> mdz: that's for sure.. it will need another round of testing everywhere
[07:12] <sabdfl> i've personally encountered it when demonstrating warty to potential hardware oem partners
[07:12] <Kamion> fabbione: forget that, even just the build time practically takes us into tomorrow
[07:12] <mdz> I think we should go out with the kernel we have
[07:12] <sabdfl> any high end new machine has all-SATA drives, and pata cdroms, right?
[07:12] <fabbione> Kamion: exactly...
[07:12] <T-Bone> mdz: the shortest path solution to that problem is reverting the patch. It needs 2 lines, and a set of builds
[07:12] <mdz> and tackle this bug after preview
[07:12] <T-Bone> mdz: that way we're back to known state
[07:12] <Kamion> T-Bone: that's not a shortest path
[07:12] <T-Bone> and not tempting anything dangerous
[07:12] <Kamion> T-Bone: it is minimum 6 hours
[07:13] <fabbione> mdz: i am afraid of one thing only...
[07:13] <mdz> T-Bone: I'm not convinced that the old state is better, and as I said, that's going to make us miss our release date
[07:13] <T-Bone> Kamion: i don't see anything shorter
[07:13] <fabbione> mdz: right now i saw cdrom I/O corruption
[07:13] <Kamion> T-Bone: listing it as an errata item
[07:13] <sabdfl> mdz: so the current kernel does not have 1440, but has a new bug that affects an unknown number of machines?
[07:13] <fabbione> mdz: allow me to run some tests on the disks
[07:13] <mdz> sabdfl: correct
[07:13] <fabbione> mdz: just be sure we are not going to trash anything else
[07:14] <T-Bone> Kamion: hell of an errata :(
[07:14] <mdz> fabbione: by all means
[07:14] <sabdfl> do we know which has the fewer number of affected machines? does it only affect a subset of the machines that 1440 affects, for example?
[07:14] <fabbione> i had say approx 40 minutes of bonnie or md5 to stuff
[07:14] <mdz> sabdfl: we have no reasonable data for that; there isn't time
[07:14] <fabbione> sabdfl: everybody that has a piix controller with the following pc ids:
[07:14] <Kamion> sabdfl: so far piix is the only chipset from which we've had a reported problem
[07:14] <Kamion> to my knowledge, anyway
[07:14] <mdz> sabdfl: it affects an entirely different set of machines
[07:14] <sabdfl> it's very common
[07:14] <fabbione> #ifdef ATA_ENABLE_PATA
[07:14] <fabbione>         { 0x8086, 0x7111, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix4_pata },
[07:14] <fabbione>         { 0x8086, 0x24db, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich5_pata },
[07:14] <fabbione>         { 0x8086, 0x25a2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich5_pata },
[07:14] <fabbione> #endif
[07:14] <Kamion> might be worth trawling the mailing lists for reports since array 6
[07:14] <fabbione> so anybody using cdrom from that chipset = BUUM
[07:15] <mdz> I really don't feel that we have a choice
[07:15] <Kamion> if it's only cdrom, it will not lose people's data
[07:15] <jbailey> fabbione: But only if they happen to get the ata_piix driver loaded instead of piix, right?
[07:15] <mdz> nobody spotted this in the past week, that we know of, and we are _out of time_
[07:15] <fabbione> Kamion: i really want to be sure
[07:15] <T-Bone> fabbione: it's worse than that. It's not 'buum', it's silent corruption. Very hard to figure out for the average user
[07:15] <fabbione> T-Bone: right.. 
[07:15] <mdz> Kamion: is the current set of CDs our best candidate, apart from this bug?
[07:16] <mako> erghh..
[07:16] <mdz> Kamion: it has all the grumble last-minute artwork changes?
[07:16] <Kamion> mdz: right
[07:16] <sabdfl> mdz: and we don't have a known-good fix, so delaying doesn't really help, we either get 1440 back or this new thing back
[07:16] <T-Bone> sabdfl: correct
[07:16] <sabdfl> mdz: let's release
[07:16] <mdz> I'd sort of like to test the release first
[07:16] <fabbione> sabdfl: please allow me to run some disk I/O tests first
[07:16] <fabbione> sabdfl: we must be 100% that the bug affects only cdrom
[07:17] <mdz> fabbione: please do
[07:17] <fabbione> we don't want to destroy people data for a mistake
[07:17] <sabdfl> ok. how about defer decision for a specific time, say 3 hours, to do tests
[07:17] <Kamion> I suppose Message-Id: <20050304045649.64C6887@orb.sasl.smtp.pobox.com> is similar
[07:17] <mdz> I'm going to do around of install+live testing
[07:17] <mdz> I need <1 hour
[07:17] <fabbione> sabdfl: i am already running the tests.. i had say one hour?
[07:17] <sabdfl> in the meanwhile, mirrors are running, and we can ask people to help us test
[07:17] <sabdfl> can we get -devel to help out?
[07:18] <sabdfl> "here is the candidate, we have a potential erratum on cdrom corruption, please test and send feedback to this email address before 21h00 UTC"
[07:18] <tseng> -devel responses pan out over days
[07:18] <T-Bone> true
[07:18] <T-Bone> the best channel is here
[07:18] <T-Bone> it has the lowest latency of all
[07:18] <thom> might be worth asking in #ubuntu if anyone has piix mobos
[07:18] <T-Bone> right
[07:18] <thom> or controllers
[07:18] <sabdfl> Kamion: is there an iso we can point people at immediately?
[07:19] <mdz> fabbione: is DMA  enabled on your CD-ROM in d-i by any chance?
[07:19] <mdz> sabdfl: yeah, the stuff in the topic
[07:19] <T-Bone> we should warn in big letters of the potential dangers
[07:19] <Kamion> we've actually had several reports of new problems on array 6, but none of them mentioned the chipset
[07:19] <Nafallo> only s-ata computers, right?
[07:19] <sivang> oops, wasn't here for sometime, what is the bug needs testing?
[07:19] <Kamion> I'm trying to find each of them and follow up now
[07:19] <sivang> (and could it possibly corrput)
[07:19] <mvo> thom: I have a test-system with PIIX4 ...
[07:20] <Kamion> there's a report of a problem on an old 233mhz laptop; would something of that vintage have piix?
[07:20] <T-Bone> Kamion: yes
[07:20] <mdz> seb128: I just had my panel crash again, when gaim failed to connect to jabber
[07:20] <mdz> I was a moron though and didn't get a backtrace
[07:20] <jbailey> Kamion: My p3 900mhz laptop has piix, so it might.
[07:20] <sabdfl> piix has been around forever
[07:20] <mdz> I clicked through it because I was in the middle of something
[07:20] <mdz> jbailey: can you test?
[07:20] <T-Bone> right
[07:21] <Kamion> I found three reports
[07:21] <Kamion> but I can't say for sure that they're the same bug
[07:21] <jbailey> mdz: Yes, but I'm trying to get an answer to my driver question.  I've asked twice with no answer. =)
[07:21] <thom> mvo: can you test current preview on that machine and see if you have problems/
[07:21] <mvo> I think I see this problem here. my cdrom/disks are detected as scsi and installation fails with "can't get stuff from cd"
[07:21] <Kamion> sabdfl: do we have mirrors running?
[07:21] <fabbione> mdz: i didn't check. i can do it now
[07:21] <mvo> (this is a p2/400, pretty old machine)
[07:21] <sabdfl> Kamion: i believe so... is the image on releases.ubuntu.com?
[07:22] <Kamion> sabdfl: no, because it is not released
[07:22] <Kamion> I need a go-ahead for that
[07:22] <seb128> mdz: weird. I'm trying to play with gaim/jabber
[07:22] <Kamion> I also want to test on powerpc first; about to do so
[07:22] <sabdfl> ok, i thought we might be putting it somewhere as a .file which we could then just symlink to
[07:22] <sabdfl> making the final mirror trivially fast
[07:23] <sabdfl> can we do that for the final release?
[07:23] <fabbione> brb
[07:23] <Kamion> yes, I can do that now too, er probably
[07:23] <Kamion> test cycles tend to leave me in a slightly awkward situation machine-wise :-)
[07:23] <mdz> seb128: jabber is flaky today, so hopefully it will happen again
[07:24] <Kamion> and I don't actually have any script support for that yet so I'll do it by hand
[07:24] <mvo> ok, I have a system that is affected. how can I help now ?
[07:24] <trukulo> gaim works well here
[07:24] <sivang> here also
[07:24] <trukulo> with last hoary package
[07:25] <Nafallo> what do I need to test, an up2date hoary kernel?
[07:25] <mdz> er
[07:25] <mdz> the amd64 live CD still has the old artwork?
[07:25] <martink> seb128, poppler 0.1.2 announce mail: "... plus the cairo version we check for now actually works with poppler" ;-)
[07:27] <seb128> martink: nice
[07:27] <mdz> mvo: :-(
[07:27] <mdz> mvo: I thought you tested the live CD earlier today
[07:27] <mvo> mdz: yes :( anything I can do now? any workarounds I could try? 
[07:27] <mdz> mvo: we haven't changed anything since then
[07:27] <mdz> in the kernel
[07:28] <mdz> mvo: or are you talking about the jabber problem?
[07:28] <mvo> mdz: yes :( I thought my cdrom in my old test-machine
[07:28] <mvo> mdz: on my "real" system live-cd works fine
[07:28] <mdz> mvo: so you are testing in a different machine than earlier?
[07:28] <mvo> mdz: I now tested on a older machine (p2/400) 
[07:28] <mvo> mdz: on my desktop (k7) everything is fine
[07:29] <mdz> mvo: what does hdparm on the cdrom say?
[07:29] <mdz> mvo: on the machine which experiences the bug?
[07:29] <mdz> ARGH
[07:29] <mdz> langpack upgrades are broken
[07:29] <mdz> and pitti is gone
[07:30] <mvo> mdz: IO_support = 0, readonly=0, readahead=256, HDIO_GETGEO failed
[07:30] <trukulo> mdz, i've got problems with langpack, removing it and installing again solved the problem
[07:30] <mdz> trukulo: yes, it's buggy
[07:30] <seb128> mdz: what is broken with langpack ?
