[12:13] <tepsipakki> I don't think that the UYVY-bug is important enough to basically update radeon_video.c from git-master
[12:18] <Mirv> nope, if it requires that. I'll reopen the bug in malone, btw.
[12:20] <tepsipakki> Mirv: yep, thanks
[12:20] <tepsipakki> soon we'll have 6.6.191 (or newer) anyway ;)
[12:27] <ogra> edubuntu amd64 server and addon are fine 
[12:30] <doko> ogra: don't need to test this one?
[12:34] <ogra> doko: CD
[12:57] <mdz> good evening
[12:59] <mvo> good evening mdz
[01:01] <mdz> mvo: how do things look?
[01:03] <mvo> mdz: good so far, I send you a mail earlier
[01:06] <mdz> mvo: the one to tollef, or a different one?
[01:08] <mvo> mdz: that one to tollef that I CCed you, he asked about the upgrade test status
[01:08] <mvo> ogra: LP is a bit slow today for me too 
[01:09] <ogra> mvo, 20min for one bug ?
[01:10] <mdz> mvo: right, ok
[01:37] <stgraber> Is anyone working on this bug : bug 105573 ?
[01:37] <ubotu> Malone bug 105573 in network-manager "gnome NetworkManager doesn't see any wired or wireless network" [Undecided,Unconfirmed]  https://launchpad.net/bugs/105573
[01:37] <stgraber> I just come accross it after an update ...
[01:38] <stgraber> hmm, just found where the problem come from, no need to check it :)
[01:39] <cjwatson> mdz: you know, I think that x-ttcidfont-conf bug may simply be that defoma-app is widdling all over stdout and confusing debconf
[01:53] <Arby> cjwatson: can you spare a moment to look at bug 99908.
[01:53] <ubotu> Malone bug 99908 in ubiquity "The ext3 file system creation in partition #1 of SCSI1 (0,0,0) (sda) failed." [Undecided,Needs info]  https://launchpad.net/bugs/99908
[01:53] <Arby> just happened to me with kubuntu 20070411 build
[01:54] <cjwatson> Arby: much of the problem seems to be that you're low on memory
[01:55] <Kmos> bug 105234
[01:55] <ubotu> Malone bug 105234 in network-manager "Netowrk manager says disconnected but is connected and working" [Undecided,Needs info]  https://launchpad.net/bugs/105234
[01:55] <Kmos> someone need to fix this
[01:55] <Kmos> 8 duplicates
[01:55] <Arby> cjwatson: fair enough as long as that's all it is.
[01:55] <cjwatson> after my earlier comment I realised my suggestion to edit commit.d/30parted was bogus, as update-dev is called by a later script
[01:55] <cjwatson> Arby: it's not entirely clear
[01:55] <cjwatson> but there are definitely signs in syslog of things failing due to out-of-memory, or processes being killed, or whatever
[01:56] <cjwatson> if that's what it is then the installer should still respond more gracefully
[01:56] <xtknight> how do i get gnome network manager to show up in my systray?
[01:56] <Arby> cjwatson: anything else I can do?
[01:58] <xtknight> ah it was  /usr/bin/nm-applet 
[01:58] <Arby> cjwatson: as long as your happy it's not a show stopper it can be dealt with $later
[01:58] <Arby> I just wanted to check
[01:58] <cjwatson> Arby: add more memory and see if it still happens? :-)
[01:59] <cjwatson> I think it probably isn't a showstopper, as I believe it's the only report I've received of this form
[01:59] <Arby> cjwatson: pass some over then :)
[01:59] <Arby> cjwatson: fair enough
[01:59] <cjwatson> though that doesn't mean I'm dismissing it
[01:59] <Arby> that's fine it just doens't need to be chased at this late stage
[02:00] <Arby> as long as I know I'm fine with that
[02:00] <Arby> if you want me to try anything in future just shout.
[02:01] <cjwatson> will do, thanks
[02:02] <cjwatson> mdz: can you reproduce that weird debconf-related hang at will? as in if you do the same install again does it still happen?
[02:03] <cjwatson> mdz: if so, I'd appreciate /var/log/syslog from an installation attempt with DEBCONF_DEBUG=developer set on the kernel command line
[02:08] <mjg59> cjwatson: Can I upload a new hotkey-setup? I'm not actually too fussed as to whether it's accepted for feisty (though it's approximately no-risk), but i'd rather have it sitting in a queue than depending on me remembering to upload it again when +1 is open
[02:08] <mjg59> (Or does the queue not work that way?)
[02:22] <Monk-e> Fujitsu, i think it's already named. But I don't know the name.
[02:39] <shawarma> Who are the soc mentors?
[02:39] <shawarma> They're wanted on slashnet..
[02:39] <shawarma> ...or anyone who can speak on their behalf.
[02:53] <BenC> anyone here experiencing bug #82314?
[02:53] <ubotu> Malone bug 82314 in linux-source-2.6.20 "pata driver in libata not mounting /home" [Critical,In progress]  https://launchpad.net/bugs/82314
[02:54] <BenC> basically Host Protected Area drives not being supported by libata-pata
[03:13] <xtknight> about bug 105234 and duplicates: i think i have a fix.  i'm not certain it's not a "hack", but brief testing shows that it works fine and even if it is slightly hacky i dont believe it has any other side effects.  should i post & have someone else review this?
[03:13] <ubotu> Malone bug 105234 in network-manager "Netowrk manager says disconnected but is connected and working" [Undecided,Confirmed]  https://launchpad.net/bugs/105234
[03:37] <xtknight> may anyone who knows about NetworkManager take a look at this patch (last comment)?  bug 105234
[03:37] <ubotu> Malone bug 105234 in network-manager "Netowrk manager says disconnected but is connected and working" [Undecided,Confirmed]  https://launchpad.net/bugs/105234
[03:39] <mjg59> xtknight: What timezone are you on?
[03:39] <mjg59> Most developers have hit bed, so tomorrow is almost certainly a better bet
[03:39] <xtknight> eastern US Wed Apr 11 9:39pm
[03:39] <xtknight> k
[03:40] <mjg59> Ask Mithrandir about it tomorrow - he's been looking at NM a little
[03:40] <mjg59> But he's pretty busy, so don't expect an immediate response
[04:49] <rpereira> Does someone know which part of Launchpad is GPL and which part it isn't?
[04:59] <mpt> rpereira, that question is better addressed to #launchpad
[05:00] <mpt> oh, nm, you're there already :-)
[05:00] <rpereira> mpt: yes, after I posted here, I used my brain.... :)
[06:24] <microphone_not_w> hi
[06:24] <Hobbsee> hi
[07:19] <Enola_Gay> hi all
[07:20] <Enola_Gay> Does anyone know when the RC will be released?
[07:23] <fabbione> when it's ready
[07:23] <Enola_Gay> :)
[07:23] <Enola_Gay> So it isn't?
[07:24] <fabbione> i haven't said that :)
[07:24] <Enola_Gay> Does the yesterday daily live build has any disadvantage? 
[07:25] <Treenaks> If you can install, you can always upgrade :)
[07:25] <Enola_Gay> So theoretically they are build like the rc or beta candidate?
[07:25] <Treenaks> afaik, all CD images are the same.. it's just that some images get to be blessed into 'Beta' images etc.
[07:26] <Enola_Gay> cool, thanks
[07:26] <Enola_Gay> Than the RC can be released when it is ready ;)
[07:26] <fabbione> yesterday's image might be RC..might be not
[07:26] <fabbione> but for sure they are pretty good
[07:27] <Enola_Gay> If only some packages differ it is no real problem. It just should have no disadvantages for upgrading.
[07:28] <pygi> I would like to congratulate everybody who got into SoC, so congrats folks ^_^
[07:28] <Enola_Gay> Thanks and cu all.
[08:05] <mdke> Mithrandir: I think bug 82335 reappeared yesterday. If not, another bug with the same symtoms :)
[08:05] <ubotu> Malone bug 82335 in network-manager "network-manager should not set offline mode when it manages no device" [High,Confirmed]  https://launchpad.net/bugs/82335
[08:06] <Mithrandir> mdke: no, that's intentional.  Apps think they're offline, but the NM doesn't manage any devices in that case, so it thinks it's offline.
[08:07] <mdke> Mithrandir: well, you originally fixed that bug... the fix was reverted intentionally?
[08:08] <mdke> all I know if that I have a big fat icon in my panel which tells me I have no network devices. That can't be a good thing
[08:08] <Mithrandir> mdke: does gaim and epiphany believe you are online or offline?
[08:08] <mdke> Mithrandir: no, that part is ok
[08:09] <mdke> (online)
[08:10] <Mithrandir> then the fix is working correctly.
[08:10] <mdke> it's just the icon which is wrong
[08:10] <Mithrandir> not really, it should just read "no managed devices" and not "no devices"
[08:11] <mdke> what's the point of that?
[08:11] <mdke> surely the whole point of the n-m icon is to allow people to quickly configure their network devices
[08:11] <mdke> I was really loving it for switching wifi networks
[08:11] <mdke> now I can't :(
[08:12] <Mithrandir> hm, you have a static, wired eth0 and an dhcp, wireless eth1?
[08:12] <Mithrandir> if so, it should list the wireless interface.
[08:13] <mdke> dhcp everything
[08:14] <mdke> even so, it seems that even people using static shouldn't get that red exclamation mark in their panel telling them something is wrong
[08:14] <mdke> s/seems/seems to me
[08:15] <mdke> oh, and the mouseover message is "No network connection"
[08:16] <Mithrandir> can you paste your /etc/network/interfaces file somewhere?
[08:17] <mdke> Mithrandir: ok. http://pastebin.ca/435901 - I've taken out all the white lines.
[08:17] <mdke> dunno what eth2, ath0 and wlan0 are - I don't have any of those
[08:17] <Mithrandir> "wireless-essid belkin54g" is a kind of static configuration and NM won't touch it because of that.
[08:18] <Mithrandir> but it should list eth0 in the list, but it might be greyed out if you don't have a cable plugged in
[08:18] <Hobbsee> how's the testing going?  kubuntu, specifically?
[08:19] <mdke> Mithrandir: that's fine. My point is, n-m is wrong to tell people they don't have a network connection, and the whole point of that icon is that i provides them with a good indication of whether their network is up or not.
[08:19] <mdke> until yesterday, I mean
[08:20] <Mithrandir> mdke: except it doesn't when you have weird setups and those kind of weird setups were more common than I thought
[08:21] <mdke> Mithrandir: was there something wrong with the way it was working since you fixed the bug and until the last upgrade? because loads of people commented on the bug report saying that their weird setups were being properly recognised
[08:22] <mdke> certainly mine started working again
[08:23] <Mithrandir> it broke with ipv6, for instance.  And it broke with mappings, and it broke LTSP and other cases where NM thought it could handle it, but couldn't.
[08:23] <mdke> I see
[08:23] <mdke> thanks for explaining that 
[08:24] <mdke> so, the icon should be fixed
[08:24] <Mithrandir> so I would have loved not to have gone back on this, but I think a small retreat is better than having something that breaks for lots of people.
[08:25] <bimberi> bug 105234 might be relevant?  In particular comment 32 (xtknight's tentatively submitted patch)
[08:25] <ubotu> Malone bug 105234 in network-manager "Netowrk manager says disconnected but is connected and working" [Medium,Confirmed]  https://launchpad.net/bugs/105234
[08:26] <sid> I purged linux-restricted-modules-common; and I can't apt-get install it back. There is no installation candidate.
[08:26] <mdke> Mithrandir: yep. We just need to make sure that people who's interfaces n-m can't handle aren't seeing the icon; maybe even show them the old Gnome icon that works properly
[08:28] <Mithrandir> mdke: but you aren't seeing your wired interface at all in the dropdown list?
[08:28] <mdke> Mithrandir: no. It says "no network devices have been found" and "no network connection" in the mouseover. (my eth1 is up)
[08:29] <mdke> the Gnome icon tells me I'm up and running
[08:29] <Mithrandir> hm
[08:30] <Mithrandir> that's weird.  I'll see if I can reproduce this on my laptop.
[08:30] <mdke> thanks. It's not so much my system that bothers me as in general though
[08:30] <Mithrandir> something else, ubuntu-docs still has the codename (feisty fawn) in the firefox home page, can we get rid of that and just refer to it as 7.04?
[08:30] <mdke> yes, if you wish
[08:31] <hile> nm does not often understand the status of links it has not itself brought up or down 
[08:31] <Mithrandir> hile: which is why it's told to stay away from them.
[08:31] <torkel> eth0 and eth1 are not marked "auto". n-m requires that for them to show up, no?
[08:31] <bimberi> Gutsy Gibbon!
[08:32] <mdke> hile: that's fine. The problem is that it's giving a misleading impression to the users, and that the icon in the panel should tell *all* users whether they are up or down, as the Gnome icon does
[08:32] <hile> of course, but I think only bug here is that nm still shows it's icon while there are no interfaces it's supposed to manage?
[08:32] <hile> ah, not that, I got it
[08:32] <Mithrandir> hile: that would be an acceptable way to fix the bug, IMO.  If there are nothing to manage, it can just hide itself.
[08:33] <hile> yeah, as long as there is no menu entry in it 'look for network interfaces now'
[08:33] <mdke> Mithrandir: it's a fix. But it's a regression from Edgy, which had a working icon for all users
[08:33] <hile> depends what you mean by working, I haven't had working wireless with default gnome app for long time :) 
[08:34] <mdke> and given that the default homepage is offline, there's no easy way for people to find out if they are online or not, short of typing a website address
[08:34] <hile> ... because I have used nm only for over a year
[08:34] <Mithrandir> mdke: they're free to add the old applet if that works better with their setup.
[08:35] <hile> nm just isn't well enough integrated with debian-based systems, I think
[08:35] <mdke> Mithrandir: if they can (a) guess it exists, and (b) find it
[08:35] <Mithrandir> mdke: can you make sure the "About Ubuntu" page just has references to the version number too?
[08:35] <Mithrandir> hile: but it'll never get there if we don't work on fixing that.
[08:35] <mdke> Mithrandir: it's not really a great time to do that - we'll break the translations. It's always contained the version number followed by the codename, afaik
[08:36] <hile> of course
[08:36] <Mithrandir> mdke: hmm
[08:36] <mdke> we could sed through all the translations too, I guess
[08:36] <Mithrandir> mdke: https://wiki.ubuntu.com/CodeNamesToVersionNumbers claims otherwise.
[08:36] <mdke> Mithrandir: as a result of that, we put the version numbers first and the codename as secondary
[08:37] <mdke> but I suppose we should remove them, it's just very late
[08:37] <Mithrandir> yes, I should have noticed earlier, but didn't since I overlooked a checklist on my checklist.
[08:38] <mdke> heh
[08:38] <mdke> I can't take care of it today, I need to get to work asap
[08:38] <Mithrandir> ok, see you.
[08:38] <mdke> maybe someone else can do it
[08:41] <sid> mdke: See if you can reproduce my bug. I didn't see it listed on launchpad.
[08:49] <mdke> sid: I don't have time I'm afraid.
[08:49] <mdke> Mithrandir: thanks for your time
[08:50] <pitti> Good morning
[08:51] <pygi> morning pitti 
[08:51] <pitti> Mithrandir: ah, it just occured to me how this lrm-video bug broke X yesterday: nvidia_new gets modprobed automatically, and thus X.org cannot load nvidia or nvidia_legacy any more, and you get that ABI mismatch
[08:51] <pitti> hi pygi
[09:00] <dholbach> good morning
[09:04] <pitti> argh, metacity hid my other xchat window again, brb
[09:04] <pitti> hey dholbach!
[09:24] <doko> pitti, lifeless: please comment on bug 105764
[09:24] <ubotu> Malone bug 105764 in python2.5 "apport exception hook negatively impacts startup of python" [Medium,Confirmed]  https://launchpad.net/bugs/105764
[09:26] <mneptok> @lart 37 Mithrandir 
[09:26] <mneptok> ukeoadi
[09:29] <Mithrandir> Hobbsee: both tickling and stomping.  Now, that's cheating.
[09:29] <Hobbsee> Mithrandir: why?
[09:29] <highvoltage> unfair advantage.
[09:29] <pygi> Hobbsee, you know that won't work with me =)
[09:29] <pygi> Hobbsee, I've got burning powers!
[09:29] <Hobbsee> pygi: nto when you're being prodded inside...
[09:29] <pygi> Hobbsee, won't help ya!
[09:29] <pitti> hi Mithrandir 
[09:30] <Hobbsee> heya pitti 
[09:30] <pygi> Hobbsee, I know all of your tricks. :)
[09:31] <zyga> good morning everyone
[09:32] <pygi> Hobbsee, I won't be able to fix your cd-recording mess for kubuntu :P
[09:32] <pygi> Mithrandir, thank you ^_^
[09:33] <Hobbsee> pygi: oh dera.
[09:33] <Hobbsee> pygi: well this is why you shouldnt eat people
[10:00] <pitti> Mithrandir: I did some tests and confirmed that http://paste.ubuntu-nl.org/15189/ really does what it should
[10:00] <pitti> Mithrandir: hmm, I just saw that BenC set the bug to fixcommitted already
[10:01] <pitti> Mithrandir: however, there's nothing in unaccepted; is this good to upload?
[10:01] <BenC> pitti: I just did an upload about an hour ago
[10:01] <Mithrandir> pitti: please.
