[00:57] <xnox> why a hunting party?
[00:59] <infinity> xnox: Because he really wants a bug fixed, apparently.
[00:59] <infinity> xnox: 1128597
[01:06] <xnox> infinity: well there is a branch attached to the bug already. It's just I happened to fix a few other things around there, while I was there and still testing will propose/merge tomorrow and we will do a ubiquity upload soon as we have a few fixes already.
[02:30] <phillw> xnox: I do hope you have a sense of humour, if not.. I apologise.
[03:20] <phillw> xnox: will these fixes be in the next re-spin?
[03:21] <phillw> aka auto cron buikd
[03:21] <phillw> *build
[10:25]  * cjwatson sorts out the image build failures
[11:14] <cjwatson> stgraber: OK, we'll start getting daily-checks mail about precise now too
[11:14]  * cjwatson remembered to deploy
[11:18] <cjwatson> slangasek,Riddell: are Kubuntu daily builds meant to still be commented out?
[11:18] <cjwatson> (and Kubuntu active)
[11:22] <Riddell> cjwatson: mm, no
[11:24] <doko> https://launchpad.net/ubuntu/+source/zipl-installer/0.0.22 doesn't have any build records. why?
[11:25] <doko> ahh, s390 only
[11:25] <cjwatson> feel free to remove that source; it won't be autosynced any mre
[11:25] <cjwatson> *more
[11:25] <cjwatson> since the autosyncer is smart about this now
[11:25] <cjwatson> Riddell: OK, re-enabled, thanks
[11:26] <cjwatson> doko: you'll probably need to explicitly remove from both raring and raring-proposed
[11:27] <Riddell> cjwatson: where should I start on getting nexus images for kubuntu-active made?
[11:27] <Riddell> poke ogra?
[11:27] <cjwatson> Riddell: talk to ogra, yeah
[11:27] <doko> cjwatson, so no need for black-listing?
[11:27] <cjwatson> doko: no
[11:27] <doko> ok
[11:27] <cjwatson> doko: most of the time you don't need to blacklist things any more
[11:27] <ogra_> Riddell, is end of the week ok ?
[11:27] <Riddell> ogra_: sure
[11:27] <ogra_> we have a dedicated builder for n7 images now, thanks to infinity, so we should now be able to do it
[11:29] <Riddell> ogra_: dedicated builder is a pandaboard or there's a nexus 7 in a datacentre somewhere?
[11:29] <ogra_> a panda indeed
[11:29] <ogra_> we dont want to have to send people to the DC to plug in power all 8h :)
[11:30] <ogra_> (since you cant charge while an USB disk is attached)
[11:30] <cjwatson> doko: for new sources, auto-sync avoids (a) anything that's blacklisted (b) anything already in NEW (c) anything not built on any architecture in the target distroseries (d) anything where an equal or higher version was already removed from the target (e) anything that was previously published in the target at a lower version (it mails me about these for manual resolution)
[11:30] <cjwatson> this was enough intelligence to let it run from cron
[11:36] <apw> ogra_, there are power pads for charging on the side of the N7, not that i know which of the 4 is which
[11:37] <apw> infinity, you have debian-installed blocked, is that for the .2 or something else
[11:54] <ogra_> apw, thats a lid switch, not power
[11:55] <apw> ogra_, there are 4 pads on teh side for the dock, they let it charge and pipe out audio according to the dock spec
[11:55] <ogra_> oh, i thought it was onlt the lid stuff
[11:55] <apw> the dock says it charges and does audio and the pictures show it sitting on the side there, nothing in the USB
[11:56] <apw> so i recon it has to have power in there to make any sense
[11:56] <ogra_> well, i know that the "jackets" use these connectors for lid open/close events
[11:57] <ogra_> anyway, happy to hear it can charge through it too
[12:00] <ogra_> cjwatson, the meta upload will fix the libO uninstallability ?
[12:01]  * ogra_ was wondering about it 
[12:22] <cjwatson> ogra_: Yes
[12:23] <ogra_> great
[15:52] <didrocks> cjwatson: do you know if the hud it still publishing? I don't see it on the page, seb128 NEWed a little bit more than an hour ago
[15:53] <cjwatson> didrocks: hmm, good question, it might have been caught by the librarian outage ...
[15:53]  * cjwatson contemplates how to check
[15:55] <didrocks> cjwatson: we'll be able to resync it from the ppa in case it's needed :)
[15:55] <cjwatson> didrocks: Ah, it got rejected due to the librarian outage, and I can't resurrect rejected syncs.  Indeed, can you just sync it again?
[15:56] <didrocks> cjwatson: sure sure, thanks for checking!