[07:30] <mvo> mdz: I'm really angry about myself because I did test the cd earlier today in this old machine but I suspected a broken cdrom :(
[07:31] <trukulo> mdz, i know, it's a file included in langpack and langpack-base collisioning
[07:31] <Kamion> mdz: amd64 live CD is up-to-date here; I'm running it now
[07:31] <trukulo> but if i remove them, and install again, it works
[07:31] <fabbione> mdz: no DMA enabled
[07:31] <mdz> gah, I missed an rsync step
[07:32] <fabbione> mdz: and i can't turn it on with libata
[07:32] <fabbione> it complains that it is an invalid IOCTL
[07:32] <Kamion> sabdfl: I'm putting them in .pool now
[07:33] <Kamion> I'll have to do some grotty removing to stop them showing up in cdimage.ubuntu.com/releases/ though
[07:33] <Kamion> but that's ok
[07:33] <sabdfl> Kamion: if they had .filenames, would they show up in HTTP and FTP LS?
[07:34] <Kamion> sabdfl: no
[07:35] <Kamion> sabdfl: really I'd rather just remove them from cdimage, it's too complicated otherwise
[07:35] <Kamion> they can still go on releases.ubuntu.com/.pool/ which is what you wanted
[07:36] <sabdfl> ok
[07:37] <sabdfl> releases.ubuntu.com should get lots more mirroring long term
[07:38] <sabdfl> so my thought was to stick them in .pool there as we get close to release
[07:38] <sabdfl> then on the last day it's just a question of creating the symlink to that
[07:38] <Kamion> yes, that's what I'm going to do now
[07:38] <sabdfl> then, when we release final, we remve preview
[07:38] <fabbione> Kamion: is there anyway to ban a module from being loaded ?
[07:38] <sabdfl> ok
[07:38] <fabbione> like an anna option or something?
[07:38] <Treenaks> bug on releases.ubuntu.com: clicking the "warty" link links to a page with the title: Ubuntu 5.04 (Hoary Hedgehog)"
[07:38] <sabdfl> and we can still point people at the url with the .pool filename for testing just before the release
[07:38] <Kamion> fabbione: er I can't check right now, could you check the anna source please?
[07:38] <mdz> fabbione: you think we can use ide-generic instead or something?
[07:39] <fabbione> mdz: i am checking if removing the modules from /lib/<kernel>/ helps
[07:39] <fabbione> Kamion: not atm...
[07:39] <Kamion> sabdfl: that doesn't work so well, hoary/hoary-preview-*.iso is a *symlink* to .pool/whatever
[07:40] <Kamion> sabdfl: so we have to rely on mirrors mirroring hard links properly, and I have no idea if they do so at the moment
[07:40] <Kamion> fabbione: generally I rm the thing
[07:40] <Kamion> fabbione: the wonderful world of hotplug hardware detection means that the installer has very little opportunity to intervene otherwise :-P
[07:40] <mdz> gah
[07:41] <tseng> =/
[07:41] <Kamion> sabdfl: let's talk about this outside a release panic
[07:41] <Kamion> Treenaks: will fix in a sec
[07:41] <sabdfl> Kamion: i did mean symlink
[07:41] <sabdfl> and said symlink too
[07:41] <Kamion> sabdfl: yeah, but if you have hoary/hoary-preview-*.iso symlink -> .pool/hoary-*.iso then you can't update .pool/hoary-*.iso or the preview ISO that people see will change
[07:42] <fabbione> Kamion: checking that too :(
[07:42] <sabdfl> Kamion: i figured to have a spare, unlinked iso in .pool
[07:42] <Kamion> sabdfl: like I say let's talk about this outside a release panic, when I have a chance to think about it :-)
[07:42] <sabdfl> Kamion: ok
[07:42] <mdz> can we hold off on changing the mirror layout and focus on the showstopper bugs in the preview release?
[07:43] <Kamion> fabbione: there might be /etc/hotplug/blacklist too, but that's not per-pci-id
[07:43] <fabbione> brb.. dinner is ready
[07:44] <mdz> lamont: ping
[07:45] <mdz> mvo: can you try fabbione's test?
[07:45] <mdz> mvo: see if suppressing the loading of the module works around the problem?
[07:46] <mdz> mvo: and do you have Martin Pitt's telephone number?
[07:46] <mdz> (or anyone else)
[07:46] <mvo> mdz: I will try to supress the module loading
[07:46] <Kamion> isn't it on the canonical wiki?
[07:46] <mdz> Kamion: no
[07:47] <mdz> ah, it is on the business card list though
[07:47] <mvo> mdz: should I call him?
[07:47] <mdz> mvo: yes, please get him online ASAP
[07:47] <lamont> mdz: ack
[07:47] <mvo> mdz: calling
[07:48] <mdz> lamont: do you have a PC with a PIIX chipset?
[07:49] <Kamion> sabdfl: syncing .pool out to syncproxy,auckland,mirnyy now
[07:49] <lamont> mdz: what's the best way to tell?
[07:50] <mdz> lamont: lspci
[07:50] <ogra> lamont: or look if the module is loaded
[07:50] <mdz> right
[07:50] <lamont> 00:04.1 IDE interface: Intel Corp. 82371AB/EB/MB PIIX4 IDE (rev 01)
[07:50] <lamont> like that?
[07:50] <ogra> yup
[07:50] <jbailey> lamont: The module on my laptop appears as piix, on my sata box as ata-piix
[07:51] <lamont> ogra: that particular machine happens to be the mail server
[07:51] <mvo> mdz: not reached on mobile and normal phone, will try again in 10 minutes
[07:51] <ogra> lamont, ouch
[07:52] <schweeb> mdz: something wrong with piix?  I have one too if you need something tested
[07:52] <mdz> lamont: yes, please get a copy of the latest live CD for testing
[07:52] <mdz> schweeb: yes, CD-ROM drives attached to those controllers seem not to work
[07:52] <mdz> schweeb: if you can get a copy of the latest live or install image, it would help us with testing
[07:52] <lamont> daily-live/current/hoary-live-i386.iso
[07:52] <lamont>    214515339  39%    1.08kB/s   83:42:26
[07:53] <schweeb> mdz: http://archive.ubuntu.com/cdimage/daily-live/20050309.1/ that the newest?
[07:53] <jbailey> Looks like we're all having the same speed issues to cdimage
[07:53] <mdz> schweeb: /daily-live/current/
[07:54] <lamont> jbailey: seb128 uploaded everything, so it's a huge beast
[07:54] <mdz> jbailey: Kamion is updating the mirrors right now I think
[07:54] <Kamion> right, they just finished
[07:54] <schweeb> alright... tell ya in about an hour
[07:54] <jbailey> CAn anyone other than fabionne currently reproduce the problem at all right now?
[07:54] <Kamion> sabdfl: synced; know of any mirrors that can be kicked?
[07:55] <mdz> mvo: were you able to leave messages for him?
[07:55] <Kamion> Treenaks: fixed your index issue now, thanks
[07:55] <mvo> mdz: no, no answer phone, I'll send him a SMS now
[07:56] <tseng> who is in charge of the new "about ubuntu"?
[07:56] <mdz> tseng: doc team
[07:56] <lamont> 0000:00:0f.1 IDE interface: ServerWorks OSB4 IDE Controller
[07:56] <lamont> grumble
[07:56] <tseng> the title doesnt show on the sidebar in yelo
[07:56] <tseng> off to search bugs
[07:56] <mvo> Kamion: is there a way to supress the loading of certain modules on the cd boot-prompt? (sorry if that was discussed before)
[07:57] <Kamion> mvo: no
[07:57] <sabdfl> Kamion: thats elmo's department
[07:57] <Kamion> hotplug makes that practically impossible for the installer to control in any sane way, AFAICT
[07:57] <sabdfl> how do i look to see if my computer is a candidate for the problem?
[07:57] <Kamion> sabdfl: lspci | grep -i piix should be a start
[07:58] <Kamion> mvo: you have to rm the module before it gets a chance to load it
[07:58] <mvo> Kamion: it's loaded before I have the chance even when running as expert :(
[07:58] <sabdfl> nup
[07:58] <sabdfl> neither box will help
[07:58] <mdz> mvo: a good way to do it is to use the install CD, and switch consoles at the hostname dialog
[07:58] <Kamion> mvo: boot init=/bin/sh
[07:58] <Kamion> mdz: that won't help if the module is in the initrd
[07:59] <Kamion> mvo: then rm the module and exec /sbin/init
[07:59] <tseng> mdz: huh, can I close 7310?
[07:59] <jdub> morning whiprush 
[07:59] <mdz> Kamion: I thought you arranged for it not to be loaded until "detecting disks and all other hardware"
[07:59] <mdz> tseng: is it a preview-critical issue?
[07:59] <Kamion> mdz: nope, don't see a sensible way to do that without killing e.g. CD-ROM detection
[07:59] <tseng> mdz: no, its fixed
[07:59] <mdz> ah, right
[08:00] <Kamion> and it is kind of nice to have the modules loaded early if possible - *usually*
[08:00] <mdz> tseng: can we talk about it tomorrow?  we're in crisis mode here
[08:00] <tseng> ok sorry.
[08:00] <mdz> schweeb: can you try what Kamion described above?
[08:00] <ogra> many people seem to have issues with the theme upgrade ?
[08:00] <jdub> ogra: two
[08:01] <schweeb> mdz: which thing?
[08:01] <tseng> ogra:you will always find someone to resist a change
[08:01] <mvo> Kamion: ok, deleted, what now?
[08:01] <Kamion> mvo: exec /sbin/init and continue
[08:01] <ogra> jdub: HiddenWolf showed me a screenshot before....it were plain gtk widgets, somehow clearlooks didnt get insalled for him
[08:01] <sivang> ogra: yes I also had it, like disturbed picture of the windows etc.
[08:01] <Kamion> ogra: update-manager doesn't do dist-upgrade
[08:01] <ogra> ah, ok
[08:01] <mdz> schweeb: boot with init=/bin/sh
[08:02] <mdz> schweeb: rm the piix module
[08:02] <ogra> thanks again, Kamion 
[08:02] <mvo> ogra: and the version pending has a warning if it detects that a dist-upgrade is needed
[08:02] <mdz> schweeb: then exec /sbin/init
[08:02] <mdz> (on the install or live CD)
[08:02] <Kamion> and new ubuntu-artwork does not depend on gtk2-engines-clearlooks
[08:02] <mdz> Kamion: is that a bug?