[10:01] <pitti> BenC: oh, weird; I checked unaccepted before starting with this
[10:01] <BenC> but I never got an accepted email, so feel free to reupload
[10:02] <pitti> BenC: while testing this I noticed another oddity
[10:02] <pitti> BenC: lrm> ok, I assume it's more or less the same patch? (see paste)
[10:02] <BenC> yeah, same thing
[10:03] <pitti> BenC: regardless of whether I modprobe _legacy or _new, lsmod always shows 'nvidia'
[10:03] <pitti> BenC: this makes it hard to figure out which module is actually loaded (and makes r-m display the wrong thing)
[10:03] <pitti> any idea why it does that? can the module change its name?
[10:04] <BenC> pitti: odd, lsmod should always show the filename
[10:05] <pitti> BenC: hmm, so I better teach r-m about this special case
[10:06] <pitti> Mithrandir: ah, my l-r-m upload is in unapproved now
[10:09] <carlos> pitti: Is there any problem if we don't generate any language pack for two weeks after Feisty release?
[10:10] <pitti> carlos: no, that wouldn't hurt at all
[10:11] <carlos> pitti: that applies to Dapper and Edgy too
[10:11] <pitti> carlos: you want to open the gutsy translations in that time?
[10:11] <carlos> no, we are not going to do it so early
[10:11] <carlos> we need the server where the language packs are generated to handle the release load
[10:12] <carlos> so the server will be unavailable for the first two weeks
[10:14] <pitti> carlos: hm, but it would be good to open them early
[10:15] <pitti> carlos: feisty was a disaster
[10:15] <carlos> pitti: I got the comprise with our translators in a public meeting to open it two months after Feisty + 1 opening
[10:16] <Mithrandir> what is the reason for not just opening it when feisty+1 opens, apart from the server load?
[10:16] <carlos> and once Feisty + 1 schedule is published, I will ask to include there when we expect to have translations open
[10:16] <pitti> carlos: hm, I still don't understand why we cannot open it right after release; it will take a long time right after release or right before beta
[10:16] <carlos> not sure whether it's already published...
[10:16] <pitti> and right after release we don't need other tarballs so urgently
[10:18] <carlos> pitti: maybe for feisty + 2 we could open it again after release, but with Debian sync the amount of files to imports will be huge, and we have another critical bug that we should address that makes hard to non ubuntu projects to use Rosetta while a huge Ubuntu import is in process
[10:18] <pitti> carlos: right, but you have to import these files anyway at some time
[10:18] <carlos> pitti: also, translators asked us to delay it those two months until everything is a bit more stable with the version updates
[10:18] <carlos> pitti: well, those two months will give us time to deploy the solution
[10:18] <pitti> carlos: hm
[10:19] <pitti> carlos: if we can rely on the two months this time, I can live with that
[10:19] <Keybuk> carlos: I'm just publishing the schedule now
[10:19] <Keybuk> carlos: what date do you ned to know?
[10:20] <carlos> Keybuk: well, the idea is to add a 'Traslations open in Launchpad' two months after Feisty + 1 development starts, but I would like to see the calendar to agree on a date with kiko and other Rosetta developers
[10:20] <carlos> do you have a draft?
[10:21] <Keybuk> carlos: no, I have the final schedule :p
[10:21] <carlos> ok, URL?
[10:21] <ogra> ENOSEB :/
[10:21] <Keybuk> why do we have to wait for two months?
[10:21] <Keybuk> carlos: http://wiki.ubuntu.com/GutsyReleaseSchedule
[10:22] <Treenaks> ooh, release on my birthday :)
[10:22] <carlos> Keybuk: because translators asked to wait until all Debian sync is done and packages are moved from universe to main and the other way and new GNOME version is imported
[10:22] <Keybuk> June 21st for debian sync done
[10:23] <carlos> Keybuk: and for this concrete release, because, as explained to Martin, we need to fix a problem that blocks non unbuntu projects to import files while we handle the initial imports for a new distro
[10:24] <carlos> Keybuk: I guess then, we would open translations around that time
[10:24] <fabbione> Keybuk: wow and the sprint doesn't even overlap with my holidays :)
[10:24] <Keybuk> fabbione: this was deliberate :p
[10:24] <Keybuk> it doesn't overlap with anyone's holiday afaik
[10:24] <fabbione> Keybuk: eheheh
[10:24] <Keybuk> (or any conference)
[10:25] <fabbione> Keybuk: i *knew* you couldn't live without me ;)
[10:25] <carlos> Keybuk: I will try to have a meeting with kiko today to discuss the final date and will send you it so you can add it to the schedule, ok?
[10:25] <Keybuk> ok
[10:25] <carlos> thank you
[10:34] <siretart> Mithrandir: an news about the idea of removing broken binary packages (like unmet deps) from the (feisty) archive? I don't remember any feedback of this
[10:36] <Mithrandir> siretart: yes, we are not going to do it.
[10:38] <pitti> Mithrandir: I just committed a fix to bug 105812, see patch linked from the bug
[10:38] <ubotu> Malone bug 105812 in restricted-manager "nvidia_new and _legacy not detected as 'in use'" [High,Fix committed]  https://launchpad.net/bugs/105812
[10:38] <siretart> Mithrandir: ok. 
[10:39] <Mithrandir> pitti: ok, looks good.
[10:40] <pitti> Mithrandir: thanks, uploaded
[10:40] <doko> ogra: is the keyboard layout reset to english after installation of the ltsp chroot on edubuntu installs?
[10:42] <Seveas> Keybuk, gnome 2.12?!
[10:43] <Seveas> Keybuk, are we going back to warty? =)
[10:43] <tsmithe> damn Seveas i was about to say that!
[10:43] <Mez> Keybuk, may I query ?
[10:44] <Seveas> tsmithe, actually warty was 2.8, breezy was 2.12
[10:44] <tsmithe> i dunno :P
[10:44] <tsmithe> but i was gonna say something along those lines ;)
[10:45] <ogra> doko, hmm, is it right after first reboot ?
[10:45] <doko> ogra: no, noticed when entering the partition where to install grub
[10:45] <ogra> and after first boot ?
[10:45] <ogra> wrong or right ?
[10:45] <cjwatson> mjg59: yes, feel free to drop things into the queue
[10:46] <doko> ogra: wrong
[10:46] <ogra> hmm, i could use  in gdm here ... weird
[10:46] <doko> pitti: ok to upload the apport change?
[10:46] <pitti> doko: hm?
[10:47] <pranav_> doko: ping
[10:47] <doko> ogra: I'll recheck after the meeting, but I'm sure I set the language to german
[10:47] <ogra> doko, are you really sure ?
[10:47] <Keybuk> Mez: sure, what's up?
[10:47] <pitti> doko: sorry, I didn't check the bug yet, overlooked it over my reboots
[10:47] <doko> pranav_: I don't know anything about SoC, sorry. Google will make things public soon
[10:47] <Keybuk> Seveas: oops :p
[10:48] <pranav_> doko: he he it ok
[10:48] <ogra> i can imagine that the run of console-setup in the chroot affects the server through /proc but that shouldnt change the /etc/default values for the host
[10:48] <pranav_> *its 
[10:48] <ogra> so it *must* be the keymap thats set before chroot building starts after the first reboot ...
[10:49] <ogra> i wouldnt know a technical reason why it should stay non-german 
[10:49] <doko> ogra: checking agin
[10:49] <ogra> thanks
[10:50] <pitti> doko: erk, doesn't this require a new binary package, if you move the hook out of the apport/ module?
[10:50] <doko> pitti: why?
[10:52] <pitti> doko: also, apport_hook as a general module is really misleading, it should be python_apport_hook or so
[10:52] <pitti> doko: since we have two other kinds of 'apport hooks' already
[10:53] <heno> cjwatson: can you have a look at bug 105814 ?
[10:53] <ubotu> Malone bug 105814 in Ubuntu "fails to resize ntfs partition on virtualbox" [Undecided,Unconfirmed]  https://launchpad.net/bugs/105814
[10:53] <pitti> doko: if the Python policy allows a python-foo package to ship components which are not in python package foo, then that's ok for me
[10:54] <doko> pitti: the file is now called apport_hook; I don't care how its called
[10:54] <pitti> doko: otherwise I'd rather ship the file in apport itself
[10:54] <doko> pitti: not a policy issue at all
[10:54] <pitti> ok
[10:55] <dholbach> hey seb128
[10:55] <mvo> good morning seb128!
[10:55] <doko> seb128: you're late, slacker ;-P
[10:55] <seb128> hi dholbach, mvo, doko
[10:55] <pitti> doko: then let's call it apport_python_hook.py
[10:55] <doko> pitti: fine with me
[10:55] <dholbach> seb128: slap him :)
[10:55] <seb128> doko: I was around at 8am european time already, just not on IRC
[10:56] <seb128> cf my reply on the distro list :p
[10:56] <pitti> doko: did you change anything in the file itself? the diff makes it hard to tell
[10:56] <doko> dholbach: he would have to come to Berliiin, which will never happen :p
[10:56] <seb128> don't say never
[10:56] <seb128> we might have an UDS there one day ;)
[10:57] <doko> pitti: just prefixed the fileutils and report imports with apport.
[10:57] <pitti> doko: alright, I commit the change to the bzr with the new name, run some tests, and do the approval/upload jumping
[10:57] <pitti> doko: right
[10:57] <seb128> pitti: is that known that software-properties is not translated?
[10:57] <doko> pitti: ok, let me know, so that I can change the pythonX.Y packages
[10:57] <pitti> seb128: bug 105611, will fix right after RC
[10:57] <cjwatson> heno: see my comment
[10:57] <pitti> doko: will do
[10:58] <ubotu> Malone bug 105611 in langpack-o-matic "software-properties land in KDE pack" [High,In progress]  https://launchpad.net/bugs/105611
[10:58] <pitti> doko: thanks!
[10:58] <seb128> pitti: danke
[10:58] <seb128> launchpad is slow for other people as well?
[10:58] <fabbione> seb128: yes
[10:58] <seb128> it takes ages to load a page since yesterday :/
[10:59] <pitti> yes, sometimes; yesterday was horrible
[10:59] <ogra> seb128, my evo was evil to me today :/
[10:59] <seb128> ogra: like crashing?
[10:59] <ogra> seems my line dropped overnight ... it showed me the online icon but didnt want to show any folder contents
[11:00] <ogra> and told me i have to be online ...
[11:00] <heno> cjwatson: ok, cool. I'll give it some more testing anyway
[11:00] <ogra> clicking the icon didnt help ... restarting evo did
[11:03] <ogra> doko, oh, btw, so you ghhad the grub question as well if you noticed the breakage there ?
[11:04] <ogra> doko, it would probably help to comment on bug 105712
[11:04] <ubotu> Malone bug 105712 in os-prober "feisty grub osprober doesnt detect an existing feisty install on edubuntu alternate" [Undecided,Unconfirmed]  https://launchpad.net/bugs/105712
[11:05] <doko> ogra: let me check after the install
[11:05] <ogra> yup
[11:05] <ogra> do you have another OS on that machine ?
[11:06] <cjwatson> other partitions are automounted which confuses os-prober
[11:07] <cjwatson> this is a long-standing bug which I'm not concerned about for release
[11:07] <ogra> not in manual partitionning mode ....
[11:07] <cjwatson> yes they are
[11:07] <ogra> at kleast if i choose "dont use"
[11:07] <cjwatson> unless you manually deselect them
[11:07] <doko> ogra: yes, having Windows and an Ubuntu install; but all other partitions are not mounted
[11:07] <ogra> right, thats what i do in manual
[11:07] <cjwatson> in any case, not release-critical IMO so I'm not going to look at it further now. If you want to help, attach syslog and partman logs from the installer
[11:08] <ogra> no, not at all
[11:08] <ogra> but something to notice and log :)
[11:16] <Mithrandir> pitti: that r-m fix is needed too, right?
[11:16] <pitti> Mithrandir: not absolutely for RC; it does give false information, but the impact is much less bad than l-r-m
[11:17] <pitti> Mithrandir: OTOH it's probably much quicker to build than lrm :)
[11:17] <pitti> Mithrandir: for the majority of users it should just work as it is, since the majority uses nvidia-glx (no new/legacy)
[11:17] <Mithrandir> ok
[11:23] <saispo> how can i say on which ubuntu version i'm ?
[11:23] <doko> ogra: no, the keyboard problem persists
[11:23] <Nafallo> saispo: lsb_release -a
[11:24] <ogra> doko, weird, how can a setting in a chroot affect a host system ? and why didnt i see it in any installs here :/
[11:25] <fabbione> ogra: because you are probably sharing /proc and /dev ?
[11:25] <saispo> thanks Nafallo 
[11:25] <fabbione> ogra: and if you are root in the chroot you can kill processes in the host
[11:25] <doko> ogra: try the edubuntu DVD? with a german keyboard? I don't know what I can offer else for testing ...
[11:25] <fabbione> etc.
[11:26] <Nafallo> np
[11:26] <ogra> fabbione, sharing /proc will rewrite settings in /etc/default/console-setup ???
[11:26] <fabbione> ogra: hmm no it shouldn't
[11:26] <ogra> i can imagine that its a temporary switch as long as /proc is mounted
[11:26] <fabbione> but if it does.. and it's realiable.. tell pitti :)
[11:26] <ogra> but after the first reboot it should work as expected
[12:10] <carlos> pitti: yesterday we had a problem with the import queue so there are some files pending to be imported
[12:11] <pitti> carlos: alright; I won't update the packs today anyway
[12:11] <carlos> pitti: when will you prepare final packages for Feisty?
[12:11] <pitti> carlos: that's unclear yet, just discussing in the meeting
[12:11] <carlos> ok
[12:12] <carlos> pitti: once you know it, tell me so I can try to prioritise things so everything is imported 
[12:12] <pitti> carlos: most probably tomorrow
[12:13] <carlos> so I should get everything imported before midnight
[12:13] <carlos> to prepare the export tomorrow morning
[12:13] <pitti> and I need to fix langpack-o-matic for this software-properties bug
[12:13] <cjwatson> heno: do you believe that the red bug icons are a complete list of serious issues?
[12:13] <carlos> which bug?
[12:14] <pitti> carlos: bug 105611
[12:14] <ubotu> Malone bug 105611 in langpack-o-matic "software-properties land in KDE pack" [High,In progress]  https://launchpad.net/bugs/105611
[12:14] <carlos> oh, that one
[12:14] <carlos> right
[12:15] <heno> cjwatson: not sure the kernel issues are reflected there, but the other items raised in the meeting have red bugs
[12:15] <carlos> pitti: hmmm, are you sure it should be in GNOME ones?
[12:15] <pitti> carlos: yes
[12:15] <carlos> pitti: isn't it used too from KDE frontend?
[12:15] <cjwatson> bug 105631 we've decided to skip for this release, right?
[12:15] <ubotu> Malone bug 105631 in debian-installer "LVM install incorrectly sets root partition in lilo.conf" [Undecided,Confirmed]  https://launchpad.net/bugs/105631
[12:15] <pitti> carlos: oh, in the base ones, sorry
[12:16] <carlos> pitti: the bug says GNOME ones, that's why I'm asking
[12:16] <Riddell> pkl_: I don't see any test results from you for kubuntu alterntae amd64
[12:16] <cjwatson> I'm a bit worried by the lack of Kubuntu upgrade tests in that list
[12:16] <cjwatson> nothing else coverage-wise really concerns me
[12:17] <pkl_> riddell: no, I'll put them in.
[12:17] <heno> Riddell: I tried doing Kubuntu upgrades yesterday, but it was non-obvious, esp. with the CD
[12:17] <hunger> Will l-backports-modules-generic and kvm get fixed before the release?
[12:17] <heno> Riddell: can you point me at some URL
[12:18] <Riddell> heno: https://help.ubuntu.com/community/FeistyUpgrades "Kubuntu Beta upgrade"
[12:19] <heno> cjwatson, pkl_, BenC: btw, we had really no testing reports from the kernel team at all yesterday, which concerns me a bit. Though I know they were busy
[12:19] <pitti> Mithrandir, CC: doko: I'd appreciate your verdict about bug 105764
[12:19] <ubotu> Malone bug 105764 in python2.5 "apport exception hook negatively impacts startup of python" [Medium,Confirmed]  https://launchpad.net/bugs/105764
[12:19] <cjwatson> heno: I think the kernel team has been sufficiently busy that it would be a good plan to assign testing away from them
[12:20] <cjwatson> hunger: clarify?
[12:20] <hunger> cjwatson: Both packages are not installable for a couple of days now.
[12:20] <heno> cjwatson: yes I had a quick chat with Ben yesterday and did that, but it did put a dent in the coverage
[12:20] <cjwatson> hunger: architecture?
[12:21] <pitti> Mithrandir: I don't think that it is really necessary to push it into feisty at this point, but it was milestoned by someone
[12:21] <hunger> cjwatson: x86, really primitive.
[12:21] <hunger> cjwatson: AFAIK I did not install that t
[12:21] <cjwatson> hmm, linux-backports-modules-2.6.20 may need to be rebuilt for the current ABI
[12:21] <doko> pitti: I did milestone it; it's a severe performance regression over edgy
[12:21] <heno> I also need to rethink timing. Assigning DVDs to Europeans is a bit pointless as they are always delayed, so thumbs are twiddled
[12:21] <cjwatson> kvm is universe, no?
[12:21] <hunger> stuff myself, so it was dragged into the system at some point.
[12:22] <Mithrandir> pitti: hm, looks simple enough.  Has it been tested by others yet?