[15:56] <cjwatson> It must have landed in accepted briefly and then been booted out
[15:56] <seb128> didrocks, perfect timing :p
[15:56] <didrocks> seb128: didn't I tell you that sometimes, I feel like the black cat? :)
[16:03] <didrocks> ok, sounds go have created the metadata in the launchpad page now, thanks cjwatson. I'll promote it in main as soon as I can see the published version (the MIR is acked)
[17:51] <cjwatson> Whoops.  Sorry for Lubuntu image build failure, fixing
[17:52] <cjwatson> bzr DTWT when resolving conflicts in an executable
[18:13] <cjwatson> ... and again
[18:14] <cjwatson> Looking happier now
[18:28] <Riddell> kubuntu 12.04.2 works on a secure boot system even though we didn't add any secure boot bits to it, 12.04.1 fails, that's unexpected
[18:33] <infinity> Riddell: Really?  That's beyond unexpected, if it's an SB system that will only boot things signed with the MS key.
[18:35] <xnox> Riddell: it may as well be a secure boot machine, maybe it's just UEFI machine where updated UEFI works and older one doesn't?
[18:36] <xnox> are you sure secure boot is fully enabled and it doesn't e.g. fallback to normal uefi.
[18:37] <Riddell> xnox: it does say enabled in the firmware settings.  it also doesn't boot 12.04.1 with the message "secure boot failure"
[18:37] <xnox> ok. i rest my case, sir.
[18:38] <xnox> Riddell: well, that mean kubuntu has secure boot now \o/
[18:39] <Riddell> 12.04.1 http://www.flickr.com/photos/jriddell/8492018551
[18:39] <Riddell> I'm not sure I trust a single data point on this issue
[18:46] <cjwatson> Riddell: You may find that the image itself boots but it (a) doesn't have a signed kernel so won't work on all systems (b) won't install in such a way as to make a bootable installed system
[18:46] <cjwatson> I think there's a shim + signed GRUB there unconditionally
[18:46] <cjwatson> But only on the live image
[19:05]  * cjwatson fixes another fairly arcane explosion due to the nusakan upgrade
[19:06] <cjwatson> (Slight lzcat interface change)
[19:45] <plars> cjwatson: I think we have a problem with the precise netboot images, they don't seem to be installing the 3.5 kernel.
[19:46] <xnox> infinity: ^
[19:48] <infinity> plars: Which ones?
[19:49] <plars> infinity: current ones at http://archive.ubuntu.com/ubuntu/dists/precise-updates/main/installer-amd64/20101020ubuntu136.9/images/netboot/
[19:49] <plars> infinity: unless I'm looking in the wrong place, that should be the latest right?
[19:50] <infinity> plars: Note the quantal-* directories.
[19:51] <infinity> plars: images/netboot will always be the precise kernel, images/quantal-netboot is the lts-q kernel.
[19:51] <cjwatson> Yep, what he said.
[19:51] <plars> infinity: ok, so that's intentional... hggdh we need to teach cobbler to cope with that somehow though
[19:52] <hggdh> ah
[19:56] <cjwatson> I did the same for the backports to lucid.
[19:56] <cjwatson> So if it's broken now it was probably broken then too ...
[19:56] <infinity> I doubt they cobbled much with lucid.
[19:59] <plars> cjwatson: yeah, it's just some automated jobs for kernel sru - not sure whether they existed that far back.
[20:00] <plars> we were not seeing the 3.5 kernel come in with the new precise tests, but the iso seemed right, but we didn't notice that quantal directory
[20:07] <cjwatson> plars: good sign that you noticed, anyway :)
[20:10] <infinity> plars: Check.  And for SRU testing, of course, you need to test both.
[20:11] <infinity> plars: So, good thing we provide two. ;)
[20:19] <plars> infinity: yes, that's the idea that we want to make sure kernel SRUs are testing for both
[21:09] <nuclearbob> it looks like the most recent daily live images have the initrd under casper/initrd instead of casper/initrd.lz  Is this intentional, and will the change be permanent?
[21:13] <cjwatson> nuclearbob: No, I'm already in the process of fixing that bug
[21:13] <cjwatson> It was a side-effect of upgrading nusakan to precise
[21:13] <nuclearbob> cjwatson: cool, thanks
[21:14] <cjwatson> Was wondering whether I'd manage to fix it before somebody noticed ;-)
[21:17] <veebers> cjwatson: would you happen to have a bug number for that?
[21:17] <cjwatson> I didn't bother
[21:18] <cjwatson> Respins are already in progress
[21:18] <cjwatson> And only a handful of images (albeit including Ubuntu desktop) were affected
[21:21] <veebers> cjwatson: cool, thanks. How long does a respin take?