[08:02] <jdub> Kamion: uuuuggghhh
[08:02] <Kamion> mdz: judging from jdub's reaction I'd say "maybe" ;)
[08:02] <jdub> Kamion: but u-d does, so, not to scary
[08:02] <schweeb> mdz: will do, give the results in about an hour
[08:02] <jdub> not too
[08:03] <Kamion> jdub: quite bad if it actually needs it; 'apt-get upgrade' will break otherwise, since ubuntu-artwork will be upgraded but not ubuntu-desktop
[08:03] <Kamion> or, for that matter, update-manager
[08:03] <HiddenWolf> ogra: i didn't dist-upgrade, artwork doesn't depend on clearlooks, but desktop does, I think.
[08:03] <lamont> gah.  even unthrottled at my end, it's 2 hours to finish the downlaod.
[08:03] <lamont> time to visit the neighbor
[08:03] <schweeb> mdz: shall I try it without that first to see if my cdrom works?
[08:03] <jdub> Kamion: yeah, for people who don't have u-d, it sucks
[08:04] <jdub> is it worth the upload?
[08:04] <Kamion> jdub: even for people who do have u-d it sucks.
[08:04] <ogra> HiddenWolf: yep
[08:04] <Kamion> jdub: because if they have an old u-d it will simply be held back
[08:04] <lamont> back online in about 10-15 minutes
[08:04] <Kamion> mdz: what PCI ids are involved here?
[08:04] <mdz> Kamion: fabbione pasted them a ways back
[08:04] <jdub> Kamion: upgrade won't work, but dist-upgrade will, right?
[08:04] <sivang> Kamion: then we should add this the the depends of -artwork no?
[08:04] <jdub> sivang: yes, there is no question about that
[08:04] <dholbach> re
[08:05] <mdz> Mar 09 10:18:34 <fabbione>      #ifdef ATA_ENABLE_PATA
[08:05] <mdz> Mar 09 10:18:35 <fabbione>              { 0x8086, 0x7111, PCI_ANY_ID, PCI_ANY_ID, 0, 0, piix4_pata },
[08:05] <mdz> Mar 09 10:18:35 <fabbione>              { 0x8086, 0x24db, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich5_pata },
[08:05] <mdz> Mar 09 10:18:35 <fabbione>              { 0x8086, 0x25a2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich5_pata },
[08:05] <mdz> Mar 09 10:18:35 <fabbione>      #endif
[08:05] <Kamion> ah yes, and modules.pcimap lists *both* ata_piix and piix for those
[08:05] <Kamion> so there is hope
[08:05] <Kamion> jdub: right
[08:05] <Kamion> jdub: really nothing else can ever rely on ubuntu-desktop's dependencies for anything much, even if u-d is installed
[08:05] <jdub> Kamion: so i can fix that straight after preview
[08:05] <Kamion> jdub: yep
[08:06] <schweeb> 0000:00:1f.1 IDE interface: Intel Corp. 82801CAM IDE U100 (rev 02)
[08:06] <schweeb> that's the PCI ID of mine if that helps
[08:07] <Kamion> schweeb: that's not a PCI id, you need to use 'lspci -n' and find the corresponding line
[08:07] <Kamion> with 0000:00:1f.1 at the beginning
[08:07] <schweeb> whoops, meant to do that, sorry
[08:07] <schweeb> 0000:00:1f.1 0101: 8086:248a (rev 02)
[08:07] <Kamion> schweeb: is that broken?
[08:08] <Kamion> pcimap only lists piix for that
[08:08] <schweeb> thought piix was the offending module?
[08:09] <Kamion> schweeb: only certain ones we think; however, if you could try it out anyway, we'd be very grateful indeed
[08:09] <zul> hold on i have ata_piix loaded on my pc
[08:09] <Kamion> schweeb: if our current hypothesis holds, your machine should be OK
[08:09] <Simira> is anyone from Ubuntu going to "The gathering", in Norway in the Easter?
[08:10] <schweeb> Kamion: alright... downloading the livecd for testing now... the drive worked fine from array 5, if that's any indication
[08:10] <mdz> ok
[08:10] <mdz> executive decision
[08:10] <mdz> preview is going out tomorrow
[08:11] <Kamion> schweeb: array 6 would be an indication
[08:11] <Kamion> array 5 predates this change
[08:11] <ogra> mdz :(
[08:11] <Kamion> although it is a useful control
[08:11] <mdz> zul: are you here?
[08:11] <zul> mdz: yes
[08:11] <Kamion> so are we changing this in the kernel, or in hotplug?
[08:11] <Kamion> s/in hotplug/hacking hotplug/
[08:11] <mdz> zul: can you do an upload to revert the patch from #1440?
[08:12] <Kamion> mdz: can I suggest an alternative?
[08:12] <mdz> Kamion: to which?
[08:12] <Kamion> mdz: a total revert
[08:12] <zul> I dont have access to main
[08:12] <Kamion> mdz: we could instead just stop those PCI ids showing up in ata_piix
[08:12] <Kamion> which is a two-line kernel patch
[08:12] <mvo> Kamion, mdz: removing the module works so far. it gives a ugly error message at startup, but otherwise it is ok so far (install-cd)
[08:13] <mdz> Kamion: hmm
[08:13] <Kamion> since so far it sounds like only piix is affected, and this way we don't have to revert the whole thing
[08:13] <mdz> Kamion: so the change is actually causing a different module to be loaded for these controllers?
[08:13] <mdz> (one which doesn't work)?
[08:13] <Kamion> mdz: yes, ata_piix gains three PCI ids #ifdef ATA_ENABLE_PATA
[08:13] <mdz> or just changing behaviour for those IDs?
[08:13] <Kamion> those PCI ids are also registered with piix
[08:13] <Kamion> in fact I wonder if the problem could be that *both* modules are being loaded
[08:13] <mdz> does ata_piix have any IDs that piix doesn't?
[08:14] <mdz> or can we just blacklist it?
[08:14] <Kamion> mdz: yes
[08:14] <Kamion> it has other IDs
[08:14] <mdz> hm
[08:14] <Kamion> I have an idea
[08:14] <mdz> I'm not very confident about pushing forward on this, rather than rolling back
[08:14] <Kamion> mvo: could you please try booting without removing the module, go through until it fails, and then run 'lsmod | grep piix' for me?
[08:14] <mdz> not least because I can't get my hands on this bug due to lack of hardware
[08:15] <mvo> Kamion: yes
[08:16] <Kamion> mdz: bear with me here just for a little bit
[08:16] <mvo> Kamion: piix and ata_piix are both loaded
[08:16] <mdz> Kamion: certainly
[08:16] <mvo> your guess was correct
[08:17] <Kamion> mvo: ok, which module did you remove the first time round?
[08:17] <mjg59> Having two devices think they're running the DMA engine on there is going to make things very unhappy
[08:17] <Kamion> yes, that's my thought
[08:17] <mvo> Kamion: libata and ata_piix
[08:17] <Kamion> mvo: this time, try removing piix
[08:17] <mjg59> I'm surprised the kernel's letting that happen. At least one driver isn't doing the right thing for PCI.
[08:18] <mdz> mjg59: e100 and eepro100 used to do this as well
[08:18] <mjg59> mdz: 2.6 makes it a lot harder to do this
[08:18] <zul> sata_nv does it as well if i recall
[08:18] <mjg59> But IDE drivers are weird and wonderful and ALL BROKEN
[08:18] <lamont_r> mjg59: but clearly these drivers have worked hard to get around 2.6's limitation... :-)
[08:18] <fabbione> re
[08:19] <mdz> fabbione: so we seem to have a clearer idea what is going on; two modules are being loaded for the same device
[08:19] <T-Bone> re
[08:19] <Kamion> the patch now ought to be trivial
[08:19] <abelli> smurfix: ping 
[08:19] <smurfix> abelli: ?
[08:19] <fabbione> mdz: yes i know that, but we can't avoid ata_piix to be loaded
[08:19] <fabbione> that means breaking
[08:20] <fabbione> i did try removing it and restarting hotplug/udev stufd
[08:20] <Kamion> fabbione: mvo is testing
[08:20] <fabbione> the cdrom detection will hang since some scsi stuff is still loaded
[08:21] <thom> smurfix: how do i debug the keyboard chooser magic? it got my UK apple usb keyboard very wrong (wound up giving me a bunch of symbols, none of which i have on the keyboard)
[08:21] <fabbione> Kamion: if test is trying to remove one of the driver, i did already once removing the ata_piix and it doesn't work
[08:21] <mdz> jdub: please upload ubuntu-artwork with the dependency fix
[08:21] <jdub> mdz: ok, preparing as we speak
[08:21] <Kamion> fabbione: instead, we're trying removing piix
[08:22] <fabbione> Kamion: i can test that too
[08:22] <smurfix> thom: I need to know which keys you pressed, and their keycodes
[08:22] <Kamion> fabbione: like I say, bear with me
[08:22] <Kamion> fabbione: please do
[08:22] <smurfix> thom: (should add some logging for that)
[08:22] <mvo> Kamion: removed piix this time. but it does not work
[08:22] <fabbione> Kamion: already doing
[08:22] <Kamion> mvo: hm, ok
[08:22] <Kamion> let me do a quick audit of pcimap entries
[08:23] <lamont_r> smurfix: I magine a report of "I couldn't identify your keyboard.  The magic answers are: ..., ..., ..., ... - please file a bug", eh?
[08:23] <Kamion> mvo: you definitely booted with init=/bin/sh and removed it before doing anything else? could you check with lsmod to make sure it hasn't reappeared?
[08:24] <mdz> seb128: there are no more translation updates, right?
[08:24] <mdz> seb128: I am rolling new langpacks by hand
[08:24] <lamont_r> 200kbytes/sec seems so slow some days
[08:24] <mvo> Kamion: just did that (lsmod). I only have ata_piix loaded, no other piix driver
[08:24] <fabbione> Kamion: no it doesn't work
[08:24] <fabbione> Kamion: the cdrom still has I/O corruption
[08:25] <smurfix> lamont_r: Yeah, something like that.
[08:25] <fabbione> anyway the disk I/O looks ok
[08:25] <thom> smurfix: pressing ` got "keycode 86 was not expected"
[08:25] <fabbione> badblocks has been running for over an hour now
[08:26] <fabbione> so what is the decision?
