[01:22] * skaet --> EOD [01:33] anybody know why there is no Quanta build on July 25 ? [01:34] babyface_: we're working on alpha 3, the dailies are on hold until it's released [01:34] babyface_: alpha 3 images available for testing on the ISO tracker [01:35] micahg, ack. thanks [01:57] skaet: the discussion in scrollback is principally about how and why the fix ogra applied this morning isn't correct; but we knew already that it wasn't correct when he uploaded it [01:57] skaet: and it doesn't appear that there were any further problems with flash-kernel once installed [02:01] indeed, I see a 'pass' on server armhf+omap4 [02:02] Yeah, it works right now, but that version of f-k should be reverted, since it breaks upgrades. [02:02] So... [02:02] We either fix the root problem, or we're shipping a broken f-k just fot A3 to "fake" the test results? [02:03] the wrong f-k is already in the archive; there's no burning need to fix that between now and tomorrow [02:04] No, I'm just not sure it's doing us any favours to have a broken f-k in the archive just for the sake of having a pass on an image, when it's something we wouldn't actually want to ship. [02:05] I think it's more the sake of having installable images, A3 or otherwise [02:06] can we just get to the bottom of the /dev issue, and /then/ revert it? [02:06] That does seem vaguely sane, yes. [02:06] ok [02:06] then I don't think there's anything here that jeopardizes A3 [02:07] Obviously, something (bind-)mounts /dev into /target later in the install process, or all the bootloader installers would explode, so probably just need to hunt that down and judiciously cargo-cult it into live-installer. [02:07] Or base-installer. [02:08] Or something. [02:08] * slangasek nods [02:08] Time to do a full Debian d-i checkout of doom, I think, so I can grep more effectively. [02:27] slangasek: Oh, it could just be a question of moving some code 20 or 30 lines up in live-installer.postinst. :P [02:27] neato [02:28] I'll have to look over logs again and make sure that's sane, but if it is, I'll commit and upload something later tonight. [02:29] Maybe after I've had some breakfast. [02:29] I recommend the skirt steak with chile relleno [02:29] Or whatever meal it is when you realise you've forgotten to eat for ~20h. [02:29] * micahg wonders where infinity is that it's breakfast now [02:30] micahg: you're assuming timezones vaguely apply to infinity ;) [02:30] micahg: he's copying your meal schedule, obviously ;) [02:30] heh [02:30] micahg: I'm infinite, which places me in every timezone. And every restaurant. [02:30] No wonder I'm getting fatter. [02:31] micahg: (PS: I'm raiding your fridge right now) [02:31] infinity: heh, joke's on you then :) [02:31] Yeah, say. Where do you keep the bacon cheeseburgers in this joint? >:( [02:32] unless you like tofu and rice tortillas [02:32] * micahg thought webkit on armhf would be done by now, but it'll take another 12 hours [02:32] I need to figure out how to incorporate peanuts into bacon cheeseburgers, so that they become not just religiously offensive, but also deadly. [02:33] eew [02:33] skaet: ^ boots, installs and still boots, good enough for a3 unless something major shows up. [02:33] infinity: well, for some people, just being around peanuts is deadly [02:33] There's something invigorating about eating food that you know can kill a rather large segment of the population. [02:33] that's why I subsist on a diet of rattlesnake meat [02:34] infinity: does the death have to be immediate? [02:34] Rattlesnake only kills when it's alive. :P [02:34] (Also, it's pretty tasty) [02:35] micahg: I dunno. Anaphylaxis is pretty dramatic. "I got a tummy ache and then died 30 years later from a steadily imperfect diet" is somehow less cool. [02:36] you could invent supergluten [02:36] umm, we already have that [02:36] we have supergluten? [02:36] Wonderbread? [02:36] that's just superlative quantities of regular gluten [02:36] infinity: you could just be muslim for a month, fast during the day, feast at night [02:36] I