[12:22] <pitti> Mithrandir: doko and my testsuite
[12:22] <pitti> Mithrandir: (it has full test coverage)
[12:23] <cjwatson> I'll rebuild linux-backports-modules
[12:23] <pitti> Mithrandir: if we do this, then we need to accept all three of apport/python2.{4,5} together
[12:23] <Mithrandir> pitti: ok
[12:24] <pitti> Mithrandir: the python2.{4,5} diff is in my last comment
[12:25] <doko> preparing the python packages
[12:25] <Mithrandir> pitti: looks fine to me.
[12:26] <pitti> Mithrandir: thanks; I did a tiny fix to the test suite as well, btw, I hope that's ok (doesn't affect any code running on installations)
[12:26] <Mithrandir> pitti: yes
[12:29] <cjwatson> Mithrandir: please accept linux-backports-modules-2.6.20_2.6.20-14.9_source.changes
[12:30] <seb128> does anybody has a keyboard with multimedia key "play,pause", "play", "pause"?
[12:31] <dholbach> seb128: I have play/pause
[12:31] <seb128> and could tell me what codekey they are using
[12:31] <iwj> seb128: Yes.
[12:31] <iwj> Err, in X, or what ?
[12:31] <seb128> dholbach: could you look what value it has (you can use the gnome shortcut dialog and try assigning it to something)
[12:31] <iwj> If you mean in X, you mean xev ?
[12:32] <seb128> iwj: system, preferences, keyboard shortcut (or whatever it's called in english)
[12:32] <seb128> iwj: try to assign them to something
[12:32] <seb128> and note the code listed
[12:32] <Mithrandir> seb128: it varies between keyboards.
[12:32] <dholbach> seb128: 0xa2
[12:33] <seb128> Mithrandir: are you sure?
[12:33] <Nafallo> seb128: if not, I'm sure :-)
[12:33] <Mithrandir> seb128: fairly.
[12:33] <seb128> dholbach: ok, thank you
[12:33] <Nafallo> XF86PLAY should work? :-P
[12:33] <seb128> Nafallo: isn't acpi_fakekey handling them?
[12:34] <Nafallo> seb128: dunno. I just remember I had problems with that in some release cycle :-). the bug got rejected since I hadn't set my keyboard to what it was or something like that :-)
[12:35] <Nafallo> I made a patch for my keyboard to the file that specified what keys it used :-)
[12:35] <seb128> Nafallo: we use standard value for volume, eject, www, etc, seems to work correctly
[12:35] <seb128> we didn't get any bug this cycle about volume key not working I think
[12:36] <elmo> my laptop multimedia keys don't work anymore, FWIW, but then my laptop is apparently cursed
[12:36] <Nafallo> things might have improved. I think it was hoary or breezy I was talking about :-)
[12:36] <seb128> elmo: which one? none of them
[12:37] <elmo> seb128: none of them, play/pause, stop, ff, rew
[12:37] <elmo> err, except I'm  using xmms, not rhythmbox \o/
[12:37] <Keybuk> elmo: do you have the double-dhclient issue?
[12:37] <elmo> Keybuk: yes
[12:37] <seb128> elmo: those are no default value assigned
[12:37] <seb128> elmo: that's what I'm trying to change right now :p
[12:38] <Keybuk> elmo: can I borrow your /e/n/i
[12:38] <seb128> s/are/have
[12:38] <Tonio_> Mithrandir: just uploading digikam, closing an important issue in the latest release (bug/102912
[12:38] <Tonio_> Mithrandir: nothing critical for the release
[12:38] <Tonio_> Mithrandir: I hope you'll approve :)
[12:38] <Mithrandir> Tonio_: if it's not critical for the release, I'll reject it.
[12:39] <elmo> seb128: I could swear they use to, but maybe I'm insane
[12:39] <Tonio_> Mithrandir: sorry I meant that's important for the release, it breaks an important part of it
[12:39] <Fujitsu> Isn't there a new dbus API that multimedia keys need to be accessed by?
[12:39] <Tonio_> Mithrandir: I'm just tired :)
[12:39] <seb128> Fujitsu: no
[12:39] <Keybuk> elmo: have your hotkeys worked before?
[12:39] <seb128> Fujitsu: that's an implementation detail
[12:39] <elmo> Keybuk: yes
[12:39] <seb128> Fujitsu: they still need to be configured to the same capplet
[12:39] <Keybuk> elmo: when you push them, do you get invalid scancode type messages in syslog
[12:39] <iwj> seb128: Plau/Pause: 0xa2 (which it thinks is pause).  Stop: 0xa4.
[12:39] <Keybuk> (maybe try from console)
[12:40] <Tonio_> Mithrandir: the all image editor is broken in fact... which for digikam is really a big issue :)
[12:40] <iwj> seb128: I have no separate pause key.
[12:40] <Mithrandir> Tonio_: how many people have tested the patch?
[12:40] <seb128> iwj: ok, thank you. Do you have a pause different of the play/pause?
[12:40] <seb128> iwj: ok
[12:40] <Tonio_> Mithrandir: lots on kubuntu-devel, patch is fro msvn upstream
[12:40] <iwj> Also missing are skip back and forth which are 0x90 and 0x99.
[12:40] <elmo> Keybuk: nope
[12:40] <iwj> `Skip to {previous,next} track'
[12:40] <Tonio_> Mithrandir: I too have tested it, works perfectly
[12:41] <seb128> iwj: right, those are on my list and ok
[12:41] <Keybuk> elmo: in gnome's keyboard prefs, can you change an item, push one of those keys and have it set?
[12:41] <Tonio_> Mithrandir: Riddell is okay for the upload (I just asked)
[12:41] <iwj> seb128: Right.
[12:42] <Mithrandir> Tonio_: ok
[12:42] <Tonio_> Mithrandir: thanks
[12:42] <elmo> Keybuk: yes
[12:43] <Keybuk> elmo: iz gtk bug
[12:44] <seb128> elmo: are they list by the keyboard shortcut configuration dialog?
[12:44] <elmo> they're listed as 0x99 and similar
[12:44] <elmo> or do you mean something else?
[12:44] <seb128> elmo: no, just wondering if there was key associated to the actions
[12:44] <seb128> since we don't do it by default atm
[12:45] <seb128> with what app did you try them?
[12:45] <Mithrandir> Riddell: what's up with https://bugs.launchpad.net/ubuntu/+source/kdebase/+bug/61946 ?  It looks like this should have been milestoned ages ago?
[12:45] <elmo> seb128: xmms
[12:45] <ubotu> Malone bug 61946 in kdebase "[Edgy Data Loss]  umount progress dialog missing" [High,In progress]  
[12:45] <elmo> seb128: but it's all my hotkeys, including brightness, lock, sleep, etc.
[12:45] <seb128> iwj: ok, so your stop button is 0xa4 and BenC mail on the distro list suggest using 0xe8
[12:46] <seb128> elmo: hum, ok, dunno then
[12:46] <cjwatson> Mithrandir: err that was on the milestone list a while back? and a fix went in a couple of days ago
[12:46] <cjwatson> I assumed the uploader would take care of closing the bug
[12:47] <pygi> hi jsgotangco :)
[12:47] <Mithrandir> cjwatson: it was?  I'm fairly sure I haven't seen that before, but malone _still_ doesn't show milestoning in the activity log.
[12:47] <cjwatson> assuming that it was in fact a complete fix
[12:47] <cjwatson> yes, I am absolutely certain that was milestoned
[12:47] <Mithrandir> but if it's been fixed, that's good
[12:47] <cjwatson> https://lists.ubuntu.com/archives/feisty-changes/2007-April/008418.html
[12:47] <jsgotangco> hi pygi!
[12:48] <cjwatson> it sounds from the bug as though the uploader may not believe it's a complete fix
[12:48] <cjwatson> nice of them not to even mention the upload in the bug though :P
[12:48] <Mithrandir> indeed.
[12:48] <Mithrandir> I'll follow up on that
[12:48] <cjwatson> thanks
[12:49] <cjwatson> http://people.ubuntu.com/~ubuntu-archive/testing/feisty_probs.html is nothing like as empty as it should be
[12:49] <iwj> seb128: Hmm.  Dunno.
[12:49] <iwj> I've just double-checked it.
[12:49] <ogra> cjwatson, well it has ppc ...
[12:49] <seb128> iwj: thanks anyway, https://bugs.launchpad.net/ubuntu/+source/xkeyboard-config/+bug/37225 also lists 0xa4
[12:49] <iwj> The keyboard is a Packard Bell thing I got free with some hard disks (!)
[12:49] <cjwatson> it's nothing like as empty as it should be even aside from powerpc
[12:50] <ubotu> Malone bug 37225 in control-center "asus multimedia hotkeys not handled" [Medium,Confirmed]  
[12:50] <cjwatson> I'll move powerpc to ports there, but still
[12:50] <Keybuk> (someone remind me how to send Ctrl-Alt-F1 to vmware)
[12:50] <cjwatson> ctrl-alt-shift-f1?
[12:50] <ogra> cjwatson, also it seems not to be up to date ... schooltool was removed from main some days ago
[12:51] <Mithrandir> ogra: not the binaries
[12:51] <Keybuk> cjwatson: that didn't work, ctrl+alt-space-f1 worked
[12:52] <cjwatson> ogra: please always check facts in the archive before suspecting that the tools are broken
[12:52] <ogra> cjwatson, well, pitti demoted everything i was told by him ...
[12:52] <cjwatson> it is not the case that the tools can never be broken, but it's much less likely than that the archive state is simply not what you think it is
[12:53] <cjwatson> schooltool | 0.11.4-1ubuntu4 |        feisty | all
[12:53] <cjwatson> CHECK FACTS
[12:53] <cjwatson> seriously, it's not that hard to fire up a browser and look
[12:53] <cjwatson> I will demote the rest of school* now
[12:53] <ogra> no, but do i really need to check everything if my colleagues tell me something ?
[12:54] <pitti> hmm, I have 'change-override.py -c universe -S schooltool' in .bash_history
[12:54] <cjwatson> pitti: failed because the binaries are out of date
[12:54] <pitti> oh, I see
[12:54] <cjwatson> ogra: if you're then going to tell another colleague that his tools are broken, yes, you do
[12:55] <ogra> cjwatson, i didnt say anything is broken ... :)
[12:55] <ogra> outdated though ...
[12:55] <cjwatson> 11:50 <ogra> cjwatson, also it seems not to be up to date ... schooltool was removed from main some days ago
[12:55] <pitti> cjwatson: so I had to demote the binaries separately? (for next time)
[12:55] <cjwatson> outdated => broken for a tool that runs every hour
[12:56] <cjwatson> pitti: yeah, unfortunately
[12:56] <ogra> cjwatson, i didnt want to blame you or anyone ... i just noticed that the report didnt match my reality :)
[12:56] <cjwatson> ogra: the output even has a timestamp at the top of it
[12:56] <pitti> alright; sorry for the confusion
[12:56] <ogra> cjwatson, ok, thats my fault then, i'll look more closely next time 
[12:57] <cjwatson> ogra: I just really want your first response to a problem to be to check the facts, rather than to say that the thing that's outputting warnings must be wrong
[12:57] <cjwatson> that's all
[12:57] <cjwatson> pitti: no worries, fixed now
[12:57] <ogra> cjwatson, i'll try :) sorry again
[12:59] <cjwatson> ogra: powerpc is in testing-ports now; thanks for the report
[01:00] <ogra> thanks for fixing :)
[01:00] <pitti> cjwatson: ah, you already took care of the API bump for linux-backports-modules?
[01:01] <pitti> cjwatson: then we only need to remove the -13 cruft and anastacia/feisty_probs should be a bit happier
[01:01] <pitti> (I put that on my plate)
[01:01] <cjwatson> pitti: yeah, it's in unapproved waiting for Mithrandir to check it
[01:05] <Mithrandir> cjwatson: accepted
[01:07] <elmo> woo, metacity crashes for me on boot reproducably \o/
[01:08] <Riddell> Mithrandir: it wasn't milestoned because it's been an upstream bug for ages so I never expected anyone to come with a fix, but fdoving surprised me
[01:10] <elmo> pitti: does apport only record one crash per program/uid ?
[01:10] <pitti> elmo: yes, until you see it in the UI it won't be overridden
[01:10] <ogra> Riddell, for gutsy i'd like to talk to you about a qt based greeter for ldm ... to make the kde people more happy that use ltsp 
[01:11] <pitti> elmo: however, it counts the number of crashes per exe/uid
[01:11] <carlos> mvo: ping
[01:11] <ogra> Riddell, are you ok if i register a spec for that during next week and subscribe you to it ?
[01:11] <pitti> elmo: I needed to prevent flooding
[01:11] <cjwatson> As the discussion of an Ubuntu bug grows longer, the probability of an invocation of Bug #1 approaches one.
[01:11] <Riddell> ogra: I have an edubuntu-kde spec on the canonical wiki for such things
[01:12] <mvo> carlos: pong
[01:12] <ogra> Riddell, well, thats unrelated to edubuntu
[01:12] <ogra> ltsp != edubuntu ;)
[01:12] <carlos> mvo: I have bad news for you
[01:12] <Riddell> ogra: then the name of the spec should be changed
[01:12] <ogra> Riddell, but its ok for a start i guess
[01:13] <carlos> mvo: even increasing the amount of time before we think our poimport script is stalled, we get a timeout with your universe import
[01:13] <carlos> mvo: our parser takes more than 1 hour to process that file
[01:13] <ogra> Riddell, nope, we *should* have an edubuntu spec as well :) but this specific task should go under kubuntu-ltsp integration or so
[01:13] <carlos> mvo: so I had to block it
[01:13] <carlos> mvo: to allow all Feisty files to be imported in time before midnight
[01:14] <carlos> mvo: we are going to fix that performance problem as a critical bug
[01:14] <pitti> seb128: gnome-doc-tools in anastacia: demote or seed?
[01:14] <carlos> but that would mean that I don't think you will be able to update that template in a week or so
[01:15] <seb128> pitti: if nothing Build-Depends on it, demote
[01:15] <carlos> mvo: talking only about universe one
[01:15] <mvo> carlos: ok
[01:15] <carlos> which is huge, not sure about the .po files...
[01:15] <pitti> seb128: ladspa-sdk wants to go as well; in edgy, gst-plugins-good needed it; I assume that's alright? or was it inadvertedly disabled in feisty's gst?
[01:23] <pitti> Riddell: compiz-kde: seed or binary-only demotion to universe?
[01:23] <pitti> Riddell: IOW, does it work at all?
[01:24] <Riddell> pitti: no idea, never tried it :)
[01:25] <Riddell> pitti: as far as I know it's just a dummy package with only a stub in it
[01:25] <pitti> -rwxr-xr-x root/root    126288 2007-04-04 09:58:19 ./usr/bin/kde-window-decorator
[01:25] <pitti> Riddell: ok, so rather demote, I figure
[01:26] <Riddell> pitti: I wouldn't call it supported whatever it is
[01:29] <pitti> ogra: can you please seed student-control-panel? it's a transitional package, but as such it needs to be in main for feisty
[01:30] <heno> *** All 20070411 images have been rejected for RC. 20070412 images being built. See https://www.stgraber.org/ubuntu/isotesting/ for status and md5sums ***
[01:39] <jsgotangco> oucchhhh
[01:40] <pitti> ogra: wb
[01:40] <ogra> my broadcom doesnt like me today :/
[01:40] <pitti> ogra: not sure whether you got that still: <pitti> ogra: can you please seed student-control-panel? it's a transitional package, but as such it needs to be in main for feisty
[01:41] <ogra> pitti, ok, no problem 
[01:41] <pitti> ogra: thanks
[01:41] <jsgotangco> ogra: i'll just wait for this day's build to be done,
[01:41] <ogra> jsgotangco, well ... you could rsync on top of yesterdays ;)
[01:41] <fabbione> Mithrandir: what images have been rebuilt? i can see sparc/server is still the old one
[01:41] <jsgotangco> ogra: sure but you asked me to do DVD hehe
[01:42] <Mithrandir> fabbione: -server is being rebuilt right now.
[01:42] <fabbione> Mithrandir: ok thanks
[01:42] <ogra> jsgotangco, exactly :)
[01:43] <ogra> pitti, suported is sufficient i guess
[01:43] <pitti> ogra: sure
[01:43] <pitti> ogra: in the ubuntu seeds we have a paragraph for transitional packages
[01:43] <pitti> so that we can clean this up over time again
[01:43] <ogra> ah, yes i see it
[01:44] <pitti> Mithrandir: I uploaded a new language-support-gl with two more oo.o l10n/help packages which linger in anastacia (doesn't affect CDs)
[01:46] <heno> fabbione: I wasn't sure about that one, setting it inactive too
[01:47] <fabbione> heno: you also want to clear netinstall/netboot
[01:48] <heno> fabbione: I'll clear the sparc one. the other's never got tested anyway. Setting up to do that here now
[01:49] <fabbione> heno: ok
[01:50] <ogra> pitti, committed
[01:50] <pitti> cheers
[01:54] <ogra> doko, can you attach /etc/default/console-setup and /opt/ltsp/$arch/etc/default/console-setup to the keyboard bug ?
[01:54] <ogra> i'd like to compare them
[02:00] <doko> ogra: same files, and XKBLAYOUT is set to "de"
[02:02] <ogra> right
[02:02] <ogra> so console-setup should set it to de on boot ...
[02:02] <ogra> cjwatson, is there anything that can override /etc/default/console-setup ? i wouldnt think so ...