[08:26] <smurfix> thom: That's the last step -- I need to log the keycodes before that.
[08:26] <lamont_r> if we revert the PATA change, what do we break again?
[08:26] <fabbione> 1440
[08:26] <jbailey> and the timeline.
[08:26] <T-Bone> at least it's a "safe" break
[08:26] <smurfix> lamont_r: On second thought that's not so easy because I also need to know which of the possible keys the user meant to press. :-/
[08:26] <mdz> the timeline is shot
[08:26] <mdz> we are releasing preview tomorrow
[08:26] <T-Bone> yeah, that one is more troublesome (re timeline)
[08:26] <fabbione> mdz: ok
[08:26] <mdz> sabdfl has decided it for us
[08:26] <Kamion> awk -F' ' '{ print $2 "-" $3 " " $1 }' /lib/modules/2.6.10-4-amd64-generic/modules.pcimap | sort -u | uniq -t' ' -W1 -D | less
[08:27] <smurfix> lamont_r: and the mind reading driver isn't ready for the kernel yet
[08:27] <Kamion> there are a fair few duplicate ids in there; some harmless/intentional (e.g. mptbase/mptscsih)
[08:27] <thom> smurfix: that was only the second question i answered
[08:27] <jbailey> piix and ata_piix on their own seem to be fine - I've had system running each of these for a couple of weeks.
[08:27] <T-Bone> so we're going to release with a *big warning*?
[08:27] <smurfix> thom: That should make things easier.
[08:27] <mdz> T-Bone: no, we're going to delay the release
[08:27] <thom> smurfix: anyway, how should i go about getting any more info
[08:27] <fabbione> mdz: should we roll out another kernel on the fly?
[08:27] <smurfix> thom: Can you tell me the symbol and keycode of the first key you pressed?
[08:27] <mdz> fabbione: as soon as we decide which fix to use
[08:27] <Kamion> now that the timeline's shot, I think I agree that we should revert the #1440 fix
[08:27] <fabbione> is there anybody other than me able to reproduce the problem?
[08:28] <T-Bone> mdz: err?? I must have missed something then
[08:28] <Kamion> fabbione: several
[08:28] <mdz> fabbione: we can either revert the patch from 1440, or back out only the PIIX changes
[08:28] <mdz> T-Bone: you missed me announcing that at least twice so far
[08:28] <Kamion> T-Bone: scroll up
[08:28] <mvo> fabbione: me
[08:28] <Kamion> 19:26 < mdz> the timeline is shot
[08:28] <Kamion> 19:26 < mdz> we are releasing preview tomorrow
[08:28] <Kamion> 19:26 < mdz> sabdfl has decided it for us
[08:28] <Kamion> like two minutes ago
[08:28] <fabbione> mdz: the patch changes 2 things. ATA_ENABLE_ATAPI and ATA_ENABLE_PATA
[08:28] <smurfix> thom: switch to text console, run "showkey", press the key in question
[08:28] <thom> ah, cool
[08:28] <fabbione> mdz: we can roll out one kernel that reverts the PATA, NOW
[08:28] <Kamion> fabbione: are they independent?
[08:28] <T-Bone> err... I didn't understand. To me tomorrow is 4h away...
[08:28] <fabbione> mdz: and see if that is enough
[08:29] <fabbione> Kamion: apparently they are
[08:29] <fabbione> mdz: and if it is not we revert the other one
[08:29] <mdz> Kamion: does tomorrow 1800UTC work for you?
[08:29] <Kamion> mdz: as good as any
[08:29] <fabbione> so in the worst is 2 kernel uploads
[08:29] <fabbione> but we a precise and test "fix"
[08:30] <thom> smurfix: don't have showkeys
[08:30] <Kamion> fabbione: would it be possible to come up with a list of the extra module/PCI-id pairs created by each of those?
[08:30] <lamont_r> fabbione: so you're thinking upload -26 with just PATA turned back off, and see what that breaks/fixes?
[08:30] <mdz> is there a length limit on the topic or something?
[08:30] <Kamion> fabbione: I could then compare that against my list of duplicated PCI ids
[08:30] <Kamion> mdz: there generally is
[08:30] <lamont_r> mdz: usually, and freenode bumps it up to something larger,but finite
[08:30] <fabbione> Kamion: for the piix there are only the 4 i listed before
[08:30] <Kamion> fabbione: kernel-wide
[08:30] <mdz> apologies for the noise
[08:31] <fabbione> lamont_r: let see what mdz suggest
[08:31] <svenl> hi all.
[08:31] <mdz> fabbione: let me look at the code a moment
[08:31] <fabbione> Kamion: i am not sure how
[08:31] <fabbione> mdz: sure
[08:31] <Kamion> fabbione: I may be able to try
[08:31] <smurfix> thom: console-tools: /usr/bin/showkey
[08:31] <T-Bone> mdz:  TOPICLEN=450
[08:31] <fabbione> Kamion: if you know a way, go for it
[08:31] <T-Bone> (from login messages)
[08:31] <Kamion> fabbione: *shrug* doesn't have to be a clean way :-)
[08:31] <svenl> Kamion: i got yaboot from netboot working on pegasos, and i guess that from cd will be ok too, need to try.
[08:32] <Kamion> svenl: ok, sorry but we're in the middle of a release panic and I haven't had any time at all to look
[08:32] <seb128> mdz: nop, translations are fine
[08:32] <fabbione> Kamion: yes.. i understand that :-) but right now is like 15 hours that i am around.. and not really a 100% clear mind to think even to extra hacks
[08:32] <Kamion> svenl: but that's very cool :-)
[08:32] <Kamion> fabbione: you and me both dude
[08:32] <svenl> Kamion: no problem.
[08:32] <fabbione> Kamion: ok.. let's try to think...
[08:32] <svenl> Kamion: release panic -> that would be hoary release ? 
[08:33] <Kamion> svenl: if that's something we can introduce cleanly post-preview, I'd like to try
[08:33] <fabbione> Kamion: what we need exactly is the pci ids listed in the kernel?
[08:33] <Kamion> svenl: hoary preview release
[08:33] <svenl> Kamion: well, it is an OF upgrade, will need no change to yaboot itself.
[08:33] <Kamion> fabbione: right. I'm grepping now
[08:33] <Kamion> svenl: rock
[08:33] <fabbione> Kamion: we need to grep for: static struct pci_device_id piix_pci_tbl[]  = {
[08:33] <svenl> Kamion: xresprobe (#7144) and yaboot-installer will need fixing then.
[08:34] <fabbione> pci_device_id is the struct that keep all the pci ids info
[08:34] <Kamion> fabbione: no, I mean grepping for ATA_ENABLE_PATA, try it it's much quicker :)
[08:34] <mdz> fabbione: ATA_ENABLE_PATA was not mentioned in the bug report; I did not even realize we had enabled that
[08:34] <Kamion> fabbione: there are like five mentions of that, in two drivers
[08:34] <svenl> mmm, netboot tries to detect keys :/
[08:34] <thom> smurfix: 'q' was the first character, 0x10 0x90
[08:34] <fabbione> Kamion: i think a simple process of that thing can give us all the pci ids
[08:34] <Kamion> piix and pata_pdc2027x
[08:34] <fabbione> Kamion: yes.. i know
[08:34] <Kamion> sorry, ata_piix and ...
[08:35] <fabbione> mdz: it has been enabled in the same patch and i am sorry because i have no idea why
[08:35] <Kamion> all those pata_pdc2027x ids show up as duplicates
[08:35] <Kamion> so I say definitely revert ATA_ENABLE_PATA
[08:35] <mdz> zul should know, I think he rolled the patch
[08:35] <fabbione> mdz: and i think reverting the PATA is enough
[08:35] <zul> fabbione: i have a kernel with pata disabled if you want to test
[08:35] <fabbione> but i need someway to test it
[08:35] <fabbione> zul: i need that kernel into a cdrom installer to test
[08:35] <fabbione> or something like that
[08:36] <fabbione> simply installing the kernel is not enough
[08:36] <fabbione> mvo: it's a combination...
[08:36] <zul> frig..
[08:36] <jbailey> Right, because initrd-tools isn't that bright.
[08:36] <mvo> fabbione: ah, ok
[08:36] <mdz> fabbione: it seems to only affect the ata_piix module from what I see in grep, so it should be sufficient to test with a replacement ata_piix module
[08:36] <fabbione> jbailey: bingo
[08:36] <Kamion> fabbione: stick the udebs in debian-installer/build/localudebs/, build
[08:36] <thom> smurfix: then ` which is 0x56 0xd6 ; and that was the unrecognised one
[08:36] <Kamion> although that won't be quite right
[08:36] <Kamion> ... what mdz said
[08:36] <smurfix> thom: what other symbols are on that key on your keyboard?
[08:37] <zul> fabbione: couldnt we just copy the .ko?
[08:37] <Kamion> you probably need to depmod too
[08:37] <fabbione> Kamion: i don't think it works.. the kernel modules are still downloaded at whatever time
[08:37] <smurfix> thom: ... and what's the correct console keymap for it?
[08:37] <Kamion> fabbione: run in expert mode and then you have time to replace them
[08:37] <thom> smurfix: ` ~
[08:37] <fabbione> mdz: it should be enough to rever that module.. yes
[08:37] <Kamion> fabbione: and anyway ata_piix is very likely in the initrd and not downloaded at all
[08:37] <fabbione> ok.. gimme a few secs
[08:37] <Kamion> although it'll depend on your image
[08:37] <fabbione> Kamion: hmmmm
[08:38] <fabbione> Kamion: i have the same images as you do
[08:38] <mdz> zul: can you make your ata_piix.ko available for download?
[08:38] <fabbione> i can craft the one from netinstall
[08:38] <mdz> Kamion: yes, and depmod
[08:38] <zul> zul: that is what im doing right now
[08:38] <Kamion> svenl: yaboot-installer is easy enough, but I have no clue about #7144
[08:38] <fabbione> and see if i can read cdrom with an older ata_piix
[08:38] <Kamion> fabbione: I mean, cdrom or netboot?
[08:38] <Alessio> sorry, what do you think about ipodder in ubuntu?
[08:38] <Alessio> for podcasting
[08:38] <fabbione> Kamion: i can use both
[08:39] <fabbione> Kamion: which one do you suggest to craft?