[21:23] <cjwatson> Minutes
[21:23] <veebers> awesome
[21:24] <cjwatson> In this case, anyway, since I don't need to rebuild the live filesystem, just the ISO9660 container
[21:29] <cjwatson> nuclearbob,veebers: 20130220.2 should be visible shortly; I've verified that its file list is back to normal
[21:30] <nuclearbob> cjwatson: awesome, thanks
[21:30] <veebers> cjwatson: noice, thanks
[21:30] <cjwatson> (For the record, the problem was that lzcat's command-line syntax got stricter from lucid to precise, confusing cdimage)
[21:31] <nuclearbob> that's inconvenient
[21:32] <xnox> (well it used to be lzma but now it's xz that provides it ;-) )
[21:35] <cjwatson> Yeah, that wasn't really the issue I think
[21:35] <cjwatson> Using -S '' was always a bit of a hack
[21:35] <nuclearbob> hmm, I think I'm using -S ''
[21:35] <cjwatson> So I wasn't desperately surprised to have to prod it
[21:35] <cjwatson> lzcat: : Invalid filename suffix
[21:37] <cjwatson> Really lzcat has a stupid interface and shouldn't require that at all, because it doesn't use the target filename gained by stripping the suffix
[21:37] <cjwatson> So (at least in raring) you can actually just pass any string as a suffix, AFAICS
[21:38] <nuclearbob> that's handy
[21:38] <cjwatson> As it happens I used a chunk from the end of the filename in question, but I think that was overengineered
[21:54]  * antarus hates releases
[21:55] <antarus> all the bugs I fixed ended up failing in release verification on weird edge cases :/
[23:20] <slangasek> why does https://launchpad.net/ubuntu/raring/+source/libreoffice/1:4.0.0~beta2-0ubuntu1 say it's published in quantal?
[23:21] <cjwatson> *blink*
[23:22] <cjwatson> Oh, I think that means libreoffice in general is published in quantal and raring
[23:22] <cjwatson> Daft presentation
[23:23] <cjwatson> Hm, but https://launchpad.net/ubuntu/raring/+source/base-files/6.12ubuntu1 doesn't do that
[23:24] <cjwatson> It's from a join between SPR and SPPH, so maybe there are copies into a PPA that are confusing it?
[23:25] <cjwatson> I don't know, very confused
[23:26] <antarus> I'm glad I'm not the only one that gets confused
[23:27] <slangasek> cjwatson: ok
[23:29] <slangasek> cjwatson: so I'm looking at this in the context of libreoffice-presenter-console, which you appear to have dispatched this morning
[23:29] <stgraber> cjwatson: https://launchpad.net/~guido-iodice/+archive/testing2/+sourcepub/2986741/+listing-archive-extra ?
[23:29] <slangasek> it's causing update-manager to display the dreaded 'partial upgrade' dialog and I'm not sure why
[23:30] <stgraber> cjwatson: basically someone copied the raring build straight to a quantal PPA, then removed it. Could that cause what we're seeing?
[23:31] <cjwatson> stgraber: Good catch, thanks, that would do it
[23:31] <cjwatson> slangasek: Yes, I did - is there still some open fallout from that?
[23:31] <slangasek> cjwatson: it's causing update-manager to display the dreaded 'partial upgrade' dialog and I'm not sure why
[23:31] <stgraber> cjwatson: google with site:launchpad.net is handy for that kind of search (launchpad's own search engine, not so much ;))
[23:32] <cjwatson> slangasek: Curious, an apt-get dist-upgrade this morning was pretty clean for me
[23:32] <cjwatson> I don't suppose it's marked for manual install or something?
[23:34] <slangasek> cjwatson: well, apt-get dist-upgrade is perfectly happy removing the package
[23:34] <slangasek> but update-manager is more particular IIRC
[23:35] <slangasek> cjwatson: remind me how to check the current mark status?
[23:36] <cjwatson> apt-mark showauto | grep
[23:37] <cjwatson> some day soon I suppose I'm going to have to actually understand how u-m works
[23:38] <slangasek> the package is indeed marked manual
[23:38] <slangasek> which, actually, is the expected behavior since it's an ubuntu-desktop Recommends: (or was)
[23:44] <cjwatson> It might need to be quirked in u-m
[23:45] <cjwatson> IIRC there's a list of packages that have been removed ...
[23:45] <slangasek> hmm yes
[23:45] <cjwatson> But I thought that was mainly for release upgraes
[23:45] <cjwatson> +d
[23:45] <cjwatson> I don't know whether it comes into play for normal u-m runs