[02:03] <ogra> (in a default install that is)
[02:04] <cjwatson> the initramfs could be broken
[02:04] <ogra> hmm
[02:04] <cjwatson> check /etc/default/console-setup in the initramfs
[02:04] <ogra> but thas generated already if ltsp gets installed
[02:04] <cjwatson> if 'sudo setupcon' fixes the keyboard layout, then the initramfs is definitely broken
[02:04] <ogra> will do, doko, can you send that to me ? 
[02:05] <ogra> or /etc/init.d/console-setup start i suppose
[02:06] <desrt> gutsy?  seriously?
[02:06] <Hobbsee> desrt: appears so.
[02:07] <ogra> desrt, "all your source is eaten by us !"
[02:07] <fabbione> desrt: yeah
[02:07] <ant_ipop> for help testing rc i should go to https://www.stgraber.org/ubuntu/isotesting/ right ?
[02:07] <Hobbsee> desrt: unless it's 13 days late
[02:07] <fabbione> ant_ipop: yes
[02:07] <ogra> ant_ipop, yep
[02:08] <heno> ant_ipop: you can get advice in #ubuntu-iso and from https://wiki.ubuntu.com/Testing/Community
[02:09] <ant_ipop> ah thanks for that
[02:10] <desrt> eh.  what's in a name? :)
[02:10] <Hobbsee> desrt: many misspellings
[02:11] <Kmos> how works in opera browser updates?
[02:12] <Kmos> to assigne a bug report for the updated version
[02:12] <doko> ogra: no, the file in the initramfs is the same as well
[02:12] <desrt> ...or should i call you hosbee...
[02:13] <cjwatson> doko: does 'sudo setupcon' at a console fix the layout?
[02:13] <Hobbsee> desrt: wonder what jdub will call it
[02:13] <heno> doko, ogra: FYI the Edubuntu amd64 DVD installed fine here. Both virtualbox and real hardware
[02:13] <Hobbsee> desrt: no, you shouldnt.
[02:13] <doko> cjwatson: wait, have to reboot
[02:13] <desrt> i want jdub at UDS.  srsly.
[02:13] <ogra> heno, no such glitches like doko saw ? 
[02:13] <heno> ogra: no, on the default option it was fine. Will try safe graphics mode too
[02:14] <heno> ogra: can I force it to use VGA?
[02:14] <ogra> heno, any keyboard weirdness ? 
[02:14] <ogra> i start to suspect dokos iso or media is broken 
[02:14] <heno> ogra: didn't heck for that yet, will do
[02:14] <ogra> thanks
[02:17] <Hobbsee> desrt: yes, jdub is fun.
[02:18] <desrt> "islamophobia"
[02:18] <jsgotangco> crazy gnome madmen are always fun
[02:27] <doko> cjwatson, ogra: reboot has to wait until my OOo build finishes; about 1 1/2h
[02:29] <StevenK> doko: Another openoffice.org upload? Have you no respect for the buildds? :-P
[02:30] <ogra> doko, fine with me 
[02:30] <pitti> heno: do you know gnome-accessibility-themes-extras? it wants to go to universe; shall I seed it (if it's any good) or demote?
[02:30] <fabbione> desrt: you coming to Seville?
[02:31] <desrt> of course
[02:31] <pitti> desrt: cool
[02:31] <desrt> watch me forget the G5 again
[02:31] <fabbione> desrt: G..... 5....
[02:31] <fabbione> ahhaha
[02:31] <cjwatson> openoffice.org at this point in the release? What universe imploded? :)
[02:31] <StevenK> Heh
[02:31] <fabbione> cjwatson: he didn't pay the heating bill.. what's best than a CPU? :)
[02:32] <cjwatson> mdz_: bug 105861, fyi
[02:32] <StevenK> fabbione: Bwahaha.
[02:32] <ubotu> Malone bug 105861 in ubiquity "migration-assistant cleanup handler manages to fork somewhere" [High,Confirmed]  https://launchpad.net/bugs/105861
[02:32] <cjwatson> investigating
[02:32] <cjwatson> Mithrandir: ^--
[02:32] <Hobbsee> Keybuk: were you going to have a separate universe UVF again, at the same time as new package universe freeze?  (aug 30)?
[02:33] <mdz_> cjwatson: under what circumstances is it triggered (if you know)?
[02:36] <cjwatson> mdz_: not quite sure; it happened to me with a simple install on a system where m-a didn't offer anything to import
[02:37] <cjwatson> mdz_: it's quite possible that the forked process (well, actually it's probably a separate thread rather than a true fork, thanks pygtk) goes off and exits harmlessly and the only ill effect is the warning dialog (a relatively recent addition to fix other critical problems)
[02:38] <cjwatson> I'll let you know when I know more
[02:40] <cjwatson> critical problems> it used to be that if partman exited non-zero then ubiquity wouldn't notice and could end up trying to "install" into the ramdisk. I fixed that ...
[02:43] <heno> pitti: it will keep coming from upstream gnome each cycle I guess, so we split it out each time. universe is fine by me though
[02:44] <pitti> heno: unless its really crackful or has dependencies not in main, we can seed it (for consistently having all binaries in main)
[02:45] <heno> pitti: right, let's keep it then. It's just themes from upstream that we opted not to install by default, not very crackful
[02:45] <pitti> heno: alright, I seed it
[02:45] <pitti> thanks
[02:46] <ubotu> Malone bug 81227 in hal "Logout screen appears twice [Feisty] " [Medium,In progress]  https://launchpad.net/bugs/81227
[02:46] <ogra> dear HW vendors, please use unified names goddamned!
[02:49] <mjg59> Unified names?
[02:51] <ogra> mjg59, acpi_PWRF and acpi_PWRB .... and who knows what else will show up 
[02:52] <mjg59> There's no requirement for any sort of consistent naming
[02:52] <mjg59> Don't attempt to depend on it. You'll just lose.
[02:52] <ogra> there is ... from me ... but i doubt they will listen to me ;)
[02:52] <mjg59> HP use ACPI identifiers that are longer than four characters and then collapse them down into C+3 digits of hex
[02:52] <ogra> gah
[02:53] <mjg59> so the name will change depending on BIOS upgrade
[02:53] <ogra> yeah
[02:53] <mjg59> The spec doesn't require any object names to be meaningful
[02:53] <ogra> well, but it seems for the HP people on the bug my workaround fixed it
[02:53] <ogra> might be that they are the lucky ones here ...
[02:54] <ogra> i simply dont want to bloat that workaround ... 
[02:55] <mjg59> If you're trying to use ACPI object names, then that's not a workaround. That's a bug.
[02:56] <ogra> well, how els should i do it ? (the patch is on the bug in my last comment)
[02:57] <ogra> according to #hal the fix has to happen in hal itself which we cant do atm ... the workaround seems to fix it for a lot of people but not for all ...
[02:58] <ogra> yeah, its horribly slow atm ...
[02:58] <mjg59> Why not just key off all acpi objects that send the power button?
[02:58] <ogra> how do i match that ? 
[02:59] <mjg59> acpi_
[02:59] <ogra> yeah, thats what i planned now (see my comment)
[02:59] <ogra> seems i should cut he paternmatching after the underscore to match all acpi_**** devices ... :/
[03:00] <mjg59> The only ACPI devices that can generate power button events are, well, power buttons
[03:00] <ogra> mjg59, so you think thats sufficient for now ? 
[03:00] <mjg59> So yeah
[03:00] <ogra> ok, i'll change to strncmp then and curt the last four chars ...
[03:00] <mjg59> You're in a glib application
[03:00] <ogra> oh, right
[03:10] <fabbione> heno: are you keeping track of the bugs that have been cleared from 11 iso testing?
[03:11] <fabbione> heno: they have been wiped from the board too
[03:11] <fabbione> heno: or do we need to relink them again?
[03:27] <fabbione> who is our avahi expert?
[03:27] <fabbione> pitti: is that you?
[03:27] <pitti> fabbione: Lathiat mainly
[03:28] <fabbione> Lathiat: you around?
[03:28] <fabbione> https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/105872
[03:28] <fabbione> (hw-cert bug)
[03:28] <ubotu> Malone bug 105872 in network-manager "Network not enabled properly by NetworkManager which configures eth1:avah" [Undecided,Unconfirmed]  
[03:28] <TomaszD> pitti, thanks for fixing that software-properties translations misplacement bug! 
[03:29] <pitti> TomaszD: you're welcome; I still need to rebuild the langpacks for that
[03:30] <TomaszD> pitti, that'd be very welcome to have, new langpacks as official update, not just the daily ones
[03:33] <TomaszD> pitti, can we expect that sometime soon? anyway it's just a few days before the final release...
[03:33] <pitti> iwj: what is the reason for the mkdir -m 0755 /dev/mapper in libdevmapper1.02.postinst?
[03:33] <pitti> TomaszD: yes, probably tomorrow we'll see another feisty langpack upload
[03:33] <TomaszD> pitti, great :] 
[03:33] <pitti> TomaszD: 'directly after RC release'
[03:34] <TomaszD> after...? I'm too tired to argue, but I see a flaw in this plan :] 
[03:34] <pitti> TomaszD: this issue is by no way a showstopper for the RC
[03:34] <pitti> TomaszD: it should be fixed in the final release, of course, and it will
[03:35] <TomaszD> no, of course it's not a showstopper, I'm not even trying to argue that. I was just thinking that the last official langpack upload was committed quite some time ago and Ubuntu'd really use a new upload, and RC would be a great testbed for that
[03:36] <TomaszD> that's why I'm wondering why "after"
[03:36] <pitti> because we have RC images since yesterday
[03:36] <pitti> sure, we just got new ones, but still; no time to push in 1000 new packages
[03:36] <TomaszD> fair enough.
[03:36] <TomaszD> :] 
[03:36] <pitti> TomaszD: and you can still dist-upgrade your system after installation, to get the latest pacakges
[03:37] <TomaszD> yes, sure, I'm no newbie. Using Ubuntu since 5.04 
[03:37] <TomaszD> bbl
[03:37] <elmo> seb128: where's 0.16.1 ?
[03:37] <heno> ogra: I did a complete install in German from the amd64 Edubuntu DVD (20070411). Keyboard and everything else worked fine
[03:38] <ogra> good
[03:38] <ogra> on the CD as well, so i guess its doko specific ...
[03:39] <doko> ogra: rechecking with today's DVD later
[03:39] <seb128> elmo: that's a good question, "Successfully built" on https://launchpad.net/ubuntu/+source/vte/1:0.16.1-0ubuntu1 
[03:39] <seb128> elmo: iz launchpad bog
[03:39] <ogra> doko, thanks ... i think a media check would also make sense
[03:40] <doko> ogra: the check was sucessful
[03:40] <ogra> hmm, k
[03:40] <ogra> still weird 
[03:41] <Mithrandir> seb128: investigating.
[03:41] <seb128> Mithrandir: thank you
[03:54] <iwj> pitti: I'm not sure.  I don't think it was my doing.
[03:54] <iwj> It seems odd that it doesn't say -p.
[03:54] <iwj> If you have udev then something has to make /dev/mapper and I have a feeling that the C code in libdevmapper itself won't.
[03:55] <pitti> iwj: weird; a mere modprobe dm-crypt sorted that out for me so far
[03:55] <iwj> Is it causing a problem ?  The way you quote it makes it non-idempotent which is clearly wrong.
[03:55] <iwj> pitti: Ah, probably that causes /dev/mapper/control to be made by udev.
[03:55] <pitti> iwj: oh, it broke the retracer chroot; which is not really libdevmapper's fault, I just wondered what it was for
[03:56] <cjwatson> mdz_: migration-assistant panic over; there's something still not right with the control flow, but I've decided it's not release-critical, and the non-zero exit status was due to having an LVM PV on the disk
[03:56] <iwj> So I bet it's just a fossil.
[03:56] <cjwatson> so it's a bug, but it can wait until after feisty
[03:57] <mdz_> cjwatson: ok
[03:57] <AlinuxOS> pitti, hello
[03:57] <pitti> iwj: it seems that Keybuk added it 10 days ago, so it's not really a fossil
[03:57] <iwj> Oh.
[03:57] <iwj> I bet there's a race then somewhere which means it ought to exist first.
[03:57] <AlinuxOS> is today LanguagePackTranslationDeadline ?
[03:57] <Keybuk> hmm?
[03:57] <iwj> Not clear why the postinst is the right place, though.
[03:58] <iwj> 14:33 <pitti> iwj: what is the reason for the mkdir -m 0755 /dev/mapper in libdevmapper1.02.postinst?
[03:58] <mvo_> pitti, Mithrandir: could you please accept the app-install-data-commercial upload to dapper-proposed once it hits (uploaded a couple of minutes ago). bugnumber is #105847
[03:58] <pitti> mvo_: will do
[03:58] <Keybuk> because you can install dmsetup into a system with existing devmapper nodes
[03:58] <mvo_> pitti: great, thanks
[03:58] <Keybuk> the /dev/.static/dev/mapper is more important, I did the other for completeness
[03:59] <pitti> Keybuk: hm, but if /dev/mapper/something already exists, why is the mkdir necessary?
[03:59] <pitti> ah
[03:59] <Keybuk> pitti: that's what the [ ! -d ]  bit is for :p
[03:59] <Keybuk> it's to make the directory if you are *not* running udev
[03:59] <Keybuk> since if you're running udev, the -d /dev/.static/dev bit is true
[04:00] <Keybuk> it makes sure that /dev/mapper exists as a directory on the root filesystem
[04:00] <pitti> Keybuk: I see; well, the library doesn't change all that often, I hope, so I can just clean up after it manually
[04:00] <Keybuk> so that when the udev /dev is mounted, /dev/.static/dev/mapper exists
[04:00] <Keybuk> pitti: "clean up after it" ?
[04:00] <pitti> Keybuk: it breaks the retracer chroots (since you cannot mkdir as normal user in /dev)
[04:00] <Keybuk> oh
[04:00] <Keybuk> you have /dev in your chroots but not /dev/.static/dev ?
[04:01] <Keybuk> (that's probably your bug -- some postinsts fail without /dev/.static/dev)
[04:01] <mdz_> Keybuk: any word on network-manager?
[04:01] <pitti> Keybuk: I didn't say it's a bug in libdevmapper NB, I was just curious what it was for
[04:01] <Keybuk> mdz_: yes, lots of words
[04:01] <Keybuk> most of them four letters long, and not pleasnt
[04:01] <pitti> Keybuk: the chroot's /dev is just the normal system's /dev
[04:01] <Keybuk> pitti: libdevmapper assumes that the directory it creates nodes in exists
[04:01] <Keybuk> pitti: right, but did you recursively mount under that
[04:01] <mdz_> Keybuk: words which meaningfully relate to the 7.04 release?
[04:01] <Keybuk> ie. does /dev/.static/dev exist ?
[04:01] <pitti> Keybuk: there is a /dev/.static/dev on ronne
[04:01] <Keybuk> *shrug*
[04:02] <Keybuk> then I don't see why you're touching that postinst at all
[04:02] <Keybuk> if there's a /dev/.static/dev then you won't be reaching the mkdir you're complaining about ;)
[04:02] <pitti> Keybuk: it's not really mounted, it's just a symlink to the real /dev, so it includes everything under it
[04:02] <pitti> Keybuk: reality just proved that wrong :)
[04:02] <cjwatson> what, a symlink out of the chroot?
[04:02] <Keybuk> sh -x it
[04:02] <Keybuk> I reckon your system is bogus'd somehow
[04:02] <cjwatson> if so then it could be a broken symlink and test -d would return false ..
[04:03] <fabbione> Keybuk, mdz_, cjwatson: #105872 coming in from hw-cert
[04:03] <Keybuk> aye, what colin said -- symlinks can't point outside chroots
[04:03] <pitti> cjwatson: yes, fakechroot supports that to be able to use /proc and /dev
[04:03] <fabbione> 3 bugs so far from 11 testing.. one is not in LP yet
[04:03] <cjwatson> pitti: maybe it doesn't support it properly
[04:04] <pitti> cjwatson: yes, it's pretty hideous
[04:04] <Keybuk> that postinst looks fine
[04:04] <cjwatson> specifically stat() of the symlink sounds like it returns the wrong thing
[04:04] <mjg59> cjwatson: 105234 looks pretty relevant
[04:04] <Keybuk> if udev is running (/dev/.static/dev exists) then it makes the directory under that if it doesn't already exist
[04:04] <mjg59> (release-wise rather than to current conversation)
[04:04] <Keybuk> otherwise it makes the directory on the root filesystem
[04:04] <Keybuk> (so when you boot with udev, /dev/.static/dev/mapper exists)
[04:04] <pitti> Keybuk: alright, I track it down the next time it stumbles over that
[04:04] <pitti> Keybuk: thanks
[04:04] <Keybuk> right
[04:04] <Keybuk> so network manager
[04:05] <fabbione> Keybuk: yeps...
[04:05] <Keybuk> this thing is so totally bogusly broken that I want to cry
[04:05] <Keybuk> basically there's a race between ifup and network manager
[04:05] <Keybuk> so you can end up with two dhclients
[04:05] <Keybuk> since ifup starts one after network manager kills them all
[04:05] <mjg59> dhclient is supposed to have pidfile handling
[04:05] <jtt> is the Ubuntu kernel 2.6.17-10 any different than 2.6.17-10 from kernel.org  i.e. does Ubuntu make any signifcant changes
[04:05] <mjg59> Why isn't that working in this case?
[04:06] <Keybuk> mjg59: the pid file has a random number in it
[04:06] <mjg59> jtt: -10 represents the Ubuntu ABI version, not 2.6.17.10
[04:06] <mjg59> And yes, very different
[04:06] <mjg59> Keybuk: Why?