[08:39] <fabbione> Kamion: netboot is the fastest one to reload imho
[08:39] <fabbione> without having to burn a cd
[08:39] <smurfix> thom: Ah, that's the additional key on the pc105 keyboard. *Sigh*. I'll have a look.
[08:39] <Kamion> fabbione: hm, ok, plan:
[08:39] <Kamion> fabbione: use netboot in expert mode; run through until after "retrieving installer components"; pause at the menu; wget new ata_piix.ko, drop into place; continue
[08:40] <Kamion> the installer will depmod for you
[08:40] <thom> smurfix: (it's a british english apple usb keyboard)
[08:40] <zul> fabbione: ata_pixx.ko without pata http://zulinux.homelinux.net/ubuntu/kernel
[08:40] <fabbione> Kamion: ok
[08:41] <smurfix> thom: Yeah, and there's no good console keymap for it, is there?
[08:41] <zul> fabbione: #define ATA_ENABLE_ATAPI  #undef ATA_ENABLE_PATA
[08:41] <Kamion> zul: that still has the dup PCI ids
[08:41] <Kamion> zul: checked with modinfo, it has 24db 25a2 7111
[08:41] <zul> Kamion: ok
[08:42] <fabbione> Kamion: while i boot can you kindly get ata_piix.ko from the morgue?
[08:42] <fabbione> just to be 100000% sure is the old one?
[08:42] <Kamion> fabbione: which version?
[08:42] <thom> smurfix: mac-usb-uk is fine, afaik
[08:42] <lamont_r> Kamion: I think -24 was pre-ata patch
[08:42] <fabbione> Kamion: linux-source-2.6.10 (2.6.10-24) hoary; urgency=low
[08:42] <zul> it was
[08:42] <fabbione> yes -24
[08:43] <lamont_r> fabbione: beate me
[08:43] <Kamion> mdz: ATA_ENABLE_ATAPI does not introduce any PCI id changes, as far as I can see
[08:43] <fabbione> Kamion: no it doesn't
[08:43] <fabbione> it's only the PATA that adds pci ids
[08:43] <mdz> zul: you are confident that ATA_ENABLE_ATAPI alone addresses #1440?
[08:44] <Kamion> fabbione: http://people.ubuntu.com/~cjwatson/ata_piix.ko
[08:45] <mdz> Kamion: that's from linux-image-2.6.10-4-386 2.6.10-24?
[08:45] <Kamion> yes, and confirmed that it does not have the offending PCI ids
[08:45] <fabbione> HMMM
[08:45] <fabbione> there is no ata_piix on the system right now
[08:45] <fabbione> so it means it will be downloaded via sata-modules
[08:45] <mdz> ?
[08:45] <Kamion> there won't be until after "retrieving installer components"
[08:45] <fabbione> mdz: sorry that's for Kamion
[08:46] <Kamion> sata-modules is not in the netboot initrd
[08:46] <Kamion> right, it will be downloaded
[08:46] <Kamion> but if you're in expert mode, you will have an opportunity to replace it
[08:46] <fabbione> but i can just tell d-i not to install it
[08:46] <Kamion> no don't
[08:46] <Kamion> do what I said earlier
[08:46] <zul> mdz: yes i looked through the mandrake patches and that was the only one there
[08:46] <Kamion> 19:39 < Kamion> fabbione: use netboot in expert mode; run through until after "retrieving installer components"; pause at the menu; wget new ata_piix.ko,
[08:46] <fabbione> ok
[08:46] <Kamion>                 drop into place; continue
[08:46] <mdz> zul: so they are enabling ATA_ENABLE_PATA?
[08:46] <zul> mdz: yes
[08:47] <mdz> interesting
[08:47] <jbailey> I woner if they do hotplug based autodetection then?
[08:47] <lamont_r> eta 10 min or so
[08:47] <Kamion> entirely possible they don't
[08:47] <Kamion> Mandrake are a kudzu shop aren't they?
[08:48] <zul> from mandrake: 
[08:48] <zul> -#undef ATA_ENABLE_ATAPI                /* define to enable ATAPI support */
[08:48] <zul> -#undef ATA_ENABLE_PATA         /* define to enable PATA support in some
[08:48] <zul> +#define ATA_ENABLE_ATAPI               /* define to enable ATAPI support */
[08:48] <zul> +#define ATA_ENABLE_PATA                /* define to enable PATA support in some                                 * low-level drivers */
[08:49] <Kamion> zul: if they don't use hotplug, they probably won't hit the bug
[08:49] <mvo> zul: trying to load your ata_piix with the install-cd gave me a "ata_pixx: disagrees about the version of struct_module"
[08:50] <zul> its based on -25
[08:50] <fabbione> can anybody paste me again the md5sum for Packages.gz on livecd dists/hoary/main/debian-installer/binary-i386 ?
[08:50] <Kamion> mvo: try http://people.ubuntu.com/~cjwatson/ata_piix.ko
[08:50] <fabbione> as well as Packages
[08:50] <Kamion> fabbione: one sec
[08:51] <Kamion> ba22054987588dc3847b52bdd794a54e  dists/hoary/main/debian-installer/binary-i386/Packages
[08:51] <mvo> Kamion: ok. it takes a bit, I need to put it on a usb-stick and copy it over
[08:51] <Kamion> 24323b5da4533d415d1821cb7a3ded24  dists/hoary/main/debian-installer/binary-i386/Packages.gz
[08:51] <Kamion> fabbione: ^--
[08:51] <fabbione> well 
[08:51] <mdz> fabbione: 24323b5da4533d415d1821cb7a3ded24  /mnt/dists/hoary/main/debian-installer/binary-i386/Packages.gz
[08:51] <maswan> jdub: :), a bit of increased traffic the last couple of hours: http://www.acc.umu.se/technical/statistics/ftp/monitordata/index.html.en
[08:51] <fabbione> let's turn off PATA bunch of wankers!
[08:52] <fabbione> and let's make this damn preview RELEASE!
[08:52] <zul> Kamion: they use hotplug
[08:52] <smurfix> thom: I *hate* these keymaps, they're incomplete as *hell*.
[08:52] <thom> smurfix: :(
[08:52] <mdz> fabbione: do we need to have the #1440 people re-test?
[08:52] <thom> smurfix: totally unsurprised
[08:52] <fabbione> mdz: yes
[08:52] <smurfix> thom: If you look at the keymap you'll find that that key is not assigned *anywhere*
[08:52] <fabbione> mdz: that fixes the I/O corruption on piix cdrom
[08:52] <zul> Kamion: probably not in the install though
[08:53] <thom> smurfix: joy :(
[08:53] <fabbione> mdz: in full hounestly i didn't have the time to read all of #1440
[08:53] <smurfix> thom: which means it gets its setting from the kernel ... or the keymap that's been installed before it. :-/
[08:53] <Kamion> let's roll a new kernel without ATA_ENABLE_PATA then, and we can roll new d-i based on it
[08:53] <fabbione> Kamion: ok
[08:53] <fabbione> lamont: ping?
[08:53] <Kamion> if mdz's ok with that plan
[08:54] <fabbione> mdz: ?
[08:54] <mdz> go for it
[08:54] <fabbione> ok
[08:54] <fabbione> it will 25.1
[08:54] <Kamion> we still have lots of time for a complete revert if need be
[08:54] <mdz> fabbione: why not 26?
[08:54] <fabbione> Kamion: exactly
[08:54] <mdz> oh, you have a 26 prepared already
[08:54] <smurfix> thom: I'll go and fix the immediate problem but a review of the whole mess will have to wait after the prerelease. :-/
[08:54] <fabbione> mdz: because i have it already in RCS with tons of other changes that we do NOT want now
[08:54] <smurfix> thom: can you email me your "dumpkeys" output please?
[08:55] <mvo> Kamion: is it still worthwhile that I test your ata_piix module with the new plan?
[08:56] <Kamion> mvo: yes
[08:56] <Kamion> mvo: the new plan will be a better test, but it's worth a shot
[08:57] <mvo> Kamion: ok, doing it now
[08:57] <ogra> smurfix: how about dumpkeys collection in hwdb-client in hoary+1 ? sounds ike you could need the data
[08:57] <thom> smurfix: sure
[08:57] <thom> smurfix: smurf@debian.org
[08:57] <smurfix> ogra: All I need to do is to preload the table with the standard kernel layout
[08:58] <smurfix> thom: sure
[08:58] <smurfix> ogra: and then check what's missing
[08:58] <ogra> smurfix, ah, so youre doing that on the fly anyway....
[08:58] <smurfix> ogra: should be a straightforward if annoying piece of programming
[08:59] <smurfix> ogra: yes and no -- I precompile the decision table
[08:59] <smurfix> ogra: The job takes too long to do it at install time.
[08:59] <ogra> ah
[09:01] <mvo> Kamion: with your ata_piix module it works here. both modules (piix, ata_piix are loaded, but ata_piix is not in use and piix seems to be in control 
[09:01] <thom> smurfix: http://people.ubuntu.com/~thom/dumpkeys-mac-usb-uk (sorry, easier than email right now)
[09:02] <mdz> mvo: great
[09:02] <mvo> mdz: I'm standing by here, if there is more I can do, just tell me
[09:03] <fabbione> stuff can wrong
[09:03] <fabbione> ops
[09:04] <smurfix> thom: Thanks.
[09:05] <smurfix> thom: keycode  86 = less             greater          bar
[09:06] <lamont> fabbione: pack
[09:06] <lamont> er, ackj
[09:06] <fabbione> lamont: ok.. i have almost done
[09:06] <fabbione> i am updating the patch now
[09:06] <smurfix> thom: ... which doesn't look like  ` ~ to me ...
[09:06] <smurfix> thom: please verify.
[09:06] <fabbione> lamont: 25.1 won't be under rcs.. not from me at least
[09:07] <fabbione> lamont: my wife is already yelling and screaming that i am still here
[09:08] <lamont> fabbione: you upload, I'll commit
[09:08] <fabbione> lamont: ok
[09:09] <schweeb> mdz, Kamion: booted into the livecd now, worked fine... cdrom's working and everything
[09:12] <fabbione> lamont, mdz: almost done...
[09:13] <mdz> schweeb: interesting; which piix module(s) were loaded for you?