want supergluten [02:37] slangasek: http://www.adm.com/en-US/Milling/Glutens/Pages/Supergluten80.aspx [02:37] made from horses [02:37] jbicha: I'd need a better beard. [02:37] heh [02:37] micahg: that's beautiful :) [02:37] Color: Light tan [02:37] I wonder if it has a theme song. [02:38] Right, enough of this silliness. Breakfast calls. Then debian-installer redux. [07:19] jamespage, hi James, are you around? [08:07] babyface_, here now [08:09] jamespage, some test failed in Jenkins on quantal server - minimal virtual http://10.98.0.1:8080/view/Quantal/view/ISO%20Testing/job/quantal-server-amd64_minimal-virtual/72/#showFailuresLink [08:09] jamespage, could you help to have a look on it ? [08:09] babyface_, minimal virtual is broken atm - there is already a bug and it should be release noted for alpha3 [08:10] jamespage, ack. could you give me the bug number ? [08:10] babyface_, bug 1028453 - its linked off the ISO qa tracker [08:10] Launchpad bug 1028453 in ubuntu-meta "Quantal Ubuntu Server minimal install oversized" [High,New] https://launchpad.net/bugs/1028453 [08:11] jamespage, got it! thanks! [08:11] jamespage, have a nice day! ;) [08:11] np [08:37] slangasek: Hrm, there might be a reason the live-installer postinst is broken the way it is, given that the setup_dev waypoint was intentionally commented out. I'm not sure I feel confident trying to fix this blind before A3 and without consulting Colin, since I don't see much of a window to really test what it breaks to make the (seemingly) obvious fix. [08:37] infinity, well, i would rather have looked for the breakage in in-target or the way its called [08:37] iirc it can mount proc sys and dev [08:38] There's no "in_target" for lots of this. [08:38] oh my [08:38] k [08:39] ah, and the doc say in-target only automounts proc and sys [08:39] *docs [08:42] Also, if we fix d-i to DTRT here, we need to scour everything that thinks it doesn't. :P [08:42] (like the flash-kernel-installer postinst, for instance...) [08:42] fun :/ [08:43] This may be why setup_dev is commented out. It might have cause issues in other packages trying to do it themselves. Dunno. [08:43] Anyhow, I'm going to try to sleep while it's dark. Cause I hear that's what normal people do. [08:43] Wish me luck. :/ [08:44] * ogra_ does so [08:47] * ogra_ finds it funny that none of these scripts check if /dev is already bindmounted ... i wonder with how many /dev mounts you actually end up in a normal install [10:16] infinity: not sure if you're already on it, but lucid kernel packages need override fixes (#1027831) === yofel_ is now known as yofel [11:39] apw, infinity, ogra_ ./install/netboot/ubuntu-installer/i386/pxelinux.cfg/default file failed MD5 checksum verification for A3 quantal-server-i386 [11:40] occurring when i 'check disk for errors' [11:40] netboot "check disk for errors" ?? [11:40] what disk would it be supposed to check on a netboot install ? you dont have any source media [11:41] its check disk for defects in usb installation [11:42] oh, sorry, i was mislead by the path [11:42] file a bug against debian-installer [11:42] ok will do thanks :) [11:54] hmm [11:54] Riddell: you don't want ppc images? [12:42] Laney: I don't but others in the kubuntu community take a "if someone tests them why not" attitude [13:01] \o [13:01] * ScottK is one of those people. [13:01] So far we got testers, some at the last minute, for at least the final images. [13:03] ok then [13:03] perhaps you could edit the manifest: https://wiki.ubuntu.com/QuantalQuetzal/ReleaseManifest [13:03] also to confirm that you want alternates [13:11] Is that the Alpha 3 manifest or the one for the release? [13:11] I think in the end we don't want alternates, but that's dependent on Ubiquity work that's not done yet. [13:12] Riddell: Mind if I add the PPC ones? I'm fine with putting myself down as sign off contact, so you needn't be bothered. [13:13] ScottK: go ahead [13:14] AFAIK there's only one manifest [13:15] you're supposed to decide what you want finally by FF [13:20] Added powerpc. [13:20] The