[00:00] GrueMaster, can you please open a bug, so we can reference it? [00:01] :) [00:01] Will do (once I figure out a few other more pressing issues). [00:01] GrueMaster, skaet: one already exists [00:02] https://bugs.launchpad.net/ubuntu/+source/unity-greeter/+bug/892394 [00:02] Launchpad bug 892394 in unity-greeter (Ubuntu) "Greeter logo needs to be updated for 12.04 (affects: 2) (dups: 1) (heat: 16)" [Medium,Triaged] [00:02] thanks stgraber. [00:02] stgraber, just heard about it from robert-ansell as well. you're faster this time. :) [00:02] hehe [00:03] * skaet --> heading for dinner, back later [00:03] * GrueMaster adds the but to the armel images on the shiny new iso tracker. [00:03] bug [00:05] This bug is old. Surprised it wasn't updated before A1. [00:06] yeah, really weird as it's an easy one (just a .png to update, took me a good 2 minutes for Edubuntu) and Robert commented on the bug (though only recently) [00:06] I could have fixed it, and I am not artistic by nature. [02:02] * skaet not seeing any blockers so far on the iso tracker, calling it an evening. [02:36] stgraber, are you the person i would talk to about which test cases are being shown for ubuntu studio on the qa tracker? [02:37] also, do you know who i would talk to about editing the instructions for the test cases? [02:39] the first sentence is referencing the 'entire disc with encryption' [02:40] the second sentence references the link from http://91.189.93.73/qatracker/milestones/205/builds/7263/testcases goes to http://testcases.qa.ubuntu.com/Install/AlternateResize, which doesn't mention jackd 'real time previliges' [02:44] ScottL: I can do changes on the tracker (rename/change testcases and make them point somewhere else), though it's probably best to talk to jibel about it as he's coordinating the testing and knows more about the right way of doing things :) [02:45] and jibel has the same rights as I do [02:50] oh, fun, looks like we just got a new empathy (new major version)... [02:54] stgraber: I think we have to go back to grilling people about freezes in DMB meetings :) [02:55] micahg: yeah, that sounds like a useful thing to do ;) [02:55] fortunately, it can't build since it requires clutter-gst [02:55] * stgraber waits for someone to upload a new clutter-gst still during the freeze [02:56] stgraber: not seeded, doesn't matter [02:57] ah, it build-deps on something new that needs to be promoted? [02:57] stgraber: worse, build deps on something we said we wouldn't promote :) [02:57] hmm, ok :) [02:58] Maybe we should send a friendly reminder for people to subscribe and READ ubuntu-devel-announce? [02:59] it's not really high traffic and should be a must subscribe for anyone with upload rights [02:59] really, I think ubuntu-devel should be required reading for devs with upload rights, but certainly ubuntu-devel-announce at a minimum [03:01] ubuntu-devel can be high traffic at times, you should at least know what threads are going on and ignore these you aren't interested in, but -announce should definitely be a must read [03:03] I guess I'll add that to be before-voting list of questions :) "are you subscribed and do you read ubuntu-devel and ubuntu-devel-discuss every day? (that's at least every day where you contribute to ubuntu)" [03:08] speaking of... [03:27] micahg: I might need educating on freezes, yes. ;) [03:28] infinity: not you :) [03:28] micahg: I dunno, my accidental mono upload made me feel special. :P [03:28] infinity: you at least caught it and apologized before I got the -changes E-Mail :), people make mistakes [03:32] * stgraber just got his first armhf build failure! [03:35] ooh, mono-devel is part of the DVD seed [03:44] infinity: I would think getting armhf bootstrapped and building in the archive would make you feel special [03:55] wait, i'm supposed to be reading ubuntu-devel-discuss? :( [03:55] argh, s/discuss/announce/g :) [03:56] though I do read -discuss every day too but it's not nearly as useful as -announce :) [03:56] ah, good :) [03:57] ooh, didn't catch that === bladernr_ is now known as bladernr_afk [06:29] Good mornintg [06:30] skaet: btw, the complete rebuild only takes 4 hours because of powerpc [06:31] skaet: i386/amd64 desktop images now take 8 minutes, but have to wait for powerpc which takes some 35 [06:31] skaet: there's an RT for switching those to a faster machine [06:32] infinity: I'm not worried at all about packages not being up to date on the a1 images; that will basically never be true anyway [06:32] infinity: as long as arch skew on mono doesn't cause uninstallability, that should not matter [06:34] skaet: so, can't see any blockers right now, except for this infamous bug 894768; anything I need to watch out for during my day? [06:34] Launchpad bug 894768 in linux (Ubuntu Precise) (and 1 other project) "Installation randomly fails with: File "/usr/lib/ubiquity/ubiquity/install_misc.py", line 621, in copy_file targetfh.write(buf) IOError: [Errno 22] Invalid argument (affects: 6) (dups: 6) (heat: 58)" [High,Triaged] https://launchpad.net/bugs/894768 [06:43] infinity: hm, seems arch desync does cause trouble.. http://people.canonical.com/~ubuntu-archive/testing/precise_probs.html [06:43] let's hope that it finishes building soon [06:44] armhf already finished, seems these builders are faster? [07:58] cjwatson, skaet: FYI, I