[09:13] <fabbione> lamont: i also committed the last changes i had to pre26
[09:13] <mdz> schweeb: it may be that your particular PCI device is not affected
[09:13] <fabbione> so it only needs 25.1 merging
[09:14] <fabbione> just check the lspci -n and compare the IDs
[09:16] <schweeb> mdz: just piix
[09:16] <schweeb> then the rest of the regular ide_* drivers
[09:20] <lamont> fabbione: ok.  we'll see how well baz does with the merge-of-duplicates. :-)
[09:21] <fabbione> we MUST get rid of that patch madness
[09:21] <T-Bone> indeed
[09:22] <zul> 2.6.11 would be a good canidate to start
[09:22] <T-Bone> indeed :)
[09:23] <fabbione> zul: we will come out with 2.6.11 when we will have the time
[09:23] <zul> fabbione: true
[09:23] <fabbione> and we redo the packaging immediatly after hoary
[09:24] <fabbione> now .10 has max priority
[09:24] <T-Bone> lamont: burn some wood, it'll produce more heat :}
[09:24] <lamont> fabbione: speaking of which, eta to upload?
[09:24] <fabbione> lamont: uploading now
[09:24] <fabbione> Uploading via ftp linux-source-2.6.10_2.6.10-25.1.diff.gz: 
[09:24] <lamont> woot
[09:25] <fabbione> Successfully uploaded packages.
[09:25] <fabbione> have fun
[09:25] <lamont> thanks
[09:25] <fabbione> i lost katie for a few secs...
[09:25] <fabbione> too bad
[09:25] <fabbione> i will just wait the ACCEPT mails
[09:25] <fabbione> and i am think i will be off for today
[09:25] <zul> later fabbione 
[09:25] <fabbione> 6am -> 10pm is a long run
[09:26] <fabbione> zul: not yet...
[09:26] <fabbione> there it is
[09:26] <fabbione> linux-source-2.6.10_2.6.10-25.1_source.changes ACCEPTED
[09:26] <T-Bone> sweet
[09:26] <lamont> fabbione: and you beat cron.daily... woot
[09:27] <fabbione> ehehe
[09:27] <fabbione> ok.. mail to changes too
[09:27] <lamont> fabbione: that means I only have to wait about 10 minutes for source, instead of 40
[09:27] <fabbione> 6 minutes actually
[09:28] <fabbione> daily is at 33 ;)
[09:28] <fabbione> 5 now....
[09:28] <fabbione> lucky eh? :)
[09:28] <lamont> yeah, plus a couple for the stuff to propogate
[09:28] <fabbione> last hit on the new album and i am off
[09:28] <fabbione> i have too much adrenaline to sleep right away
[09:29] <schweeb> odd
[09:29] <fabbione> schweeb: what?
[09:29] <schweeb> in OO.o 1.1.3 CUPS printing appears not to be working
[09:29] <schweeb> works everywhere else
[09:29] <fabbione> Kamion: i think we should all get some good night sleep and meet tomorrow morning early to test together again
[09:30] <fabbione> Kamion: what about 6 UTC?
[09:30] <schweeb> and in the OO.o 2 preview
[09:30] <fabbione> i definetely cannot wait 2 hours for CD images this night
[09:30] <schweeb> the odd thing is the job gets queued and sent to the printer, but the receive light blinks a couple times and does nothing
[09:30] <fabbione> or my wife will ask for divorce :)
[09:31] <schweeb> lol
[09:31] <T-Bone> fabbione: this would be quite soon for that heh :)
[09:31] <T-Bone> s/soon/early/
[09:32] <lamont> fabbione: at this stage, many jurisdictions would permit an annulment... but you don't want to go there either
[09:32] <T-Bone> lol
[09:32] <fabbione> ehhehe
[09:32] <T-Bone> lamont: is it because you're living in the us that you see the legal matter before everything else? :^)
[09:32] <zul> fabbione: already some celebrities last a month
[09:33] <fabbione> lamont: see.. after the first 2 weeks we knew eachother she told me: "I hate italians and i don't like nerds..."
[09:33] <fabbione> lamont: that's why we got married ;)
[09:33] <zul> lol
[09:33] <lamont> lol
[09:33] <T-Bone> fabbione: rotfl
[09:33] <fabbione> lamont: so now i would like to show that even a nerd can have a life ;)
[09:33] <fabbione> T-Bone: at this speed your fortune will have 99% of my entries
[09:34] <T-Bone> fabbione: heh. jbailey is quite a candidate you know :)
[09:34] <T-Bone> i'll make sure to make it available somewhere anyway :)
[09:34] <fabbione> ok i am off
[09:34] <T-Bone> see ya
[09:35] <fabbione> lamont: can you make sure with Kamion that the new kernel is in any of the CD (live or install) by 5 UTC?
[09:35] <fabbione> lamont: i usually wake up at that time and i can start testing immediatly after
[09:35] <mdz> fabbione: good night and thanks
[09:35] <fabbione> mdz: no problem
[09:35] <mdz> fabbione: welcome back ;-)
[09:35] <fabbione> I WANT THIS PREVIEW OUT!
[09:36] <fabbione> mdz: ehehhe
[09:36] <lamont> night fabbione - np on the kernel being there.
[09:36] <fabbione> lamont: perfect
[09:36] <fabbione> cya in a few hours
[09:36] <fabbione> and get some rest while the kernel propagate
[09:36] <fabbione> it's pointless to watch the logs ;)
[09:41] <mvo> mdz: I have pitti at the phone, he asks what's wrong with the lang-packs
[09:42] <mdz> mvo: bug #164595
[09:42] <mdz> mvo: dpkg: error processing /var/cache/apt/archives/language-pack-en-base_20050308_all.deb (--unpack):
[09:42] <mdz>  trying to overwrite `/usr/share/locale-langpack/en_CA/LC_MESSAGES/gnome-panel-2.0.mo', which is also in package language-pack-en
[09:43] <mvo> mdz: pitti says he solved it with a pre-depends? does it not work?
[09:44] <mvo> mdz: he is at a place with not network, he'll see to it tomorrow morning. is there a deadline for him?
[09:48] <mdz> mvo: it does not work
[09:48] <mdz> mvo: the pre-depends needs to be versioned
[09:50] <mdz> mvo: it can be dealt with first thing in the morning
[09:50] <mvo> mdz: ok, he'll deal with it then
[09:58] <svenl> Kamion: 7144 is under control, i just need to find time, but there is this OF work, and the debian 2.6.11 kernels, and the 2.4.27 mess and so on.
[09:59] <mdz> jdub: what is the delay on the new ubuntu-artwork?
[09:59] <HiddenWolf> mdz: it's uploaded, isn't it?
[09:59] <svenl> 7144 is tzo-sided. one is simply to detect that a pegasos needs BusType "PCI", and add it to the config file, the second is to fix xresprobe, and io will provide a patch for it.
[10:00] <ogra> HiddenWolf: nope
[10:00] <mdz> HiddenWolf: no, it's missing a dependency
[10:00] <HiddenWolf> mdz: ah, that.
[10:06] <Kamion> schweeb: good, according to our guessed diagnosis your machine should not have been affected, so I'm glad that's how it really is
[10:06] <Kamion> fabbione: I can't be there by 6 UTC; Kirsten is ill and I need to do a certain amount of looking after her
[10:06] <mdz> I'll be here at 6 UTC
[10:07] <Kamion> I think others may have to take care of the d-i build and a new CD build
[10:07] <Kamion> I have disabled the CD cron jobs (apart from Kubuntu), so any builds will have to be manual
[10:07] <Kamion> anyway, I'll be around for a couple more hours
[10:08] <zul> later
[10:09] <Kamion> mdz: should we mail ubuntu-devel@ to explain the delay, and to say that preview freeze is still in effect?
[10:09] <Mithrandir> Kamion: I think that would be good.
[10:09] <Kamion> I can do that if you want
[10:09] <mdz> Kamion: yes, that's a good idea. I'll do it
[10:09] <Kamion> ok, thanks
[10:12] <lamont> the 7th grader will be a bit inconvenienced.
[10:13] <T-Bone> lol
[10:14] <lamont> hrm... I really have to do something about the 23 minute startup time in my mirror-archive script
[10:17] <daniels> svenl: the BusType PCI thing shouldn't be needed, though -- it's indicative of broken hardware
[10:18] <svenl> daniels: it is indicative of broken radeon drivers in the ubuntu packages. Xfree86 4.3 as used in debian had no such problem.
[10:18] <svenl> daniels: the pegasos hardware uses a NB which had no agp bus, so fakes a pci-x bus into being an agp one.
[10:19] <svenl> daniels: this is a feature, not broken hardware.
[10:19] <svenl> daniels: and the agp-express guys have a similar solution on x86 even.
[10:20] <svenl> daniels: so the sane thing would be to do like the debian X packages, if for whatever reason loading agpgart fails, fall back to pcigart transparently.
[10:24] <daniels> svenl: how is it a feature?  do you even use that agp bus?
[10:24] <daniels> svenl: i agree it should fall back to pcigart transparently, and my understanding was that that was happening
[10:24] <daniels> but setting up an agp gart was either hanging or reporting success
[10:24] <tritium> Hi.  What's the policy on new packages in universe regarding .menu and .desktop files?  Do they need both?  Only .desktop?
[10:25] <svenl> daniels: it is a feature because with a NB without agp bus, you can use agp cards, which are more easy to find.
[10:25] <svenl> daniels: there is thus no agpgart.
[10:25] <lamont> (which won't upload until after the preview...)
[10:26] <svenl> daniels: and the agp express do the same, they add pci-faked-as-agp slot on intel pci-express motherboard for compatibility reason, so you will see this case outside of the pegasos too.
[10:26] <svenl> daniels: it doesn't happen. neither michel's dri-trunk, not ubuntu's X.org have the transparent fallback magic in it.
[10:27] <daniels> svenl: i think it's a pathologically dumb case, but there you go -- bear in mind that i've never heard of this needing to be done anywhere else, so either no-one owns the intel pcie motherboards, or they did it properly
[10:27] <daniels> svenl: well, patches welcome.
[10:27] <svenl> daniels: it is a new thing for compatibility with older hardware.
[10:27] <daniels> yes, I know, I have a PCIE motherboards
[10:28] <svenl> daniels: xresprobe and yaboot support in OF have priority :)
[10:28] <daniels> s/.$//
[10:28] <svenl> daniels: but not on with this new AGP-express thingy. I believe it is from ECS or something.