alternates are already listed with an appropriate note, so I think we're good. [13:21] yep. I thought you'd decided to keep them, but if it's correct alreay then more's the better. Thanks. [13:30] good morning Laney [13:30] howdy [13:31] * skaet goes to stare at the iso tracker, and pad [13:32] can't see anything that's going to move [13:32] good luck with that, and i hope you'll win the no-blinking competition [13:38] knome, apparently so, looks like Xubuntu didn't get respun while I slept [13:38] skaet, awwh [13:38] webkit still isn't finished [13:39] if armel didn't get restarted then it probably would have been this morning [13:40] knome, did you want the desktop ones to go out at least? [13:40] ie. are they usable enough for developers? [13:41] * skaet knows alternates are no hope until the respin [13:41] um, you should ask micahg... [13:41] :) [13:41] or maybe pleia2 [13:42] testing report published on the wiki https://wiki.ubuntu.com/QATeam/ReleaseReports/QuantalAlpha3TestReport [13:43] knome, ? thought you were xubuntu lead this cycle... did I miss something? [13:43] jibel_, thanks! :) [13:44] skaet, i am, but unfortunately i'm not quite knowledgeable with the current situation - i've been away quite a lot lately :) [13:44] Laney, am not seeing ScottL/scott-work online, the Ubuntu Studio images haven't been tested at all, can you help me figure out if someone's going to test them, or if they're a no ship for A3. [13:46] I'll go ask in their IRC channel [13:47] skaet: Laney: I intend to give either one a spin in a few hours. [13:48] astraljava: Oh OK. Do you fancy asking in the channel then to see if anyone else wants to do it? It would be good to have results sooner rather than later. [13:48] astraljava, ok, if you make it in before the announce we can include them. Won't hold it up though at this point. [13:48] Laney: Just did, waiting for answers. [13:48] thanks [13:48] skaet: When were you planning to announce? [13:50] astraljava, trying for before EOD in london, since we need one of the folks there to help with the web side. [13:50] skaet: Excellent, since I'm two hours ahead, it'll give me some time. [13:52] astraljava, key word in there was "before". :) we'll leave it late binding then. [13:52] knome, can you make sure that the Technical Overview has the key feature changes listed then for Xubuntu. [13:52] Sure. [14:00] skaet: so I can mark as ready the ones which have all mandatory tests done and nothing concerning in the rport? [14:00] Laney, yes [14:00] thanks. :) [14:04] infinity: ok - leaving live-installer as-is for now, then? [14:09] slangasek, we could probably just add a bindmount right before the dpkg-reconfigure console-setup in post-base-installer.d/25live-installer-console-setup [14:09] that would only wrap the one call that breaks [14:10] which sounds like another kludge? [14:10] well, at least it wouldnt break other scripts if adam thinks generally having it mounted is bad [14:12] we already have a kludge in place; we don't need more kludges, we should understand what should really be happening and upload that [14:13] laney: gst-plugins-bad0.10 is still on images ATM :( [14:14] ogra_, can you add to the https://wiki.ubuntu.com/QuantalQuetzal/TechnicalOverview/Alpha3 a note on how to bring up arm server in the known issues part. [14:15] skaet, sure, np [14:16] ogra_ , thanks. will you be reverting those last two fixes, so when we turn on the dailies again, we don't cause problems with the hacks? [14:18] micahg: woe indeed, but it won't cause any problems at this point [14:18] skaet: what about images with only one mandatory test uncompleted? [14:18] like kubuntu which are missing wubi [14:19] Riddell, ScottK, ^ what do you feel comfortable letting get released on these. [14:19] skaet: yes I'm fine with that [14:19] Personally, I call "missing wubi" a feature. [14:19] skaet, that will make the images uninstallable until another fix is added, if thats fine i can upload to proposed immediately [14:20] but I guess I should work out how to get wubi