[04:06] <jtt> mjg59: thanks
[04:06] <Keybuk> no idea
[04:07] <mjg59> Fixing that would presumably fix the issue to some extent
[04:07] <Keybuk> I don't think so
[04:07] <Keybuk> because I'm seeing other weird behaviour caused by dhcdbd
[04:07] <Keybuk> it seems to kill dhclient itself if it gets a status message it wasn't expecting
[04:08] <Keybuk> and I don't understand how network manager kills dhclient
[04:08] <Keybuk> since it does "killall dhclient", which won't match dhclient3 (which is what ifup starts)
[04:09] <Keybuk> I also don't understand why NM seems to take over eth0 when there's nothing even logged in yet
[04:09] <Keybuk> I thought it needed nm-applet to tell it to do that
[04:09] <mjg59> My understanding was that it claimed interfaces on startup
[04:09] <mjg59> In the absence of policy, the policy is no network
[04:14] <Keybuk> so it does appear to "claim" the eth0 interface
[04:14] <Keybuk> bring it down
[04:14] <mdz> it seems to bring up interfaces before login for me (which is fortunate because I want that)
[04:14] <Keybuk> (seperate to it actually killing dhclient, it tells dhcdbd to bring it down)
[04:14] <Keybuk> and then bring it back up again
[04:14] <Keybuk> this happens whether or not it was already up or down
[04:14] <Keybuk> so policy seems to be to always have an interface up
[04:14] <xtknight> hey
[04:14] <mjg59> Interesting
[04:14] <Keybuk> so it seems we have four different things trying to do things to a network interface at the same time
[04:14] <Keybuk> ifup called by udev
[04:14] <Keybuk> ifup -a
[04:14] <Keybuk> dhcdbd
[04:14] <Keybuk> and network manager
[04:14] <Keybuk> it's no surprise it's a bit buggered, tbh
[04:14] <Keybuk> (the ifup -a bit doesn't clash with network-manager, just the ifup-by-udev and we have locking for that)
[04:14] <azazello> why does acpi-support-0.95 tarball have a top level dir called acpi-support-0.94?
[04:14] <mjg59> Because we hate freedom
[04:14] <xtknight> there may be a diff.gz to update it to 0.95?
[04:14] <Mithrandir> cjwatson: #105861> phew
[04:15] <azazello> xtknight: the files inside do mention 0.95
[04:15] <cjwatson> there is another m-a-related bug in that the installation summary doesn't properly display what's to be migrated
[04:15] <xtknight> azazello, apt-get source gives me onloy a 0.95 dir
[04:15] <cjwatson> I'll commit the fix but I don't think it's release-critical
[04:15] <mjg59> azazello: The top level directory name is pretty unimportant
[04:16] <xtknight> what's the status of the network-manager issue?   
[04:16] <ogra> heh
[04:16] <ogra> which of the 1000 ?
[04:16] <ogra> :P
[04:16] <xtknight> heh the one with 1000 duplicates ;)
[04:16] <cjwatson> azazello: the top-level directory name is just whatever the directory happened to be called when the developer built the source package.
[04:17] <xtknight> bug 105234
[04:17] <ubotu> Malone bug 105234 in network-manager "Netowrk manager says disconnected but is connected and working" [Medium,Confirmed]  https://launchpad.net/bugs/105234
[04:17] <cjwatson> read scrollback
[04:17] <mjg59> xtknight: There's a large number of n-m issues that people are working on
[04:17] <Keybuk> I also, as an experiment did the following
[04:17] <Keybuk>  - configured an eth1 and eth2 
[04:17] <Keybuk>  - both auto/dhcp
[04:17] <Keybuk>  - ifup brings them both up, both with dhclient3 processes running
[04:17] <Keybuk>  - started network manager
[04:17] <Keybuk>  - network manager removes the ip from both
[04:18] <Keybuk>  - and then "brings up" eth2 again
[04:18] <Keybuk> dhclient kills the existing eth2 process (correctly, itself) and starts a new one
[04:18] <Keybuk> you're then left in a state with
[04:18] <Keybuk> eth1: UP, no IP, dhclient3 running (!!)
[04:18] <Keybuk> eth2: IP, dhclient, etc.
[04:18] <Keybuk> (and ifup thinks both are up)
[04:19] <Keybuk> after a short while (dhcp lease time?) dhclient for eth1 gets an IP again
[04:19] <bddebian> Heya
[04:19] <Keybuk> which sends NM into an absolute spawn
[04:19] <Keybuk> spasm, even
[04:22] <seb128> Mithrandir: sound-juicer could use a Depends on libgnomevfs2-extra (required to do cddb), that's low importance though, ubuntu-desktop install it, ok to upload the change now or rather next cycle?
[04:22] <xtknight> what determines whether NetworkManager handles an interface or not?
[04:23] <hunger> xtknight: As far as I could find out so far: The phase of the moon.
[04:23] <seb128> xtknight: the interface being list to /etc/network/interfaces or not
[04:23] <seb128> listed
[04:23] <mvo_> Mithrandir: could you please do a publisher run for app-install-data-commercial in dapper-proposed ? 
[04:23] <Mithrandir> mvo_: the publisher is on auto.
[04:24] <Keybuk> seb128: except that's bogus
[04:24] <Keybuk> since NM thinks it's fun to kill dhclients
[04:24] <xtknight> hrmm but /etc/network/interfaces is for ifup() at startup too, right?
[04:24] <Treenaks> Keybuk: can't it just keep track of its own children, and just kill THOSE, instead of killing anything that looks remotely like a dhclient?
[04:25] <ogra> Keybuk, well, at least it leaves static interfaces alone atm
[04:25] <Keybuk> then we end up in the situation where you can't use network manager until you've removed things from /e/n/i
[04:25] <mvo_> Mithrandir: oh, sorry. I thought it was still in manual mode
[04:25] <ogra> which was the edgy situation, no ? 
[04:27] <Keybuk> which was a pain, but at least it worked
[04:27] <ogra> yeah
[04:31] <Keybuk> I don't really have any release-useful suggestions at this point
[04:31] <TheInfinity> when was the update where /usr/share/X11/fonts/... was changed to /usr/share/fonts/X11/... ? because german and english wiki entrys have this old version ...
[04:31] <Keybuk> with half a dozen tests, I've convinced myself that NM only works if you don't use ifup for anything other than lo
[04:32] <Keybuk> and even the work to go back to that would be more than 1 week including testing
[04:32] <Nafallo> TheInfinity: early feisty
[04:33] <ogra> Nafallo, i thought that came with 7.0 already ? 
[04:33] <Nafallo> ogra: that change was when we merged back with Debian. I'm fairly certain on that :-).
[04:33] <xtknight> Keybuk, did the patch ( http://librarian.launchpad.net/7297631/patch_network-manager_0.6.4-6ubuntu6.patch ) produce satisfactory results?  I assume there is another more serious underlying problem?
[04:35] <TomaszD> there is a problem with network-manager.  "static configuration" option isn't translatable
[04:35] <tepsipakki> TheInfinity: if you mean the change in new xorg.conf-files, it was made in February when 7.2 was merged
[04:35] <Mithrandir> TomaszD: already fixed.
[04:35] <TomaszD> Mithrandir, checked lp this morning, not available
[04:36] <Mithrandir> TomaszD: that's because LP is lagging.
[04:36] <TheInfinity> teps ... okay ... then we have to make a switch in wiki ...
[04:36] <TomaszD> Mithrandir, so should be available, say, tomorrow?
[04:36] <Nafallo> tepsipakki: are you sure? I still think the change was very early feisty when we merged back with Debian and they had that other path...
[04:36] <cjwatson> I think it was early *edgy* when we merged back with Debian and they had that other path
[04:37] <tepsipakki> Nafallo: hah, you are right :)
[04:37] <cjwatson> oh, hmm, maybe not
[04:37] <tepsipakki> the ones we merged from in February had both paths
[04:37] <Nafallo> yay memory! :-)
[04:38] <cjwatson> I must be thinking of xkb
[04:38] <Keybuk> more than I hate NM
[04:38] <Nafallo> Keybuk: webm? webmin?
[04:38] <Keybuk> Nafallo: literate programming
[04:38] <StevenK> Nafallo: Don't use language like that around me.
[04:38] <Nafallo> ah ;-)
[04:39] <Keybuk> noweb thingy
[04:51] <iwj> Is it just me or is LP being hopeless today ?  I just got an ECONNRESET (according to this here livecd's firefox).
[04:51] <TomaszD> iwj, it
[04:52] <TomaszD> it's on its knees
[04:52] <StevenK> iwj: Agreed, I've gotten four or five
[04:52] <Mithrandir> iwj: it's slow.
[04:52] <iwj> For me it's beyond slow.
[04:53] <cjwatson> #launchpad-code
[04:53] <Hobbsee> iwj: probably getting as bad as australia's often is - or used to be
[04:53] <cjwatson> (irc.canonical.com)
[05:02] <ogra> iwj, seems i'm joining in ... connection reset as well here now 
[05:02] <fabbione> i just spoke with SteveA about LP
[05:03] <fabbione> there is going to be somekind of improvement within the day
[05:15] <ogra> good
[05:16] <dholbach> amd64 desktop look good
[05:16] <jsgotangco> yeah
[05:22] <seb128> cjwatson: what component has the ubiquity translations?
[05:38] <iwj> Hmm, I'm glad I saved a copy of that bug text now.
[05:40] <cjwatson> seb128: debian-installer
[05:41] <seb128> cjwatson: ok, thank you
[05:43] <iwj> Oh, looks like I'm reporting this bug twice since the email from the first attempt has just arrived.
[05:45] <iwj> Does langpack installation in ubiquity do some kind of LP transaction ?
[05:45] <iwj> I have an install here which seems to have got stuck at that stage.
[05:46] <pitti> iwj: downloading the packages at most
[05:46] <pitti> iwj: but that shouldn't happen for English
[05:47] <iwj> I'm installing in German just for a change.
[05:48] <pitti> iwj: do you see a 'cancel' button?
[05:48] <iwj> Yes.
[05:48] <iwj> I didn't select it and it has just woken up.
[05:48] <pitti> iwj: if so, then it's the stage where it downloads language-support-de and dependnecies
[05:48] <iwj> How big is that data ?
[05:48] <pitti> iwj: hmm, magnitude of 30 MB
[05:48] <iwj> Ah, in that case the bug is just poor progress display.
[05:49] <pitti> iwj: hm, IIRC there was a textual ETA somewhere
[05:49] <pitti> it usually works quite well for me
[05:49] <iwj> Not in this case.
[05:49] <ogra> i usually get a question if i want to download it 
[05:49] <pitti> ogra: in alternate, yes, but not in ubiquity
[05:49] <ogra> ah
[05:49] <pitti> it just does and allows you to cancel
[05:58] <andre_pl> with the latest feisty kernel my tifm card reader doesn't work at all.  the modules load, but otherwise theres nothing in dmesg.. i had read on launchpad that the latest kernel had included a newer tifm driver that was supposed to solve most of these problems but in my case it is worse.
[05:59] <reitblatt> andre_pl: you would get better help in #ubuntu-kernel
[05:59] <andre_pl> reitblatt: thanks. I thought this was the right place. :P
[06:00] <Hobbsee> andre_pl: likely too late to fix now.  and this may be the right place, too
[06:01] <andre_pl> ok, well here's another one. since last nights updates when i boot my computer, network manager shows not connected, but it is connected, and it works.
[06:02] <andre_pl> if i disable/enable it sows the right status
[06:02] <Nafallo> andre_pl: that's a feature from the last changelog :-)
[06:02] <Nafallo> andre_pl: people are working on it.
[06:02] <andre_pl> haha, a "feature"
[06:02] <andre_pl> i wanted to go bug crazy today but launchpad is dead.
[06:04] <Nafallo> there already are a couple of thousands bug reports against network-manager :-P
[06:04] <Hobbsee> that many?  there were a couple of hundred, last time i looked...
[06:04] <andre_pl> yeah i've got quite a few issues right now though, not just network manager.
[06:05] <Hobbsee> andre_pl: are they release-critical?
[06:07] <andre_pl> well... maybe? lol.  about half of the time I start my laptop it just hangs with a _ in the top left corner after the splash finishes.. then I alt-f1 and it says its failed to read some image file i have to press enter to get it start the boot process.
[06:07] <Hobbsee> did you get the syslog?
[06:08] <andre_pl> nope.  i probably can though, it happens often enough.
[06:08] <Hobbsee> --> #ubuntu+1 anyway
[06:08] <Hobbsee> seeing as this isnt a supoprt channel
[06:08] <Hobbsee> yay, it's friday!
[06:10] <jsgotangco> Hobbsee: yay!
[06:10] <Hobbsee> jsgotangco: we can now EOL most of https://launchpad.net/ubuntu/+bugs?field.searchtext=breezy&orderby=-importance&search=Search&field.status%3Alist=Unconfirmed&field.status%3Alist=Confirmed&field.status%3Alist=In+Progress&field.status%3Alist=Needs+Info&field.status%3Alist=Fix+Committed&field.assignee=&field.bug_reporter=&field.omit_dupes=on&field.has_patch=&field.has_no_package=
[06:10] <Hobbsee> when LP gets better
[06:11] <Nafallo> that's a small URL ;-)
[06:11] <Hobbsee> indeed
[06:11] <TomaszD> andre_pl, don't worry about LP being dead slow, do what I do. Take a magazine, click a link, read an article or two, look if the page loaded. Repeat.
[06:11] <jsgotangco> my eyes! my eyes!
[06:11] <Nafallo> TomaszD: haha
[06:11] <Hobbsee> TomaszD: i used to routinely do it.  every page would take 40-50 seconds to load.
[06:11] <andre_pl> TomaszD: its not even loading now.
[06:11] <andre_pl> and i cant get any search results for anything.
[06:12] <Hobbsee> TomaszD: and i got all of the bugs on the packages i was interested in, emailed to me
[06:12] <Hobbsee> night all!
[06:13] <mvo_> Mithrandir: ok if I do a g-a-i upload so that the desktop data is updated? or should I wait with this
[06:13] <Hobbsee> TomaszD: bugmail isnt bad, just selecting all the pages you want to load
[06:13] <Hobbsee> then looking later
[06:13] <TomaszD> Hobbsee, how about translating in Rosetta? :] 
[06:13] <TomaszD> getting parts of the .po file in mail lol
[06:13] <Hobbsee> TomaszD: never tried.  i only speak a bit of german, and english
[06:14] <TomaszD> I know Polish, English, German and a bit of Spanish
[06:14] <Hobbsee> neat :)
[06:14] <TomaszD> though it becomes apparent in a few moments that English isn't my native tongue
[06:14] <TomaszD> :P
[06:15] <Hobbsee> heh
[06:15] <TomaszD> 13 years of learning and I still can't get the hang of some things
[06:15] <TomaszD> and I'm teaching the bloody language too
[06:15] <TomaszD> :] 
[06:15] <Hobbsee> yeah well.  5+ years of learning german, and i still cant hold a conversation in it
[06:15] <Hobbsee> haha
[06:16] <Hobbsee> mind you, we learnt about things like saltmines, instead of useful conversation
[06:16] <Hobbsee> TomaszD: your english seems fine so far, anyway :P
[06:16] <Hobbsee> it's understandable
[06:16] <TomaszD> well I hope so, I'm an upstream GNOME translator, I should know my thing
[06:16] <TomaszD> :P
[06:17] <TomaszD> and about German, it's the same here, four years gone to waste
[06:17] <Hobbsee> hehe
[06:17] <TomaszD> Was hasst tu gestern gemacht? Ich habe Computer gespielt und Bucher gelesen
[06:17] <TomaszD> every single lesson she would ask us that
[06:17] <Hobbsee> ah crud...i should know that....
[06:18] <Keybuk> Apr 12 09:56:18 localhost NetworkManager: <information>^ICouldn't send DHCP 'up' message because: name 'com.redhat.dhcp.OperationInProgress', message 'interface eth0 is being released. Please try again later.'.
[06:18] <Keybuk> hmm
[06:18] <iwj> Woah!   `Retour   Revenir en arriere #-#-#-#-# fr.po (debian-installer) #-#-#-#-# Retour'
[06:18] <Hobbsee> TomaszD: i have played on the computer and learned from the book...  dont know what the other one is
[06:18] <Keybuk> and then ...
[06:18] <Keybuk> Apr 12 09:56:25 localhost NetworkManager: <information>^IActivation (eth0) Stage 3 of 5 (IP Configure Start) started... 
[06:18] <Keybuk> Apr 12 09:56:25 localhost avahi-autoipd(eth0)[11501] : Successfully claimed IP address 169.254.7.163
[06:18] <Keybuk> Apr 12 09:56:25 localhost avahi-autoipd(eth0)[11501] : fopen() failed: Permission denied
[06:18] <Keybuk> Apr 12 09:56:25 localhost NetworkManager: <information>^IDHCP daemon state is now 11 (unknown) for interface eth0 
[06:19] <Keybuk> Apr 12 09:56:25 localhost NetworkManager: <information>^IDHCP daemon state is now 14 (normal exit) for interface eth0 
[06:19] <Keybuk> Apr 12 09:56:27 localhost NetworkManager: <information>^IActivation (eth0) Beginning DHCP transaction. 
[06:19] <Keybuk> Apr 12 09:56:27 localhost dhclient: There is already a pid file /var/run/dhclient.eth0.pid with pid 134993440#] 
[06:19] <TomaszD> What have you been doing yesterday, that was the question
[06:19] <Keybuk> kablooey
[06:19] <Hobbsee> TomaszD: ahhh...of course.