[10:28] <daniels> yes, well that would be the reason why I don't have it
[10:28] <daniels> anyway, gotta go now; follow up in the bug report if you like
[10:28] <svenl> daniels: for hoary it will be more easy to simply detect a pegasos machine and but the bustype accordyingly though.
[10:29] <svenl> daniels: does this mean that you will do nothing about it unless i provide a patch ? Is your pegasos machine running ?
[10:38] <lamont> what was the magic incantation to turn off the archive copier?
[10:38] <T-Bone> kill -9 -1 ;)
[10:38] <lamont> archive-copier/copy=false
[10:39] <mdz> Kamion: at what point do you want to tackle the WinFOSS-on-livecd stuff?  I think we should start shipping it quite soon after preview
[10:39] <mdz> T-Bone: fuser -k ;-)
[10:39] <T-Bone> mdz: hehe ;)
[10:39] <lamont> it's just a tarball to unpack right before the mkisofs
[10:40] <lamont> fuser -k mdz :-)
[10:40] <T-Bone> LOL
[10:40] <mdz> lamont: did the last round of kubuntu cloops build OK?
[10:43] <Kamion> mdz: damn, I forgot about that
[10:43] <Kamion> mdz: soon after preview is good
[10:43] <tseng> mako: i started a draft on the wiki as UnsignedGpgKey, id appreciate your review at some point
[10:43] <zenwhen> in a half hour I will have all the preview packages
[10:43] <zenwhen> :D
[10:43] <tseng> mako: unless you are part of the release crunch.
[10:43] <lamont> mdz: the last [k] ubuntu builds of cloops worked on !ia64
[10:43] <lamont> that is our (full) report :-)
[10:44] <Kamion> tritium: nowadays probably just .desktop is fine
[10:44] <Kamion> tritium: although others might disagree here
[10:45] <Kamion> mdz: I'd actually meant to do that for preview
[10:45] <Kamion> think I can squeeze it in?
[10:45] <tritium> Kamion, thanks.  Okay, I did run it by ogra.  It is a new package, so I believe I'll still with .desktop.
[10:46] <tritium> s/still/stick
[10:49] <svenl> Kamion: yaboot on pmacs/cds, it has just a plain iso9660 on the cds, or some extra magic ?
[10:49] <Kamion> svenl: hybrid iso9660/hfs (same as Debian)
[10:50] <Kamion> svenl: I think pegasos OF is fine with it, although you might want to check
[10:50] <svenl> Kamion: because yaboot sees that there are no partitions on my cd, and then tries each filesystem it knows, but not iso9660.
[10:51] <Kamion> yaboot doesn't implement ISO9660 itself
[10:51] <svenl> ext2, reiserfs, xfs, and then OF fallback, but it adds a extra :00 to the of call, which confuses my OF :/
[10:51] <Kamion> I believe it defers to OF to do that
[10:51] <svenl> yep, but adds bogus arguments to the OF prom_open call :/
[10:52] <Kamion>      strncpy(buffer, dev_name, 768);
[10:52] <Kamion>      strcat(buffer, ":");
[10:52] <Kamion>      if (part) {
[10:52] <Kamion>           char pn[3] ;
[10:52] <Kamion>           sprintf(pn, "%02d", part->part_number);
[10:52] <Kamion>           strcat(buffer, pn);
[10:52] <Kamion>      }
[10:52] <Kamion> hmm
[10:52] <svenl> Kamion: also, do you know how the cds can boot on ibm chrps, if they don't have a yaboot.conf in a separate /etc ?
[10:52] <Kamion> svenl: I suspect they don't
[10:52] <svenl> Kamion: where is that ?
[10:52] <svenl> Kamion: :)))
[10:52] <Kamion> I should probably stick a symlink in or something
[10:53] <Kamion> svenl: yaboot/second/fs_of.c:of_open()
[10:53] <Kamion> svenl: probably the question to ask is how come part is non-NULL
[10:54] <svenl> Kamion: that i know :)
[10:54] <svenl> because partition.c partitions_lookup sets it to a iso9660 partition.
[10:54] <zenwhen> do any of you devs use reiserfs?
[10:55] <mdz> Kamion: if you want to squeeze in the winfoss stuff, sure
[10:55] <mdz> it doesn't get much safer
[10:55] <svenl> zenwhen: reiserfs is a data eater.
[10:55] <Kamion> I'm happy to fix yaboot in Ubuntu if it's clear the code change doesn't affect pmac
[10:56] <Kamion> can Pegasos OF be fixed not to mind the :00?
[10:56] <svenl> Kamion: can you build a yaboot with full debug and try it on a pmac from a CD ?
[10:56] <Kamion> well, "fixed", whatever
[10:56] <Kamion> svenl: sure, but not today; I've made a note
[10:56] <svenl> Kamion: well, it could, but who knows what else may be broken.
[10:57] <svenl> Kamion: i can hack OF until tomorrow 16h00 only.
[10:57] <Kamion> yes, and I have a release to sort out now
[10:57] <Kamion> I will try to do it tomorrow morning
[10:57] <svenl> Ok, cool.
[10:57] <Kamion> svenl: oh, do you still want that yaboot.conf?
[10:57] <svenl> I plan to see if i can fix it.
[10:57] <thom> smurfix: i suspect that the console keymap i have doesn't actually look much like the keyboard :/
[10:57] <svenl> Kamion: i think not.
[10:58] <svenl> Kamion: err, yes, i want it.
[10:58] <Kamion> I can get you one now if need be, I have a recent installation
[10:58] <svenl> but less urgently.
[10:58] <svenl> Kamion: ok, get it and mail it to me.
[10:58] <svenl> so i can play with it.
[10:59] <smurfix> thom: *sigh*. Can you write one that does? I'd do it but I don't have your keyboard. :-/
[10:59] <Kamion> mailed
[10:59] <Kamion> there's a whole bunch of entries at the end that are worked out from os-prober, you can ignore those
[10:59] <svenl> I guess this code you pasted above is just plain broken, the right fix would be to set the partition number to -1 in partitions.c, and then do a check for that there.
[10:59] <svenl> ok.
[11:00] <svenl> or maybe check for a partition of size 0, i don't think there are any valid partitions entries where the first partition has a size of 0.
[11:00] <svenl> oh well.
[11:03] <mvo> Kamion, mdz: when do you think there will be something to test for me (and my piix system)? is it worth to stay up? or should I do it as first thing in the morning?
[11:04] <Kamion> mvo: some of the kernel builds are done, so maybe a couple of hours if mdz is happy to do another d-i byhand (or if elmo is up)
[11:05] <thom> smurfix: i'll add it to my todo *sigh*
[11:05] <lamont> Kamion: is the install CD actually checking gpg sigs and such on Releases?
[11:05] <thom> i can byhand if necessary, but if elmo is around i'd prefer to defer
[11:05] <thom> to him
[11:05] <Kamion> lamont: CD doesn't bother, netboot does
[11:06] <Kamion> lamont: it does check that the Packages md5sums match what's in Release though
[11:06] <smurfix> thom: Sorry that I can't be somewhat more helpful. Of course you could package up the keyboard and UPS it to me ;-)
[11:06] <lamont> Kamion: Menu item 'apt-setup-udeb' failed...
[11:06] <mvo> if apt-cdrom add is used, it will also check the signature and only copy the gpg file if they match
[11:06] <lamont> what did I b0rk?
[11:06] <lamont> (rewritten Packages...)
[11:07] <Kamion> lamont: that's not d-i
[11:07] <thom> smurfix: heh
[11:07] <lamont> Configuriing same exited 1
[11:07] <Kamion> lamont: yeah, that whole thing kinda sucks
[11:07] <lamont> oh wait.
[11:07] <thom> smurfix: well, i do have two of them, but that seems a little excessive :-)
[11:07] <lamont> lack of net connectivity wouldn't do that, would it?
[11:07] <Kamion> lamont: I want to turn off *all* authentication on the CD somehow, if possible; it's too damn hard to customise it at the moment
[11:07] <mdz> elmo: are you around?
[11:08] <lamont> Kamion: built a new Release file with the right md5sums and such, I think
[11:08] <Kamion> mvo: will apt-cdrom add be happy with a missing Release.gpg?
[11:08] <Kamion> lamont: it'll be more the invalid Release.gpg sig, I'd expect
[11:08] <Kamion> well, that too, at least
[11:08] <mvo> Kamion: yes
[11:08] <thom> mdz: if he's not in the datacenter he has no laptop
[11:08] <Kamion> I killed some of the authentication on the CD for #5723
[11:08] <mdz> thom: is he one of those weirdos who doesn't have a real computer?
[11:08] <thom> mdz: unless he's arranged to steal one
[11:09] <Kamion> and the fact that I remembered that bug number without even looking is scary
[11:09] <mdz> only a laptop?
[11:09] <thom> mdz: naw, but he's in london rahter than leeds
[11:09] <mdz> ah
[11:10] <mvo> Kamion: indeed :)
[11:10] <lamont> Kamion: well, I built a new Release.gpg, but of course the key is mine, not the CD image key...
[11:10] <lamont> where do I replace/add the key?
[11:12] <Kamion> you might be able to glue it into ubuntu-keyring then
[11:12] <lamont> but I'd like to have it nicer...
[11:13] <lamont> that is, what package actually delivers /etc/apt/trusted.gpg?
[11:13] <mvo> lamont: you could just add it to the ubuntu-keyring package
[11:14] <mvo> lamont: /usr/share/keyrings/ubuntu-archive-keyring.gpg
[11:14] <lamont> mvo: to ubuntu-archive-keyring.gpg?
[11:14] <lamont> cook
[11:14] <lamont> cool, even
[11:15] <svenl> Kamion: if i mail you a small yaboot patch, can you try it tomorrow on your pmac ? 
[11:15] <lamont> good to know for the next time.
[11:15] <lamont> but I'm manually past the annoyance for this pass
[11:15] <Kamion> svenl: yeah
[11:15] <lamont> "Configuring base system" isn't lsb-ized
[11:16] <lamont> s/base/the base/
[11:16] <Kamion> "lsb-ized"?
[11:16] <svenl> Kamion: basically i test in the above function if the partition is 0 sized and the only one, which is the iso9660 configuration.