tested before beta [14:20] ScottK: you don't like it? [14:20] Riddell: Since all it does now (AIUI) is point you to the web site to download the real wubi, it seems somwhat pointless. [14:20] Also, I've never been convinced it's worth all the trouble it causes. [14:21] It could be that more people use it than I suspect though. [14:21] http://testcases.qa.ubuntu.com/Install/DesktopWubi [14:21] it's that [14:21] Laney, at this point, since its Alphas, if the leads recommend it as ok, and if there's been some testing we'll let it go out with the alphas, things start to tighten up for the betas, since they're not targetted at developers. [14:22] ok [14:26] Daviey: Server OK from your side? [14:28] make sure to not release the other side then :) [14:28] Daivey: Cloud Images are good to go [14:42] skaet: so, webkit is done on all archs except for armel, that'll be another 7.5-15 hours, since it's alpha3, I don't know if it's worth it to play around with the builds to get alternates [14:43] are we on track to get rid of the alternates this cycle [14:44] micahg, last I heard from slangasek, yes. But he'll know more than me. ;) [14:45] micahg, so do you want to let Xubuntu desktop out as is? and put some comments in the release notes? or punt on shipping for this release? [14:45] skaet: I think it's fine for alpha with release notes [14:46] micahg, ok, can you add now, and I'll go mark up the iso tracker. [14:46] * micahg thought pleia2 and knome did that already [14:49] vanhoof, netboot for highbank hasn't been tested, any data from your team? [14:50] micahg: I think skaet means that the release notes are still TODO [14:50] albeit with astraljava's name [14:54] thanks Laney, yes micahg, release notes about the issues still left (in known issues section). [14:56] saket: Cloud Images are ready for release [14:58] thanks utlemming :) [15:04] ogra_: how do I reinstall a panda after being installed? if I take out the memstick (/dev/sda) I get dropped to busybox [15:05] you write the image newly to SD indeed [15:05] at least the first partition [15:07] * skaet dropping i386 Server from list, see release notes. [15:09] oh, I read that as saying that they wanted it in A3 [15:09] but were declaring it unofficial [15:12] Daviey, arsoales - MAAS (juju and server) tests haven't been marked as run for the amd64 server image. any data sitting elsewhere? [15:12] Daviey, do you want us releasing i386 server image or not as part of A3? seems to be some confusion. [15:15] Is arsoales intended to be a pun? [15:16] *sigh* typo on my part [15:16] arosales, ^ [15:16] hahaha [15:18] skaet: I'll follow up wth roaksoax. I think he had been doing some work around this area. I will confirm. [15:19] thanks arosales [15:33] skaet: re dropping i386 for server for A3 let go ahead and proceed (pending any objections from Daviey) in order to get feedback from users. [15:33] thanks arosales [15:33] proceed with having it? [15:34] proceed with dropping it [15:34] sorry proceed with dropping i386 for A3 [15:34] dropping it in order to get feedback doesn't seem to compute [15:34] except to collect complaints? [15:35] Laney: I believe Daviey announced on the -devel list, would also like to get feedback for A3 if folks are missing it [15:35] ok then [15:35] so yes to get feedback on it being a large disruption to users [15:35] rather than waiting for the beta [15:35] at least that was the thought [15:36] fair enough [15:36] so for server, just waiting on remaining tests for amd64 [15:36] not sure about desktop armhf+omap4 [15:36] ogra_: any thoughts? have you tried it? [15:39] Laney: trying the manual partition, just had the panda turn off DVI and completely lock, will reboot and try again [15:39] * arosales is posting and update for MAAS testing. MAAS in quantal that is currently broken pending https://code.launchpad.net/~andreserl/maas/bzr777_packaging_updates [15:39] sentence or two on the release notes would be welcomed [15:39] Laney, yes, i had logged my results too [15:40] good to go [15:40] arosales, please add it to the