[06:19] <TomaszD> pitti is the root of all evil
[06:19] <cjwatson> iwj: oh damn silly French translators
[06:19] <Keybuk> Treenaks: ping
[06:19] <Hobbsee> Keybuk: blame the innocent who arent here, of course!
[06:19] <iwj> cjwatson: Shall I file a bug ?
[06:20] <cjwatson> iwj: hang on, phone
[06:20] <iwj> cjwatson: OK.
[06:20] <iwj> cjwatson: It seems really very ugly and if it's on the common path I might even call it RC (although I suppose it only affects French).
[06:21] <iwj> cjwatson: I see this in the "confirm resize partitions" dialogue in ubiquity.
[06:22] <cjwatson> iwj: sigh, I suppose so, though I can only work around it
[06:22] <cjwatson> Mithrandir,mdz: ^--
[06:22] <Keybuk> elmo: so yeah, somewhere along the line, you end up with a bogus pid for dhclient
[06:22] <Keybuk> I have a vague suspicion it's related to that previous avahi-autoipd error, but I could be wrong
[06:23] <mdz> cjwatson: I'm not sure I can tell the difference between an ugly text bug and standard French punctuation rules :-P
[06:24] <Keybuk> (though that might be spurious, it's entirely possible it's a dhclient bug)
[06:25] <iwj> cjwatson: Do you want a screenshot or is what I write in bug 105903 enough ?
[06:25] <ubotu> Malone bug 105903 in debian-installer "crazy french translation for "retour" and "suivant"" [Undecided,Unconfirmed]  https://launchpad.net/bugs/105903
[06:32] <mdz> Mithrandir: what's the reason for this:
[06:32] <mdz>         if (! applet->device_list)
[06:32] <mdz>                 state = NM_STATE_DISCONNECTED;
[06:34] <iwj> mdz: This one is obviously really crazy.  It looks like the computer has gone completely haywire.
[06:34] <cjwatson> mdz: it's a totally fucked multi-line translation due to somebody stupidly unfuzzying a translation
[06:35] <cjwatson> iwj: no need
[06:37] <Keybuk> elmo: so, err, that's totally freaky
[06:39] <Mirv> iwj: what is the English string and how is it translated in French (if re-translated into English)
[06:41] <Mirv> oh wait, the bug, if it's there I read it from there
[06:42] <iwj> cjwatson: I haven't seen this crazy string again so I think it must only show up in manual partitioning (or perhaps also auto-resize).
[06:43] <Mirv> iwj: oh, ok, so someone having had a stupid fuzzy (I've seen those) with all the ## letters etc., and the translators as unfortunately marked is as "reviewed"
[06:56] <iwj> This flashing /var/lib/os-prober/mount icon is a bit distracting really.
[06:56] <iwj> Not just during migration-assistant (as I said in bug 105531) but during grub installation too.
[06:56] <ubotu> Malone bug 105531 in migration-assistant "/var/lib/os-prober/mount briefly appears on desktop" [Undecided,Confirmed]  https://launchpad.net/bugs/105531
[06:57] <evand> iwj: fix committed, but it's up to cjwatson if he wants it included in the release.
[06:57] <dholbach> iwj: i'll take a break after this installation - will do test the oem d-i i386 dvd?
[06:59] <iwj> dholbach: You mean will I test it ?
[06:59] <dholbach> iwj: sorry, I forgot the "you" :)
[06:59] <iwj> Sure but are there special instructions for oem installs ?
[07:00] <iwj> It looks like my rsync got the dvd ok.
[07:00] <fabbione> i have some spare vmware cycles to spare...
[07:00] <fabbione> iwj: just select oem at the dvd boot IIRC
[07:00] <iwj> OK, and nothing in particular else needed ?  NP
[07:01] <dholbach> iwj: after that it gives you instructions on how to run oem-config-prepare (or whatever it is)
[07:01] <fabbione> yeah it's pretty straight forward
[07:03] <heno> fabbione: what arch?
[07:03] <fabbione> heno: ?
[07:03] <fabbione> heno: for vmware? i386
[07:03] <heno> fabbione: i386, amd64, sparc, all of the above?
[07:04] <heno> ah, ok
[07:04] <fabbione> there is no sparc vmware :)
[07:04] <fabbione> and amd64 support is experimental in ws 5.x
[07:04] <heno> fabbione: can you test kubuntu alternate?
[07:04] <fabbione> heno: i only have ubuntu images here.. if it's not urgent i can download and test it
[07:05] <heno> virtualbox is open source (ish) you might get that to run on the sparc
[07:05] <fabbione> nah... i am waiting for hw virtualization support on Niagara
[07:05] <heno> fabbione: yep, ubuntu alternate i386 needs testing love too
[07:05] <fabbione> ok
[07:05] <fabbione> ubuntu alternate is
[07:06] <heno> thanks!
[07:06] <dholbach> iwj: gracias
[07:07] <Mithrandir> mdz: avoid filling the .xsession-errors with a load of stuff about it not being able to find the proper icon.
[07:11] <cjwatson> iwj: it'll be presented by a bunch of different dialogs - partitioning, migration-assistant, misc failures
[07:11] <iwj> cjwatson: Oh dear.
[07:12] <cjwatson> Mirv: I'm pretty sure I tried to fix this ages ago but Rosetta will only let me present it as a "suggestion" despite me being the maintainer and knowing that it's broken
[07:12] <cjwatson> and translation syncs are typically too big for me to be able to notice all these problems every time
[07:13] <Mirv> cjwatson: yeah, the launchpad requires you'd be member of the translation team in question also. anyway, translators should spot those broken suggestions and not mark them as reviewed, but of course it can happen
[07:14] <cjwatson> there's a similar problem with Turkish
[07:14] <cjwatson> no others
[07:16] <Mirv> it depends also on how much "random" people there are in each translation team, as it's generally known that in the early times of Launchpad many translation teams accepted too many members even though Rosetta QA was (and is) lacking
[07:19] <keescook> doko: can you look at bug 102786?  Someone ran into problems with the python-uno upgrade (it looks like the pycentral call in the prerm is failing?)
[07:19] <ubotu> Malone bug 102786 in openoffice.org "python-uno fails during upgrade from 2.0.4-0ubuntu4 to 2.0.4-0ubuntu5" [Undecided,Confirmed]  https://launchpad.net/bugs/102786
[07:20] <keescook> oh, you're already on it.  our comments to the bug crossed paths.  :P
[07:20] <fabbione> crossing the fluxes is BAD....
[07:21] <fabbione> bad is the life the way you know  it will stop instantly and each atom in your body will explode at light speed
[07:21] <keescook> okay, that's bad.  important safety tip
[07:22] <lappy> hey ogra thanks for the acceptance to PyStart
[07:22] <fabbione> keescook: i guess you have never seen Ghostbuster 1.. did you? :)
[07:22] <ogra> thanks for doing it :)
[07:23] <keescook> fabbione: you're being sarcastic, right?  "okay, that's bad.  important safety tip" was the next line.  :)
[07:23] <fabbione> keescook: oh i didn't remember it in english.. i saw it in italian and doing a rought translation on the fly :)
[07:24] <keescook> fabbione: hehe. cool.   "If someone asks you if you're a god ...."
[07:24] <fabbione> ahha yeah i remember that
[07:24] <keescook> "you say YES!"  :)  one of my favorites.  :)
[07:25] <fabbione> ehhehe
[07:44] <iwj> `Text mode install for manufacturers' is oem, I take it.
[07:44] <ogra> yeah, that was changed recently iirc
[07:45] <ogra> iwj, are you currently in a fresh install ? i'm looking for someone who can test something for me with RC i'm not sure its my machine that wrong ...
[07:45] <ogra> *that's
[07:46] <lappy> ogra: what do you need tested?
[07:47] <ogra> screensaver preview ... but be careful it crashes X for me over here if you click the "leave screensaver" button i'm not sure its my crappy ati card or its the screensaver 
[07:50] <iwj> ogra: Both my test systems are currently installing.
[07:51] <iwj> Ask me again in a bit :-).
[07:51] <ogra> iwj, yeah, will do :)
[07:51] <ogra> if thats happening for others its an RC ... i'm a bit worried ...
[07:52] <superm1> currently firefox's update-notifier implementation copies over a file to /var/lib/update-notifier/user.d, but how does it signal to the running update-notifier process that it needs to show a popup?
[07:52] <superm1> i seem to be missing it somewhere in the postinst or something
[07:55] <superm1> is the update-notifier process perhaps supposed to be holding a inotify on /var/lib/update-notifier/user.d?
[07:57] <cjwatson> Mithrandir,iwj: bug 105903 fixed in unapproved queue
[07:57] <ubotu> Malone bug 105903 in ubiquity "crazy french translation for "retour" and "suivant"" [Undecided,Unconfirmed]  https://launchpad.net/bugs/105903
[07:57] <Mithrandir> cjwatson: yay, great.
[08:08] <iwj> cjwatson: Great.
[08:11] <superm1> asac, ping
[08:12] <iwj> ogra: So what was that test you wanted someone to do ?
[08:12] <iwj> I have a nice fresh `erase disk' ubiquity-created install here which isn't doing anything else right now.
[08:13] <ogra> iwj, just start the screensaver preview and survive closing it 
[08:14] <heno> ogra: I just tested it on a headless box I access via VNC. Just clicking the Preview button kills X there :(
[08:14] <ogra> argh
[08:14] <ogra> i was hoping that was only me
[08:15] <heno> can I get you some logs?
[08:15] <iwj> ogra: All seems to work just fine for me.
[08:15] <ogra> i have it here so i can dig through it ...
[08:15] <ogra> phew
[08:15] <ogra> ok, two alternate reports ... for now i'll blame vnc ...
[08:16] <ogra> heno, what kind of graphics card is that ? 
[08:16] <ogra> i have an ati here ... 
[08:16] <heno> I'm doing another install so I can test it on real hardware soon
[08:16] <heno> ogra: nvidia
[08:16] <ogra> hmm
[08:16] <Treenaks> Keybuk: pong
[08:16] <heno> but the vnc stuff is not very stable
[08:16] <iwj> I have some very boring P4 onboard graphics here.
[08:17] <ogra> i'll try it with fglrx soon ...
[08:17] <heno> tested in in virtual box, no problems there
[08:17] <ogra> strange ... but something isnt completely stable there it seems ...
[08:17] <Keybuk> Treenaks: s'ok, pretty sure it's not avahi
[08:22] <iwj> What exactly seems to me most likely to make it crash ?  Any particular screensaver ?  I've been in and out of the screensaver pref and the fullscreen mode, and selected different ones, and it works perfectly for me.
[08:22] <iwj> s/to me/to make/
[08:23] <heno> ogra: I can confirm the X crash on two physical machines
[08:23] <Treenaks> Keybuk: ok :)
[08:23] <ogra> iwj, i used the 3D-bubbles one (which is a 2d screensaver unlike the name indicates)
[08:24] <ogra> but i dont think it matters ... 
[08:24] <ScottK> Keybuk: I tested your new network-manager_0.6.4-6ubuntu7 packages with Kubuntu both wired and wireless with no issues (was not having problems before either).
[08:24] <fabbione> Keybuk: the upstart ttyS0 transition from edgy to feisty is not fixed yet.. Release Note?
[08:24] <heno> I got it on a Toshiba laptop and and amd64 /w nvidia
[08:24] <fabbione> Keybuk:  a reboot in that condition can leave users without access to the machine for a while
[08:25] <Keybuk> fabbione: it's on my list
[08:25] <fabbione> Keybuk: ok
[08:25] <iwj> ogra: Bubble3D I take it.  Yes, no problem.
[08:25] <Keybuk> I don't quite understand why that doesn't work
[08:25] <fabbione> Keybuk: do you want access on a machine to test?
[08:25] <fabbione> Keybuk: i can set that up for you by tomorrow
[08:25] <fabbione> Keybuk: or later this evening
[08:25] <fabbione> you tell me..
[08:26] <keescook> Keybuk: I'm still seeing snapshot races; they're just much more rare.  where should I record that? new bug or reopen old?
[08:27] <Keybuk> new bug
[08:27] <mdz> any other feedback on the new N-M?
[08:28] <mdz> it needs regression testing as well as confirmation that it has fixed the main issue
[08:28] <ogra> mdz, you have an ati card in your laptop, right ? 
[08:28] <mdz> ogra: yes
[08:28] <ogra> i fear i found an X crasher bug :(
[08:29] <mdz> my X hasn't crashed in ages
[08:29] <ogra> could you try to open the screensaver preview and close it again ? 
[08:29] <ogra> (colse all important stuff before)
[08:29] <mdz> it does corrupt the screen occasionally with the current driver
[08:29] <ogra> it seems for me with an ati card/driver it crashes X ... heno saw it on an nvidia
[08:30] <mdz> ogra: no problems
[08:30] <ogra> heno, was that an nv or nvidia driver ? 
[08:30] <ogra> mdz, phew, ok
[08:30] <mdz> ogra: I close it by clicking 'leave fullscreen'; is that what you do?
[08:30] <ogra> yes
[08:30] <heno> ogra: it was nv
[08:30] <mdz> ogra: and which screensaver do you run?
[08:30] <ogra> thats what dropys me out of X
[08:30] <ogra> mdz, Bubble3D
[08:30] <mdz> I tried with blank, random and Bubble3D
[08:31] <heno> a clean ubuntu install on an amd64 desktop
[08:31] <ogra> or any other i guess, i have stuff open i need to finish before testing more
[08:31] <mdz> probably specific to certain cards
[08:31] <ogra> heno, aha
[08:31] <ogra> i have an amd64 as well here
[08:31] <mdz> in any case, not a release issue.  possible SRU if further analysis warrants it
[08:31] <ogra> but i386 install
[08:31] <ogra> yeah
[08:31] <ogra> i was just cautious ... since i didnt see it before at all 
[08:32] <heno> i got in on an amd64 install and an i386 install on an i386 laptop
[08:32] <ogra> but if its not affecting all setups its ok for now
[08:32] <heno> I got it on 2 out of 3 :)
[08:32] <heno> plus the vnc one
[08:33] <ogra> its surely not the most used functionality
[08:33] <ogra> i.e. its no NM :)
[08:33] <heno> true, but it can loose you data, which is bad
[08:33] <heno> say you have a long OOo doc open and X bails out
[08:33] <ogra> yeah
[08:33] <ogra> i know what you mean 
[08:34] <heno> solution might be to remove the preview button ...
[08:34] <ogra> hmm
[08:35] <ogra> that will make me drown in regression bugs :)
[08:41] <keescook> Keybuk: done. see bug 105936, with example test scripts.
[08:41] <ubotu> Malone bug 105936 in lvm2 "snapshot creation failure race "in use: not deactivating"" [Undecided,Unconfirmed]  https://launchpad.net/bugs/105936
[08:44] <ogra> so its the ati driver for me ... 
[08:44] <ogra> fglrx works flawless
[08:44] <tepsipakki> ogra: file a bug and attach the log :)
[08:45] <iwj> This oem install is freaky.  I've never done this before.  Is there a document saying what I'm expecting ?
[08:46] <iwj> Hello, anyone ?
[08:47] <cjwatson> iwj: https://wiki.ubuntu.com/Testing/InstallMethods (fairly briefly)
[08:47] <cjwatson> typically the manufacturer does everything up to 'sudo oem-config-prepare', and the end user gets the next bit
[08:47] <iwj> Oh FFS!  I typed  `sudo oem-config prepare'  and it did a very weird thing.
[08:48] <cjwatson> ha
[08:48] <cjwatson> been meaning to add a .desktop for that
[08:48] <harpreet> Colin: Harpreet here from GlassFish. Just wanted to know the status for the packages
[08:48] <iwj> What does the `prepare' argument mean there ?
[08:48] <cjwatson> 'sudo oem-config prepare' should crash with AttributeError
[08:48] <cjwatson> iwj: frontend
[08:49] <iwj> cjwatson: It didn't.
[08:49] <iwj> It produced a gui and updated initramfs and then the gui vanished.
[08:49] <iwj> Halfway through me fiddling with it.
[08:49] <iwj> I'll report a bug.
[08:50] <cjwatson> oh, yeah, ok, -f is a frontend - yes, it should reject that, 'prepare' is meaningless
[08:50] <cjwatson> harpreet: I did some review earlier today and what I saw was fine (and accepted), but it's been a very busy day
[08:51] <cjwatson> harpreet: I'm looking at sunwderby at the moment
[08:51] <gnomefreak> there is no support for ppc in feisty right?
[08:51] <ogra> community support :)
[08:52] <harpreet> Colin: Ok - thanks. I will just wait up for an email from you then
[08:52] <gnomefreak> :) but will it run it
[08:52] <iwj> If I move something to the panel is it supposed to stick after the reboot ?
[08:52] <gnomefreak> i havent seen ISO;s for ppc 
[08:52] <cjwatson> harpreet: are you prepared to test the binary packages from our archive?
[08:52] <ogra> oh, wow, compiz doesnt generate a white screen for me anymore ...
[08:52] <ogra> not that it would work ... 
[08:53] <cjwatson> gnomefreak: http://cdimage.ubuntu.com/ports/daily/current/ http://cdimage.ubuntu.com/ports/daily-live/current/
[08:53] <cjwatson> and generally under /ports/
[08:53] <gnomefreak> ty hmmm
[08:53] <harpreet> Colin: That was my followup question. Where do we get the binary from your archive to test.
[08:53] <cjwatson> harpreet: do you have an Ubuntu Feisty system up and running?