[11:16] <Kamion> that's not an init script, it's something that base-config says
[11:21] <sabdfl> hi guys
[11:21] <sabdfl> just read the scrollback
[11:22] <sabdfl> mdz: what times did you set for the release tomorrow?
[11:22] <zenwhen> preview is slick guys
[11:22] <zenwhen> :)
[11:23] <Kamion> zenwhen: no preview yet ...
[11:23] <zenwhen> huh
[11:23] <zenwhen> i thought I saw an announcement
[11:23] <zenwhen> was it premature?
[11:23] <Kamion> where?
[11:23] <zenwhen> a link
[11:23] <zenwhen> in this channel
[11:23] <zenwhen> that spoke as if the preview was done
[11:23] <Kamion> that was a draft, and had "Draft" in the URL
[11:23] <zenwhen> and linked to images
[11:23] <zenwhen> Oh
[11:24] <zenwhen> well whatever this is, is really slick
[11:24] <zenwhen> ;)
[11:24] <Kamion> real preview should be tomorrow, very small change from what you have though
[11:25] <Kamion> sabdfl: he put it in the topic: "preview release: 2005-03-10 1800UTC"
[11:25] <zenwhen> its nice to see gnome 2.10 on my desktop on release date instead of three weeks later when I was running slackware.
[11:26] <zenwhen> I switched to ubuntu because quick gnome 2.8 inclusion, and am impressed as hell to be running 2.10 on release day
[11:26] <zenwhen> of^
[11:26] <Kamion> sabdfl: bonus of the delay, though: releases.u.c mirrors should all have today's image, and the rsync to tomorrow's will be pretty small
[11:26] <Kamion> few tens of MB per CD maybe
[11:27] <sabdfl> Kamion: perfect!
[11:27] <Kamion> not that I'm suggesting we go through this hell every time ;)
[11:27] <Kamion> hmmm. should change the cdebconf password widget to display *s.
[11:27] <sabdfl> Kamion: everything on our release checklist sorted?
[11:28] <sabdfl> do we need to add anything to that list for final?
[11:28] <sabdfl> other than:
[11:28] <Kamion> where's the checklist again?
[11:28] <sabdfl>  - get final artwork in A WEEK BEFORE
[11:28] <jon1012> hello :)
[11:28] <sabdfl>  - get kernel sorted TWO WEEKS BEFORE :-)
[11:28] <T-Bone> lol
[11:28] <Kamion> er, hmm. should we remove "Development Branch" from lsb-release and stuff?
[11:28] <Kamion> and we should change Release
[11:29] <mako> tseng: i hacked on https://www.ubuntulinux.org/wiki/UnsignedGpgKey after dholbach pointed it out to me
[11:29] <Kamion> I guess "Development Branch" can stay until just before final
[11:29] <mako> tseng: i added a bunch
[11:30] <sabdfl> http://www.ubuntulinux.org/wiki/ReleaseChecklist
[11:30] <sabdfl> Kamion: ^
[11:30] <mdz> sabdfl: 1800 UTC is the target
[11:30] <sabdfl> so final build in by 12:00 UTC?
[11:31] <Kamion> perhaps we should s/Development Branch/Preview/g
[11:32] <Kamion> sabdfl: I think that's doable; need to get elmo to change dists/hoary/Release, ideally
[11:32] <Kamion> although "Ubuntu Hoary Unreleased" is not too bad
[11:32] <mdz> sabdfl: we can do the final build tonight in fact
[11:34] <mdz> new kernel build is up, at least on i386
[11:34] <sabdfl> ok
[11:34] <Kamion> still waiting for powerpc and (to a lesser extent) ia64
[11:35] <mdz> mvo: still here?
[11:35] <mvo> mdz: yes
[11:35] <mvo> mdz: but a bit sleepy :)
[11:35] <mdz> mvo: are you able to do a test run with the new kernel?
[11:35] <mvo> mdz: yes, what needs to be done?
[11:35] <sabdfl> let's let kamion, elmo get a good nights rest
[11:35] <sabdfl> there will be lots to do tomorrow
[11:35] <sabdfl> make the most of the delay
[11:36] <mdz> Kamion: which udebs will mvo need to swap in in order to test?
[11:36] <Kamion> mdz: sata-modules
[11:36] <mdz> oh, never mind, it'll be in the initrd, won' t it
[11:36] <mvo> mdz: I think so, it is with the current cd
[11:37] <Kamion> yes
[11:37] <mdz> we could do a daily d-i build for i386
[11:37] <Kamion> mdz: ok, I'll start kicking them off
[11:39] <Kamion> i386 and amd64 d-i builds kicked
[11:39] <jdub> morning
[11:40] <Liblit> Hey, jdub.
[11:42] <T-Bone> seems that Linus doesn't agree :)
[11:42] <T-Bone> which is way cool: maybe we'll have a rock solid linux kernel on ppc now :)
[11:42] <thom> T-Bone: he has a big power5 box now doesn't he?
[11:43] <T-Bone> yeah
[11:43] <dredg> T-Bone: feh, what does linux know? :)
[11:43] <dredg> er, linus
[11:43] <T-Bone> dredg: nothing much, He just "switched"
[11:43] <T-Bone> his main platform is no longer x86 :)
[11:43] <T-Bone> http://www.zdnet.com.au/news/0,39023165,39183867,00.htm
[11:44] <Kamion> linus has had a powerpc box for ages
[11:44] <Kamion> it's why powerpc went mainline
[11:44] <mdz> and still we have random SIGILLs
[11:44] <T-Bone> Kamion: yeah, but making it his main machine is brand new
[11:44] <T-Bone> mdz: i hope that this will be fixed soon, if that's effectively a kernel bug. I bet he'll notice and get annoyed enough to fix it :)
[11:45] <whiprush> anyone know the right mirror contact email offhand? mirrors@canonical.com is bouncing and I'd like to fix the wiki entry.
[11:45] <Kamion> I've never seen those SIGILLs outside the context of a buildd
[11:45] <T-Bone> Kamion: yeah. That's a hint imho
[11:45] <Mithrandir> whiprush: uhm, just edit the wiki? :)
[11:45] <whiprush> I don't know what the right address to fix it to is though.
[11:46] <lamont> elmo/thom??
[11:46] <whiprush> Or should we just add mirrors and forget the email contact?
[11:46] <Kamion> T-Bone: probably just that it only occurs under fairly heavy load ...
[11:46] <thom> lamont?
[11:46] <mxpxpod> fabbione: good work on the linux-source-2.6.10 package :)
[11:46] <Kamion> mdz: i386 and amd64 d-i builds done
[11:46] <lamont> thom: on rookery, please: find ~lamont/public_html/Archives ! -user lamont -perm -0200 ! -perm -020 | xargs chmod g+w
[11:46] <lamont> and lart fabbione for me.
[11:47] <T-Bone> Kamion: buildd workload is quite specific. There are other things generating much more load
[11:47] <mdz> Kamion: shall I byhand them, or pull out an initrd for mvo?
[11:47] <Kamion> T-Bone: true
[11:47] <Kamion> mdz: byhanding would be good
[11:47] <T-Bone> Kamion: that sigill stuff could be toolchain related, i think
[11:47] <thom> done
[11:51] <Kamion> lamont: how far along are the powerpc and ia64 kernel builds?
[11:51] <ogra_live> hmm, piix works fine on my old lappie .... funny
[11:52] <lamont> linux-source-2.6.10_2.6.10-25.1_20050309-2037 04:04:41 (2 entries, sigma 00:03:10)
[11:53] <lamont> that's ppc.
[11:53] <mdz> Kamion: done, syncing
[11:53] <lamont> ia64, if it goes true-to-form, will just miss the kubuntu CD build, since it will upload right around 0415 or so
[11:53] <lamont> and ppc will be in the archive at about 0103 your time
[11:54] <lamont> ia64: linux-source-2.6.10_2.6.10-25.1_20050309-2037 07:29:45 (6 entries, sigma 00:09:07)
[11:54] <lamont> i386: linux-source-2.6.10_2.6.10-25.1_20050309-2040 02:25:58 (13 entries, sigma 01:01:29) (and builds the most kernels, plus arch: all)
[11:55] <lamont> amd64: linux-source-2.6.10_2.6.10-25.1_20050309-2037 01:23:13 (12 entries, sigma 00:40:43)
[11:55] <thom> ia64 takes 7 hours?!?
[11:55] <lamont> thom: it's a P-O-S box
[11:56] <lamont> thom: 1GHz CPU, 2GB RAM
[11:56] <lamont> and _SLOW_ disks
[11:56] <lamont> uniprocessor
[11:56] <thom> true
[11:56] <lamont> but it is itanium2
[11:56] <lamont> oh, and minimal cache
[11:57] <T-Bone> lamont: IDE?
[11:57] <ogra_live> jdub, is it intentional that gstreamer-properties is in the desktop settings ?
[11:58] <lamont> T-Bone: dev/hda is the CDROM, and there is /dev/sda*, so, no.
[11:58] <T-Bone> then why "slow"?
[11:58] <jdub> ogra_live: it will be out before final
[11:58] <ogra_live> ah, good
[11:58] <lamont> because it's special?
[11:58] <lamont> zx1600, iirc
[11:58] <T-Bone> lamont: i2 comes at least with 10k disks, neh?
[11:59] <lamont> T-Bone: 10k is _SLOW_
[11:59] <T-Bone> 1600? Does such a thing exist? :)
[11:59] <T-Bone> lamont: hell. I wonder how you describe the Xserve G5 SATA disks then :)
[11:59] <Kamion> lamont: could you please kick a d-i build on ross once the powerpc kernel reaches the archive, and ask mdz to byhand it?
[11:59] <lamont> T-Bone: take itanium 2 products from HP.  find the lowest price point.  that's the one in the data center
[11:59] <lamont> certainly
[11:59] <lamont> the others are done already?
[11:59] <T-Bone> lamont: can't be. IDE disks on i2 *are* slow
[11:59] <Kamion> i386 and amd64 are
[11:59] <Kamion> ia64, obviously, will have to wait
[12:00] <lamont> T-Bone: those disks are scsi
[12:00] <T-Bone> lamont: in the xserve?
[12:00] <Kamion> lamont: hmm. elmo didn't byhand this morning's d-i builds, at my request. Will that have confused things?