known issues for server then too. [15:40] skaet: will do [15:40] hggdh: fun [15:40] * hggdh is starting to hate pandas [15:41] hey, they are endangered animals ! [15:41] ogra_, some of the manditory tests haven't run for arm desktop, hggdh is looking into one right now, but there is another without results, so not so sure its good to go... [15:41] hggdh: try feeding them bamboo [15:41] skaet, ah, well ... imho its good to go, but if you guys insist :) [15:42] I think cyanide broth would be better [15:43] ogra_ feel free to take it up with QA if you think the tests shouldn't be mandatory ;) [15:43] seriously, its an alpha release, if there is *a* way to install that works it should be fine, even though if others might have issues [15:44] ogra_ its a matter of warning folks too about known issues, which the manditory tests do expose. [15:45] for aplhas i'm personally happy if accepting the defaults results in a working system, everything beyond that are nice to have, bug should be filed and fixed but that shouldnt be a blocker [15:48] ogra_ if the test would result in a corrupted system, probably want to know that and then make a decision to release or not. [15:49] if it did that i would release note it :) but definitely not block the release ... for a beta i would be more strict ... [15:50] anyway, not my call :) [15:50] hggdh: for bug 1029201, is that blocking a milestone? Should the fix go to quantal or to quantal-proposed? [15:50] Launchpad bug 1029201 in qemu-kvm "qemu-system-x86_64 crashed with SIGSEGV in virtio_pci_mask_vq()" [Critical,In progress] https://launchpad.net/bugs/1029201 [15:53] the release will happen pretty soon, so there won't be a freeze for much longer [15:53] hallyn: it is only blocking my usage of qemu-kvm; I think -proposed is OK [15:54] hggdh: thanks [15:54] Laney: how soon? a few hours? [15:55] at most a couple I guess [15:56] Laney: ok, thanks. i'll wait then rather than cause overhead with having to copy out of proposed [16:03] skaet: Laney: there is hope for manual partitioning on omap4 desktop, finally installation is proceeding [16:03] thanks hggdh. :) [16:07] skaet, micahg: we are, shall we say, doing our best to get rid of alternates this cycle; but none of the code has landed yet so it's a bit early to say yes/no [16:15] slangasek: how does this look? https://bugs.launchpad.net/ubuntu/+source/update-manager/+bug/979661/comments/29 [16:15] Ubuntu bug 979661 in update-manager "oneiric to precise: debconf: unable to initialize frontend: Gnome and falls back to Dialog" [Critical,Fix committed] [16:15] or anyone [16:16] astraljava, are you going to be able to get to ubuntu studuio amd64? or should we just release with i386? [16:29] skaet: If you can give me 1.5 hours, I can confirm, or is that too late already? [16:32] astraljava, go ahead and confirm. we're close, and its likely to be ok, if the other one is. [16:33] skaet: Actually, it seems that it is just about done. Please for a few more minutes for resukts. [16:33] astraljava, :) I like that answer. you have your minutes [16:33] Please wait*, funny how I can't multi-task efficiently, was on the phone. [16:36] No one can multi-task efficiently. It's a myth. [16:37] server has all tests run and those not passing have bugs, so marking it ready now. [16:38] skaet: Studio can be marked ready now. [16:39] thanks astraljava. [16:39] ScottK: True, I've seen some footage on tests for that. People drive like idiots on tracks when speaking on the phone, even with hands-free. [16:39] skaet: Have you received release notes from Studio, from Scott L? [16:40] astraljava, https://wiki.ubuntu.com/QuantalQuetzal/TechnicalOverview/Alpha3 has them. I think someone else put them in this time. [16:41] * skaet just happy they're there. ;) [16:41] Well, me too, obviously. These new contributors seem to rock plenty. :) [16:42] indeed. :) [16:45] bdmurray: looks good to me [16:52] slangasek: Yeah, I think leaving it be until post-A3 is sane, and then we can experiment a bit with what the right answer is. [16:54] Does anyone