[08:55] <harpreet> Colin: yes
[08:55] <xan_> Hi, I asked this question in many ways: #ubuntu, #ubuntu+1, etc and no one answer me. Can you answer this tech question to me?
[08:56] <xan_> I only want to know how to disable that checkfs.sh checks all the filesystems and achieve that it only checks /
[08:56] <cjwatson> harpreet: System -> Administration -> Synaptic Package Manager, make sure multiverse is turned on in Settings -> Repositories, search for your package names and install them
[08:57] <iwj> Anyway, it seems to have worked anyway and that strange rune didn't do any harm.
[08:57] <iwj> So that's the dvd done apart from winfoss.
[08:57] <cjwatson> xan_: set the pass field in /etc/fstab to 0 for the filesystems you don't want to check; see 'man fstab'
[08:58] <xan_> I have no field
[08:58] <harpreet> Colin: Ok
[08:59] <xan_> I can't show you the fstab because I have no internet connection in the other computer
[08:59] <mjg59> xan_: This isn't a support channel
[08:59] <xan_> but cjwatson I have no field (nop 0, nop 1, ...) in fstab
[09:00] <cjwatson> xan_: please read the man page to which I referred you
[09:00] <cjwatson> it has all the necessary information
[09:00] <cjwatson> harpreet: the imq binary should be available within an hour
[09:01] <cjwatson> harpreet: I just accepted the sunwderby source, but the binary will need manual approval later
[09:01] <harpreet> Colin: Ok. I will need all of them to test this out.
[09:01] <xan_> cjwatson: collin, please, I believe that all filesystems appeared or not in fstab are checked from /etc/init.d/checfs.sh
[09:01] <xan_> is it correct?
[09:01] <harpreet> Colin: What does that mean? I apologize I do not understand the entire process
[09:01] <cjwatson> xan_: I have already given you your answer.
[09:02] <cjwatson> xan_: your belief is incorrect (unless there is a serious bug)
[09:02] <cjwatson> see the fsck man page
[09:02] <desrt> johanbr; u
[09:02] <desrt> erp.
[09:02] <cjwatson> harpreet: progress :-)
[09:02] <xan_> in feisty beta so?
[09:02] <cjwatson> harpreet: the binary is what you need to actually install it from the package manager, but the source is the thing we really seriously review
[09:03] <cjwatson> source gets automatically built into binaries on our build daemons
[09:03] <cjwatson> the cron jobs that fire that off happen roughly hourly (there's lots of detail there but it isn't relevant)
[09:06] <harpreet> Colin: Where are you located?
[09:07] <cjwatson> harpreet: Cambridge, England
[09:07] <harpreet> Colin: Cool.
[09:08] <harpreet> Colin: So I wait up an hour for getting all the binary packages and test it out. Once we are okay - I let you know. 
[09:09] <soho> found a bug, totem wont play mpgets-files in feisty. it opens the add-software dialog and suggests a gstreamer-plugin which i installed.
[09:09] <cjwatson> harpreet: glassfish Depends: sunwderby (>= ${Source-Version}), imq (>= ${Source-Version}), sun-java5-jre, glassfish-bin (>= ${Source-Version})
[09:10] <harpreet> Colin: Yes
[09:10] <cjwatson> harpreet: I won't reject this upload for that, but ${Source-Version} is wrong there
[09:10] <cjwatson> you may only use that for binaries from the same source package
[09:10] <cjwatson> sunwderby, imq, and glassfish-bin are all from totally separate source packages and thus their version numbers are (theoretically) independent
[09:11] <cjwatson> harpreet: as it is, any future upload of glassfish without also uploading the others will break
[09:11] <cjwatson> I would suggest just writing normal version numbers in place of ${Source-Version} there
[09:11] <Solarion> are all of you able to reach C3 and/or C4 processor powersaving states?
[09:11] <Treenaks> I am
[09:12] <harpreet> Colin: the future uploads of GF will also upload the others as we actually build it lock-step
[09:12] <Solarion> Treenaks: do you have hci_usb?
[09:12] <cjwatson> harpreet: likewise in glassfish-bin
[09:12] <Solarion> something on my box is blocking C3
[09:12] <cjwatson> harpreet: you mean that you intend to merge all of those into one source package?
[09:12] <Treenaks> Solarion: what's that?
[09:12] <Treenaks> Solarion: kernel module? yes.
[09:13] <Solarion> yes
[09:13] <harpreet> Colin: No not right away. What happens is we build this from one codebase and we have packaged them separately.
[09:13] <Solarion> Treenaks: feisty?
[09:13] <Treenaks> Solarion: yes
[09:13] <cjwatson> harpreet: ok, in that case the Depends lines are definitely a bug and need to be corrected
[09:14] <cjwatson> if they're separate source packages, you really can't assume that other source packages will be the same versions
[09:14] <pitti> ugh, back from mega install testing
[09:14] <pitti> so, this expert install was kind of messy, but the rest went well
[09:14] <cjwatson> harpreet: an hour> it will take a little longer for all of them to be available, I'm afraid
[09:14] <cjwatson> harpreet: imq will be available within the hour
[09:14] <Solarion> Treenaks: may I msg you?
[09:15] <Treenaks> Solarion: I don't think I can help you fix your bug
[09:15] <Solarion> I just want to know what things are keeping me from C3
[09:16] <harpreet> Colin: but we do no need the sunwderby, glassfish-bin, imq for glassfish.
[09:17] <micahcowan> What can you do with a coredump, if you lack the program that dumped it? (Assuming you're an expert developer/debugger)
[09:18] <pitti> Keybuk: hm, did you ever try a raid-1 install in expert mode? I just did, and booting the result threw me at an initramfs prompt; I had to manually do 'mount /dev/md0 /root' to continue
[09:19] <pitti> Keybuk: yesterday I did something similar with a different layout and it worked, so it's a rather special case, I figure (not RC critical)
[09:19] <pitti> micahcowan: not much, I'm afraid
[09:20] <micahcowan> pitti, surely there must be a way to get assembly dumps from the core program, at least? I mean, the code is in there, along with stack values, etc. ? What tools would you use to get that?
[09:21] <pitti> micahcowan: but the code is not contained in the core dump
[09:21] <pitti> micahcowan: it just has references to the executables and libraries
[09:21] <harpreet> Colin: are you indicating that this needs to be fixed for this release.
[09:21] <cjwatson> harpreet: no
[09:21] <cjwatson> I'm just saying it's a bug
[09:22] <cjwatson> harpreet: OK, I've accepted all the source packages now
[09:22] <harpreet> Colin: whew ... I will file it :-)
[09:22] <cjwatson> harpreet: I'm going out for the evening shortly (it's late here), but I'll accept the binaries when I get back
[09:22] <harpreet> Great
[09:22] <cjwatson> (assuming there are no obvious showstoppers there)
[09:22] <fabbione> pitti: what bootloader did you chose?
[09:22] <pitti> fabbione: grub, same as yesterday
[09:22] <fabbione> pitti: what layout did you use?
[09:23] <fabbione> raid-1 has been working forever...
[09:23] <harpreet> Colin: So will you send an email that says this to me. Everyone here is on pin and needles and your email will calm things down
[09:23] <harpreet> Colin: I mean after you accept it ofcourse :-)
[09:23] <pitti> fabbione: /dev/hda5 and /dev/hda7 -> /dev/md0, with /dev/hda6 as /boot
[09:23] <micahcowan> pitti, but isn't the executable code loaded into VM at the time the process is running? I had thought that the coredump included everything in that process's VM at the time of dump?
[09:24] <fabbione> pitti: hmm interesting... did you grab the boot logs without splash/quiet?
[09:25] <pitti> fabbione: no, I didn't, it wasn't a vmware install; I can do some photos of it, though
[09:25] <fabbione> can you reproduce it reliably?
[09:25] <fabbione> if so it would be worth to look at it
[09:25] <cjwatson> harpreet: done
[09:25] <fabbione> i am kind of curious if initramfs gets confused by /boot not being on raid1
[09:26] <cjwatson> harpreet: oh, sure, I'll tell you when it's (about to be) available
[09:26] <pitti> fabbione: reproduce in the sense of booting the broken installation again, yes
[09:26] <harpreet> Colin: Thanks. I will check my email in an hour/two and test it out and send you a reply.
[09:26] <pitti> fabbione: I wasn't sure whether the mdadm d-i module would get along without a separate /boot
[09:26] <pitti> fabbione: on my production server I don't have a separate /boot, everything is on raid (but I use lilo there)
[09:26] <cjwatson> harpreet: FWIW, cjwatson@ubuntu.com is a perfectly good e-mail address for me (and actually I prefer it), so I got both your mails ;-)
[09:27] <pitti> fabbione: I already wondered whether md devices get proper uuids, but since it worked yesterday, I guess that's not it
[09:27] <fabbione> pitti: mdadm get along with root/boot on raid
[09:27] <fabbione> also in d-i
[09:27] <harpreet> Colin: Okay. I have figured you folks at Canonical prefer irc over emails. :-). 
[09:27] <fabbione> i used that setup for ages on sparc
[09:27] <harpreet> Colin: Thanks for the email. Have a good evening ahead.
[09:28] <fabbione> pitti: ok.. if you can get the data out it's worth looking at it.. it might be a corner case that's not so corner as we think
[09:28] <cjwatson> harpreet: IRC can be faster although e-mail is better if you need reliability
[09:28] <cjwatson> (IME)
[09:28] <pitti> fabbione: I prefer it as well (separate /boot is a big ugly imho), but it should still work
[09:28] <pitti> fabbione: no problem, let me boot again and do some shots
[09:28] <fabbione> pitti: yeps
[09:28] <harpreet> Colin: yep. So will test this stuff and let you know before I go.
[09:30] <keescook> micahcowan: the core file only contains the program memory and various maps, not the executable itself.
[09:31] <pitti> hi keescook 
[09:31] <keescook> hiya pitti!
[09:31] <micahcowan> Okay. Thanks keescook, pitti.
[09:33] <keescook> pitti: related to core files, shouldn't I be able to generate coredumps of installed binaries that have prior crashes?  (ulimit -c unlimited; sleep 120 &; kill -SEGV %1)  I was only able to do this after copying sleep to the local directory, or after removing /var/crash/*sleep*
[09:33] <keescook> ack
[09:38] <Solarion> any recommendations for finding out what's preventing the processor from reaching C3?
[09:47] <pitti> fabbione: http://people.ubuntu.com/~pitti/tmp/raid1-bootfail-1.jpg http://people.ubuntu.com/~pitti/tmp/raid1-bootfail-2.jpg
[09:48] <fabbione> pitti: thanks
[09:48] <pitti> fabbione: pretty messy, no fstab, some modprobe failure, I wasn't sure about the root cause
[09:48] <imbrandon> Keybuk, got a sec?
[09:48] <fabbione> pitti: no fstab?
[09:49] <fabbione> looks like something went foo bar with the installation 
[09:49] <fabbione> but totally
[09:49] <pitti> fabbione: nope
[09:49] <fabbione> probably you will never be able to reproduce that install
[09:49] <pitti> fabbione: oh, I probably will, wasn't too complicated
[09:49] <fabbione> ok
[09:49] <pitti> fabbione: since the actual fstab is on /dev/md0, which wasn't mounted at that time, I assume that the initramfs has a copy?
[09:50] <pitti> fabbione: I can access the initramfs from here
[09:50] <fabbione> pitti: initramfs doesn't need fstab
[09:50] <fabbione> it only needs to find root via root=/dev
[09:50] <pitti> fabbione: why does it try to access it then?
[09:51] <fabbione> pitti: it looks to me that initramfs things that root is mounted.. but the raid starts later
[09:51] <fabbione> can you check what's in /proc/cmdline?
[09:51] <pitti> fabbione: hmm, it started before IIRC
[09:51] <fabbione> pitti: note on pic 2
[09:51] <pitti> fabbione: I can reboot to find out, but should it differ from the options given in grub?
[09:51] <fabbione> md: md0 stopper
[09:52] <pitti> right
[09:52] <pitti> but that's after the failure
[09:52] <pitti> it was there before
[09:52] <fabbione> it might differ.. you don't need to reboot
[09:52] <pitti> fabbione: and on the initramfs prompt I can mount it
[09:52] <fabbione> yes it's after the failure only on printk
[09:52] <fabbione> do you have to wait 3 minutes to get there?
[09:52] <fabbione> or it prompts quickly into the initramfs?
[09:53] <pitti> fabbione: no, almost immediately
[09:53] <pitti> fabbione: kernel boots, checks for resume image, and wham, there I am
[09:53] <fabbione> pitti: then file a bug on initramfs.. it thinks that the raid is there but it's not
[09:53] <fabbione> pitti: otherwise it would wait for the raid to be formed
[09:54] <fabbione> i wonder if i can reproduce it in vmware
[09:54] <pitti> fabbione: hm, indeed it says 'md0 stopped' before the error as well
[09:54] <fabbione> but not tonight
[09:54] <fabbione> i am way too tired
[09:54] <pitti> fabbione: yesterday I did a vmware install and that went fine
[09:54] <fabbione> pitti: i think you need to talk to iwj too.. i have the feeling that what happens is:
[09:54] <pitti> fabbione: no worries; thanks for having a look
[09:54] <fabbione> raid is started with one disk
[09:54] <fabbione> initramfs finds the raid
[09:55] <fabbione> at the same time mdadm stops the raid to add the second disk
[09:55] <pitti> yay race conditions
[09:55] <fabbione> initramfs tries to mount a device that's there but not working
[09:55] <pitti> fabbione: I think I can find that out by watching the booting again
[09:55] <fabbione> and bam
[09:55] <fabbione> yes.. worth looking
[09:55] <pitti> fabbione: I'm pretty sure that the raid was up before, but the 'md0 stopped' seems to support your theory
[09:56] <fabbione> once the raid is mounted you can't stop it.. but it might be a race
[10:02] <pitti> fabbione: you are right
[10:02] <pitti> fabbione: it boots, md doesn't find anything, so md fails and initramfs says 'waiting for root fs'
[10:02] <pitti> fabbione: then it binds sda5, unbinds sda5, binds sda7 and then it's up
[10:02] <pitti> but at that time I'm already at the prompt
[10:02] <fabbione> pitti: you really want to talk to iwj about it
[10:03] <pitti> right, will do; thanks
[10:03] <pitti> fabbione: that would be a bug against what then?
[10:03] <fabbione> i think both mdadm and initramfs can be blamed.. but make it RC
[10:03] <fabbione> this is definetely a blocker
[10:03] <fabbione> not being able to boot on raid is a problem for servers
[10:04] <fabbione> pitti: file it against mdadm/initramfs.. affects also.. and make sure both Scott and Ian are aware of it
[10:04] <pitti> alright
[10:04] <fabbione> i need to get some rest
[10:04] <fabbione> 6am -> 10pm is a bit too much
[10:06] <pitti> fabbione: bug 79204 sounds similar
[10:06] <pitti> fabbione: good night!
[10:06] <ubotu> Malone bug 79204 in initramfs-tools "boot on md raid drives fails" [Undecided,Unconfirmed]  https://launchpad.net/bugs/79204
[10:07] <kofler> What package gives me /lib/modules/`uname -r`/build/ and all the files therein?
[10:07] <kofler> i.e. Makefile.
[10:08] <kofler> I can't seem to find any package that does this and I have all these installed: 2.6.17-11-generic
[10:09] <ogra> ogra@edubuntu:~$ ls -l  /lib/modules/2.6.20-14-generic/build
[10:09] <ogra> lrwxrwxrwx 1 root root 40 2007-04-12 18:56 /lib/modules/2.6.20-14-generic/build -> /usr/src/linux-headers-2.6.20-14-generic
[10:09] <kofler> Er, oops, I meant: http://phpfi.com/226155
[10:09] <ogra> ogra@edubuntu:~$ dpkg -S /usr/src/linux-headers-2.6.20-14-generic
[10:09] <ogra> linux-headers-2.6.20-14-generic: /usr/src/linux-headers-2.6.20-14-generic
[10:09] <kofler> Hmm.
[10:11] <kofler> Thanks ogra.
[10:20] <pitti> yay, my guess about the hanging desktop was right
[10:20] <pitti> cjwatson: still awake?
[10:21] <avoine> there is a expert of gtk that know why the library libgtk-directfb was remove in feisty?
[10:21] <seb128> pitti: what hanging desktop?
[10:22] <pitti> seb128: I did an expert install and with that it took a minute for gnome to even begin loading
[10:22] <seb128> avoine: likely because it was deprecated, we will make gtk+2.0 build it next cycle
[10:22] <seb128> pitti: lo not correctly configured?
[10:22] <pitti> seb128: now I tracked it down to a missing loopback iface
[10:22] <seb128> right
[10:22] <pitti> seb128: right
[10:22] <seb128> it's the classic reason
[10:22] <pitti> ooh, that bugs is already known and ages old, bug 9532
[10:23] <ubotu> Malone bug 9532 in netcfg "skipping over netcfg means that lo doesn't get configured" [Medium,Confirmed]  https://launchpad.net/bugs/9532
[10:23] <seb128> yeah, as said, that's the usual reason for gnome-session hanging
[10:23] <seb128> that and clock set 1907 or something
[10:23] <seb128> ah
[10:24] <seb128> known for netcfg as well
[10:24] <pitti> seb128: right, I often get that clock problem on my laptop when I managed to drain the battery while it was suspended :)
[10:24] <seb128> I should have a look at making gnome-session display a dialog or something when lo is not configured rather than hanging
[10:25] <pitti> cjwatson: unping
[10:26] <seb128> carlos: around?