other than the not present cjwatson have experience unbreaking MoM? It died a week ago. [16:55] ok, I think I'm done making edits for xubuntu on the release notes [16:55] thanks pleia2 :) [16:56] skaet: desktop armhf-omap4 is good to go [16:56] thanks hggdh [16:57] skaet: ack, dropping i386 for this milestone [16:58] * ogra_ hugs hggdh [16:58] thnaks for the effort ! [16:58] ogra_: welcome :-) [16:58] * hggdh hugs ogra_ back, and jots down the need for a beer next UDS [16:58] ++ ! [16:59] hggdh, arent you at the QA sprint next week ? [16:59] ogra_: yes. You will be there? [16:59] yeah,m we can have it there then :) [16:59] s/next UDS/next week/g [17:00] :-) [17:09] Is it okay for me to apporve an upload to -proposed that I sponsored? [17:09] bdmurray: no [17:10] you can verify it though :) [17:11] Actually I'm not positive I uploaded it... [17:12] bdmurray: you should have an e-mail saying waiting for approval if you did [17:12] That or check the signature on the changes file. [17:13] that would be more reliable than e-mail :) [17:13] this? http://launchpadlibrarian.net/110566856/parted_2.3-8ubuntu5.1_source.changes [17:13] there is no signature [17:15] Or the .dsc [17:15] I don't think it gets stripped from that. [17:15] Signature stripping is a security feater. [17:15] (but it only applies to .changes) [17:16] bdmurray: hey there. I just noticed that lxc is kind of broken once the fix in bug 974584 lands... I'll now push a matching lxc SRU to precise-proposed so you can accept both at the same time and both can land in -updates in the same batch (or container creation will fail) [17:16] Launchpad bug 974584 in sysvinit "Semaphores cannot be created in lxc container" [High,In progress] https://launchpad.net/bugs/974584 [17:18] bdmurray: (the fix will also land in quantal later today as part of a bugfix uploaded by hallyn) [17:42] bdmurray: lxc fix uploaded now. Would greatly appreciate if you could review sysvinit and lxc today as creating quantal containers on precise is currently broken (that's breaking my daily QA tests) [17:43] stgraber: I'll try [17:44] bdmurray: thanks [17:54] slangasek: is a source daily something we should spin? [17:55] publish-images gives me a line to publish them, but apparently it does not exist [18:01] Daviey, stgraber ^ any thoughts? [18:01] I'm wondering exactly what Laney is trying to do and what he's running :) [18:01] so you run publish-images.py [18:01] and at the end it gives you this: for-project ubuntu publish-release source current src no alpha-3 [18:01] because I believe at this point he sound be running publish-image-set from ubuntu-archive-tools, but the script is broken (working on fixing it now) [18:01] *should [18:02] yeah, that [18:02] it's not that broken :P [18:02] anyway that line errors out [18:02] ERROR: Cannot handle product Ubuntu Desktop Preinstalled armhf+ac100 [18:02] aren't you getting that one? [18:03] yep, but I hacked it [18:04] oh, I see the source bit... weird, never noticed that before :) [18:04] yeah. what is it? [18:06] looks like the last time it was built was for 12.04's release [18:06] stgraber: how are images synced to the cdimage.u.c machine? [18:06] is there a cron job? [18:07] IIRC one of the scripts triggers a rsync back from the various cdimage machines [18:07] right [18:07] so considering we haven't had a source build for a1 or a2, I wouldn't worry about a3 ;) [18:07] cool [18:08] sync-mirrors syncs the mirrors. [18:09] publish-release even helpfully reminds you to run it. [18:09] yeah. [18:09] Want to make sure nothing automatic was going to do it. [18:09] Nope. [18:10] It only get automagically called from image building. [18:10] s/get/gets/ [18:19] why is there a 12.10 directory on releases.u.c ? [18:19] with some arm thing [18:21] ubuntu-12.10-no-preinstalled-desktop-armhf+ac100.bootimg and friends [18:21] Laney, ^ [18:21] where? [18:21] it's probably me cocking it up [18:21] although I didn't sync anything yet. [18:22] * skaet notes that all the results are now in for A3 on the iso tracker [18:22] bdmurray: thanks! [18:23] Laney: it's on acai at least - not all machines [18:23] Laney: which is alarming in of itself [18:23] looks like it came across in a sync a few hours ago [18:24] I invoked the script wrong for that image [18:24] surprised that it synced it though ... [18:24] Well, uninvoke and sync. :/ [18:25] (I suppose it's not critical, no one's polling releases.u.c for new ac100 releases and getting excited, I'm sure) [18:25] But, yeah. Is it fixed on nusakan? [18:25] yes. [18:25] Laney: so yeah, we should have a source image built... dunno why it doesn't exist [18:32] Laney: I "think" this is the fix we need for publish-image-set: http://paste.ubuntu.com/1112420/ [18:32] Cloud Images are published [18:34] Laney: hmm, I might be missing something, the publish path for server on arm looks wrong with that patch [18:43] infinity: one more for you, https://code.launchpad.net/~stgraber/ubuntu-archive-tools/fix-preinstalled-images/+merge/116943 [18:43] Laney: ^ that one gives me the output I'd expect [18:45] nice [19:11] stgraber, looks like all the precise download links are broken [19:12] stgraber: Has that server publishing change been discussed somewhere? [19:12] the syntax is http://cdimage.ubuntu.com/precise/daily-live/ not http://cdimage.ubuntu.com/daily-live/precise. I started manually updating them, but that's silly.. can you script a change/ [19:12] ? [19:13] balloons: yeah, I should be able to script a change, I believe I have a python script I used to do something like that in the past [19:13] balloons: probably not today though [19:15] stgraber, right no worries [19:15] just a heads up :-) [19:15] infinity: what change? all I changed for server is make the armhf images publish at the same place as the i386 and amd64 ones [19:15] stgraber: Yeah, that's a change... [19:15] infinity: I believe this wasn't a problem in that past as daily-preinstalled was hardcoded to ubuntu-server/daily-preinstalled, but daily-live didn't have it hardcoded [19:16] Laney: shall I nuke the 12.10 stuff from www/simple? [19:16] please [19:17] done [19:17] double check the image diff for me? [19:17] infinity: so I'm just trying to have the server daily armhf publish under the same product as the server preinstalled armhf used to [19:18] stgraber: Alrighty. Merging. [19:18] given previous cock ups I'm now scared to push it [19:18] Laney: we should really be archiving off the alpha-1 and alpha-2 stuff; dunno why this wasn't already done at alpha-2 time [19:19] is that in the checklist somewhere? [19:19] infinity: I "think" the logic is right and the output looked sane here, but these scripts as such a mess of mapping dicts and regexps that I'm never sure I didn't miss a corner case... [19:19] yes it is [19:20] "if there is a previous milestone [...]" yah [19:20] we don't have particularly good scripting for that part [19:20] shall I? [19:20] goferit [19:20] will sync-mirrors delete that borked thing on releases? [19:22] grr, the alpha-1 directories are all owned by the wrong user [19:22] elmo: ^^ who should I pester about this? uid 3237 doesn't even have a name, but owns all the alpha-1 dirs, so I can't move them away for archival [19:23] it's probably tiaz; he was fixing something for me [19:23] will check with him [19:23] ok [19:23] alpha-2 is all moved, anyway [19:24] slangasek: 26/07 20:24:49 fixed [19:25] yay [19:28] Laney: so the diff looks like what I expect, but I never used that www.prev and don't know whose idea it was to add it to the publish-image-set script :) [19:28] heh [19:28] I like the safety net [19:29] I don't find it provides much safety because the diff is too large to eyeball [19:30] it lets you go back to the old state in case of problems [19:30] * Laney shrugs :) [19:30] just 'sync-mirrors'? [19:31] are the HEADER.html supposed to be updated manually? [19:32] yeah, the script gives you seddery [19:33] full/releases/quantal/alpha-3/HEADER.html doesn't look updated yet [19:33] will you do the seddery? [19:34] there we go [19:34] heh, nice, ac100 made it to /release ? [19:34]