[10:26] <pitti> great, so the two breakages on my expert install are filed, and the rest went without a hitch
[10:26] <avoine> thanks seb128 
[10:26] <seb128> pitti: when are we going to roll language packs for feisty?
[10:26] <pitti> seb128: right after RC
[10:27] <seb128> and no update after that?
[10:27] <pitti> seb128: today I fixed langpack-o-matic for software-properties, so tomorrow's dailies should be good
[10:27] <pitti> seb128: well, the usual monthly updates in -proposed/-updates
[10:27] <seb128> k, just to know until when we can get translation dones
[10:27] <seb128> k
[10:27] <seb128> not cool
[10:27] <seb128> update-manager is still not translatable
[10:27] <pitti> -able?
[10:28] <seb128> pitti: no .pot built during build, I fixed it yesterday but rosetta still didn't import it
[10:28] <seb128> so the "Static configuration" option is not translatable
[10:28] <pitti> seb128: I think carlos has to manually approve new templates
[10:28] <seb128> I chassed the missing french translations yesterday
[10:28] <seb128> he did approve it this morning
[10:29] <seb128> I'm wondering is rosetta is overloaded also or something
[10:34] <BenC> BTW, if anyone can test these kernels, I'd appreciate it. Release timeline depends on a good round of testing
[10:34] <BenC> http://people.ubuntu.com/~bcollins/kernels/feisty-release/
[10:34] <Solarion> a-HA
[10:34] <BenC> main thing is that your storage controller isn't regressed
[10:34] <Solarion> r300 and uhci_hcd are the culprits
[10:35] <BenC> e.g. you can get to all your static disks on IDE and libata driven controllers
[10:35] <BenC> as well as CDROM
[10:38] <Mirv> seb128: is the update-manager translation now in the general language packs instead of the kde packs, or is that a problem too, still?
[10:38] <seb128> Mirv: update-manager was no to the kde pack
[10:39] <dholbach> brb with new kernel
[10:40] <carlos> seb128: the queue is burning
[10:40] <seb128> carlos: like too many items there?
[10:41] <seb128> carlos: how come it happens today?
[10:41] <carlos> seb128: well, the problem we had yesterday delayed things a bit
[10:41] <carlos> seb128: and it's finishing with evolution .po files
[10:41] <Mirv> seb128: oh yes, it's the software-properties, and it's still in the kde pack. ok.
[10:42] <carlos> which are big and we are slow to import them
[10:42] <seb128> carlos: is there a public location with the queue order?
[10:42] <carlos> seb128: did you upload a new evolution version? or it's just a small fix done in Ubuntu?
[10:43] <carlos> seb128: https://translations.launchpad.net/translations/imports/+index?target=all&status=APPROVED&type=all&start=0&batch=75
[10:43] <seb128> carlos: when? we did upload GNOME 2.18.1 monday and tuesday
[10:43] <carlos> so that's a new evolution
[10:43] <carlos> let me check it anyway
[10:44] <carlos> if this is just a local update I can remove those entries from the queue
[10:44] <carlos> so we speed it a bit
[10:44] <seb128> carlos: is the queue going to catch up before tomorrow?
[10:45] <carlos> once evolution finish, it should be quite fast
[10:45] <Solarion> bug 105011
[10:45] <ubotu> Malone bug 105011 in Ubuntu "Unable to reach C3 or C4 powersaving states when uhci_hcd is in kernel, or radeon driver (r300) is in use." [Undecided,Unconfirmed]  https://launchpad.net/bugs/105011
[10:45] <carlos> but I'm not sure as we only have a couple of hours and each evolution .po file takes a lot of time..
[10:45] <dholbach> BenC: both my i386s are happy with the new kernel - the amd64 is not - it hangs in busybox and says 'job control stopped' or something - is there anything i can do to debug it?
[10:46] <seb128> BenC: new kernel works fine on my laptop (i386) do you want details on the config or something?
[10:46] <carlos> seb128: those .po files were uploaded yesterday
[10:46] <carlos> so I guess that was a local upload for an already existing version in Ubuntu, right?
[10:47] <seb128> carlos: you can ignore the po, we need to .pot though because we distro added a string
[10:47] <carlos> ok
[10:48] <BenC> dholbach: Can you boot without quiet/splash and see if you get some kernel messages?
[10:48] <dholbach> BenC: sure, hang on
[10:49] <Solarion> is emacs 22 going to be in gibbon?
[10:49] <tepsipakki> Solarion: it's released?
[10:49] <heno> BenC: works fine on an i386 install here (amd64 hw) with a sata drive
[10:50] <Solarion> tepsipakki: in the next few weeks
[10:50] <BenC> heno, seb128: thanks
[10:50] <tepsipakki> Solarion: sweet
[10:50] <Solarion> I had thought the betas would be in universe
[10:50] <carlos> seb128: everything removed except the .pot file
[10:50] <carlos> that would speed a bit the imports
[10:50] <seb128> carlos: cool, thank you
[10:51] <Solarion> http://lwn.net/Articles/229825
[10:52] <carlos> np
[10:52] <Solarion> heya seb128 
[10:53] <seb128> hi Solarion
[10:53] <Solarion> at least I figured out where my C3 went, I guess.  :/
[11:01] <dholbach> BenC: http://daniel.holba.ch/temp/PICT1777.JPG
[11:01] <mjg59> dholbach: Need the page before that
[11:01] <mjg59> dholbach: shift+pgup should work
[11:02] <dholbach> mjg59: ok, hang on
[11:02] <Arby> BenC: new kernel is fine here i386 with SATA drive
[11:04] <dholbach> BenC: http://daniel.holba.ch/temp/PICT1781.JPG
[11:05] <dholbach> mjg59: ^ too
[11:05] <mjg59> dholbach: Hm. Looks like even earlier than that :)
[11:05] <BenC> yeah, see if you can find some hpa resize stuff
[11:05] <mjg59> The section where it probes the hard drive
[11:05] <pitti> dholbach, mjg59: something like that for you as well? http://people.ubuntu.com/~pitti/tmp/14.23-boot-failure.jpg
[11:06] <BenC> looks like we might have an ABI bump to get the hpa revalidation right :/
[11:06] <mjg59> pitti: That's consistent with the most likely place for it to fail, yeah
[11:07] <pitti> interesting 64 bit numbers there
[11:07] <dholbach> ah yeah - it looks very much like what pitti has there
[11:07] <mjg59> pitti: Uninitialised memory, at a guess
[11:07] <BenC> yeah, not sure how that huge number came back from the native size read
[11:07] <pitti> BenC: on my second line there's another failed comparison which isn't that far off
[11:08] <mjg59> BenC: Urp. Worryingly, take a look at the hpa_sectors line on Pitti's grab.
[11:09] <BenC> mjg59: yeah, some odd numbers
[11:09] <dholbach> http://daniel.holba.ch/temp/PICT1783.JPG
[11:09] <pitti> BenC: do you need the geometries of my HDs?
[11:10] <BenC> same numbers on dholbach's
[11:11] <BenC> give me a minute to read this code again, see if anything is glaringly wrong with the logic
[11:11] <BenC>         /* if no hpa, both should be equal */
[11:12] <BenC> mjg59: that seems to be an incorrect assumption :)
[11:12] <mjg59> The alternative is that there's a bug in the reading code
[11:12] <pitti> 0xFFFFFFFFAFA19EB0
[11:12] <mjg59> pitti: What do you get if you mask off the top ones?
[11:13] <BenC> mjg59: is the hpa capability check just to see if the drive supports HPA, or to see if the drive has an HP area?
[11:13] <mjg59> Just checks the capability, I believe
[11:13] <BenC> 2147483647
[11:13] <mjg59> You'd need to check the ATA specs to be sure
[11:14] <pitti> 2946604720
[11:14] <mjg59> Yeah. Still not right.
[11:15] <pitti> BenC: your number is for dholbach's?
[11:16] <BenC> pitti: it's the hex you pasted minus 0xfffffff00000000
[11:16] <BenC> err, 0xffffffff00000000
[11:16] <pitti> BenC: hm, I calculated it twice and I get 2946604720, but anyway, it doesn't seem 'more' correct
[11:16] <pitti> BenC: I raise you to 3000000000!
[11:17] <Nafallo> haha
[11:17] <BenC> I think you've got me covered though
[11:17] <pitti> I originally hope that converting it to hex would reveal a sensible string or so
[11:19] <pitti> BenC: oh, 14.23 is already uploaded?
[11:20] <BenC> pitti: Yes, but that's a minor point...Tollef wont get a chance to do d-i/cd stuff till the morning anyway
[11:20] <BenC> pitti, dholbach: Can you test another kernel in about 30 minutes?
[11:20] <dholbach> BenC: no problemo - just tell me what you need
[11:20] <BenC> i386 or amd64?
[11:20] <dholbach> amd64 is the broken one
[11:20] <pitti> BenC: yes
[11:21] <pitti> dholbach: amd64
[11:21] <seb128> linux upgrade works correctly on my amd64 desktop using i386 distro as well
[11:21] <BenC> odd that it seems amd64 specific
[11:22] <heno> BenC: I have a stall on an amd64 as well now
[11:22] <BenC> mjg59: do you think something is borked in ata_tf_to_lba48() or ata_tf_to_lba()?
[11:22] <mjg59> Could well be
[11:23] <mjg59> Symptoms are similar to when that was just returning gash
[11:23] <BenC> we're using u64 everywhere, but those functions are twiddling bits
[11:23] <heno> so, erm, guys how do I boot from grub with usplash off
[11:23] <heno> ?
[11:23] <mjg59> Remove splash from the command line
[11:23] <Kmos> heno: sudo apt-get remove usplash
[11:23] <Kmos> :)
[11:23] <Nafallo> remove quiet
[11:24] <Nafallo> from the kernel commandline
[11:24] <heno> ah
[11:24] <BenC> heno: 'e' to edit the entry, 'e' to edit the command line, edit it, return and 'b' to boot
[11:24] <mjg59> Nafallo: No...
[11:24] <Nafallo> no?
[11:24] <mjg59> Though you'll want to do that as well
[11:24] <heno> ok
[11:24] <mjg59> Removing quiet doesn't imply a lack of splash
[11:25] <Nafallo> ah, right. that changed some time ago :-P
[11:25] <kylem> BenC, all the testing was done on amd64...
[11:25] <kylem> by me, that is.
[11:26] <BenC> I tested it on my amdt64 too
[11:26] <BenC> but it may be that only lba or lba48 is broken
[11:26] <BenC> hence, maybe we didn't hit the codepath
[11:26] <BenC> pitti's is at least showing LBA48
[11:27] <kylem> unlikely to be hitting the LBA codepath
[11:27] <BenC> hmm, so am I
[11:29] <BenC> pitti, dholbach: What's the actual size of your hd's?
[11:29] <pitti> BenC: hda is 10 GB in total, hdc is 120 GB (imho, let me check again)
[11:30] <kylem> LBA28 is (1 << 28) sectors * 512 bytes/sector
[11:30] <kylem> hmm, so conceivable your hda might be lba28.
[11:31] <pitti> BenC: sorry, hdc (mapped to sdb) is 160 GB
[11:31] <pitti> BenC: and hda (sda) is 13 GB
[11:31] <BenC> it's the larger one that shows the 0xffffffff00000000 craziness
[11:31] <pitti> [   28.005896]  ata1.00: 25408559 sectors, multi 16: LBA 
[11:31] <pitti> [   28.357830]  ata2.00: 312581808 sectors, multi 16: LBA48 
[11:32] <pitti> ^ with 14.22 kernel
[11:32] <pitti> [   28.404219]  SCSI device sda: 25408559 512-byte hdwr sectors (13009 MB)
[11:32] <pitti> [   28.547102]  SCSI device sdb: 312581808 512-byte hdwr sectors (160042 MB)
[11:32] <BenC> -14.23 is showing a native size of 25410672 for the 13G drive
[11:33] <pitti> BenC: right, and 25408559 is on the left-hand side, that's what .22 reports
[11:33] <kylem> BenC, uh. where's that patch again?
[11:33] <BenC> kylem: it's in git
[11:34] <BenC> the hpa on the second one is showing smaller for ata_hpa_resize, for the ata_same_device() it's showing some whacked crap
[11:34] <pitti> BenC: (left-hand side of the comparison in 14.23, I mean)
[11:35] <BenC> if it's smaller, that's fine, it gets ignored
[11:35] <BenC> guess the question is why are we showing some whacked number in ata_same_device() compared to the previous native read in ata_hpa_resize()
[11:36] <bf> after following instructions in Bug #52685, I know have none of hte duplicate files installed in /etc/Xsession.d, despite that fact that dpkg -L x11-common suggests otherwise.  Anyone have any ideas on why apt/dpkg claims to have installed the files yet they are not on disk?
[11:36] <ubotu> Malone bug 52685 in xorg "x11-common installs dup files" [Undecided,Fix released]  https://launchpad.net/bugs/52685
[11:36] <bf> s/know/now
[11:39] <kylem> BenC, i think the disk isn't set up the same when we call ata_same_device the second time...
[11:40] <kylem> probably why we were saving the n_sectors_boot...
[11:41] <BenC> ata_hpa_resize is called, then it does some device checks (LBA, ID's, etc), then it does ata_same_device()
[11:41] <pitti> "REBUILDING: 20070412 are REJECTED" -> ah, waiting for the new kernel? or any other problems?
[11:41] <BenC> not sure what could be disabled in between there
[11:41] <BenC> not sure why it is even recovering devices during this process anyway
[11:42] <BenC> pitti: new n-m too I think
[11:42] <pitti> ah, right
[11:45] <dholbach> BenC: [   21.666336]  SCSI device sda: 781422768 512-byte hdwr sectors (400088 MB)
[11:45] <BenC> pitti, dholbach: Give me a little bit to rebuild a new kernel
[11:45] <dholbach> BenC: np
[11:45] <pitti> just a little warning, if I drop off the network, n-m screwed me
[11:46] <Nafallo> pitti: :-)
[11:49] <mvo> I guess I'm a bit tired already, but why is 20070412 rejected as image?
[11:49] <pitti> mvo: new kernel + network manager, I assume
[11:49] <mvo> thanks
[11:49] <mdz> mvo: see Colin's post to -devel-announce
[11:50] <pitti> ok, booting again for testing Keybuk's new n-m, brgb
[11:50] <pitti> brb, even
[11:51] <mvo> mdz: thanks, my impression was that this mail was about 20070411
[11:52] <bf> any help with my apt/deb x11-common question? Why might the package skip installed conf files?
[11:53] <tepsipakki> bf: maybe just because of that.. conffiles are not removed on upgrade
[11:53] <tepsipakki> it should be done by the package scripts
[11:53] <tepsipakki> bf: I'll reopen the bug
[11:54] <bf> tepsipakki: is there a good way to force it to?  Might not help the bug, but I'd like to have a usable X.
[11:54] <tepsipakki> bf: oh, so you removed all of them :)
[11:54] <bf> well, that was the suggestion approach...
[11:55] <bf> err.  suggested
[11:55] <tepsipakki> no, it was suggested to remove the *xorg-common* files
[11:56] <tepsipakki> not *x11-common*
[11:56] <bf> "remove (sudo rm) all duplicate files (from both x11-common and xinit packages)"
[11:56] <bf> Did i misinterprete that?
[11:56] <tepsipakki> maybe,
[11:56] <tepsipakki> or maybe not
[11:56] <mdz> mvo: well, the issues apply to both unfortunately
[11:57] <mdz> bf: dpkg --force-confmiss
[11:58] <tepsipakki> right
[12:00] <bf> mdz: whee
[12:01] <bf> mdz:  that seems to have worked.  thanks.  i suppose I should update the bug report?
[12:01] <tepsipakki> bf: well, I updated it already
[12:01] <BenC> pitti, dholbach: Ok, built and uploading...should be 5 minutes
[12:01] <dholbach> rock on
[12:01] <pitti> yay
[12:03] <j1mc> i've been away today . . .  will an additional RC candidate image be released tomorrow? 
[12:06] <dholbach> BenC: same place? finished uploading?
[12:06] <BenC> almost done
[12:07] <pitti> BenC: can you give me the md5sum, just to be sure? my ISP's transparent proxy is sometimes funny
[12:08] <BenC> 9008b8fe14597a713fd0b8778b3638a8  linux-image-2.6.20-14-generic_2.6.20-14.23_amd64.deb
[12:08] <BenC> same location
[12:08] <pitti> uploaded?
[12:09] <tepsipakki> BenC: .23 worked for me
[12:09] <dholbach> pitti: looks like
[12:09] <dholbach> I have the same md5sum
[12:09] <tepsipakki> the previous one
[12:10] <BenC> tepsipakki: thanks
[12:10] <pitti> yep, md5sum matches
[12:10] <pitti> BenC: is there any way to disable pata at the kernel command line?
[12:10] <pitti> booting live CD and dpkg -i'ing older kernel in a chroot is a bit painful
[12:10] <BenC> pitti: you mean pata in general, or disable libata-pata in favor of ide?
[12:11] <BenC> oh, no, you are just at my mercy for this testing phase :)
[12:11] <pitti> hm, I don't know; having normal IDE drives back would do it, I guess
[12:11] <dholbach> BenC: looks GOOD!
[12:11] <BenC> yay!
[12:11] <dholbach> you ROCK!
[12:12] <BenC> no, I just reverted the patch back to kyle's original
[12:12] <BenC> I such because I broke it :)
[12:12] <BenC> suck