updated publish-image-set.py for the new tracker [07:58] still had to use screenscraping, the XMLRPC API is missing some bits for this still (in particular, retrieving builds) === doko_ is now known as doko [09:36] skaet: FYI, added "general" and filled out the "Ubuntu" part of https://wiki.ubuntu.com/PrecisePangolin/TechnicalOverview [10:54] pitti: Yeah, armhf has all the pandas right now. [10:55] pitti: And sorry about the skew. Was a complete oops on my part. [10:57] infinity: np, it's all good now [11:04] * ogra_ wonders if hf actually gains us compiler speed too [11:04] does that use the floating point engine massively in any way 8i would have thought not) [11:35] skaet: also added some known bugs to TechOver now [11:36] jibel: so aside from the dreaded EINVAL thing, I don't see any showstoppers so far; did you? [11:39] pitti, I'm currently facing a problem with i386 installed on a server. jamespage says it could be due to powernap, I'm testing a workaround now [11:40] problem = the server hangs and tells me bad things like http://paste.ubuntu.com/754743/ [11:40] jibel: I'm pretty sure thats what its is - same symptoms as I saw in the openstack lab [11:40] ah, offlining CPUs [12:13] pitti, jamespage bug 898127 [12:13] Launchpad bug 898127 in linux (Ubuntu) "system hangs and errors at /build/buildd/linux-3.2.0/arch/x86/kernel/apic/ipi.c:113 default_send_IPI_mask_logical+0xdc/0xf0() (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/898127 [12:30] jamespage, no hangs with your workaround, could you update the bug report, and I'll add a release notes task. [13:57] jibel, jamespage: do we want to upload a powernap workaround for this, or just release-note it? [13:58] Daviey: any opinion on ^^ [13:59] * Daviey looks [14:00] jamespage: Only impacts a subset of machines? [14:00] we've only seen it on Dell? [14:00] Daviey: in this instance it was machines with the ubuntu-orchestra-client installed which still recommends powernap [14:00] jibel: do you know what type of machine that issue happened on? [14:01] ubuntu-orchestra-client is still universe.. we don't install powernap by default, but shipped server-ship [14:01] jamespage, PE R210 [14:01] so a dell then [14:01] I really don't think this warrants a rebuild. [14:01] right [14:01] Daviey: I don't think so either - is powernap even on the ISO? [14:02] * jamespage goes to lookl [14:02] server-ship [14:04] so I see [14:04] * jamespage should read things better [14:06] I don't think it even needs release notes providing there is a fix in the archive at release. === bladernr_afk is now known as bladernr_ [14:17] putting several of the armhf builders on manual to minimize my pain from some maintenence to happen shortly === Ursinha is now known as Ursinha-lunch [15:17] How do the meta-release files at changelogs.ubuntu.com? I fixed one of them yesterday and hit hasn't been updated. [15:17] pitti, good morning. Thanks for the updates to TechnicalOverview. :) [15:18] oh, technicaloverview alsmost forgot that one :) will need to update (not that there's much to say at this point) [15:18] stgraber, thanks! [15:19] * skaet crosses one of the teams to remind off her list ;) [15:20] mvo: ^? [15:29] bdmurray: what did you fix, sorry I missed the context [15:29] bdmurray: oh, indeed [15:30] bdmurray: its a manual process currently, anyone in the changelogs group can update it [15:30] I did that now [15:30] mvo: okay, thanks [15:52] jibel, mvo - has anyone started off the https://wiki.ubuntu.com/UpgradeTestingProcess? [16:03] skaet: I haven't but I think that jibel has [16:08] skaet, I did excepted cdrom upgrade [16:13] thanks jibel, mvo === Ursinha-lunch is now known as Ursinha [16:51] skaet, wubi.exe is not on the desktop images and not listed in http://cdimage.ubuntu.com/daily-live/current/precise-desktop-amd64.list [16:52] skaet, did I miss something or is it really missing ? [16:52] pitti, ^ [16:52] * pitti looks [16:53] jibel, should be there, will look into it too with pitti [16:53] right, it's missing indeed [16:54] * jibel adds a task to test the content of the images [16:56] jibel: how can I find out what failed here? https://jenkins.qa.ubuntu.com/view/Precise/job/precise-upgrade/PROFILE=main-all,label=upgrade-test/ [16:57] pitti, go to the last failed or select job 58 [16:57] pitti, the logs are attached as artefacts [16:57] jibel: ah, found it; thanks! [16:57] in that case main.log says that 2011-11-30 06:54:46,720 ERROR Dist-upgrade failed: 'E:Error, pkgProblemResolver::Resolve generated breaks, this may be caused by held packages.' [16:57] so you'll need to look at apt.log [16:58] and you'll see there some dependency issue with tdsodbc [16:58] Broken tdsodbc:i386 Breaks on libiodbc2 [ i386 ] < 3.52.7-2 > ( libs ) [16:58] Considering libiodbc2:i386 3 as a solution to tdsodbc:i386 0 [16:58] Holding Back tdsodbc:i386 rather than change libiodbc2:i386 [16:58] jibel: ok, will do that tomorrow; but sounds just like arch desync [16:59] pitti, I'll look at it tomorrow morning with mvo because we also need to fix the base image for amd64 [17:00] I need to run for today, Taekwondo time [17:00] good night pitti [17:00] skaet: no idea about missing wubi.exe right now, I don't know how this works; need to RTFS tomorrow; perhaps ev knows [17:01] RTFS? [17:01] read the fine source [17:01] cdimage scripts and so on [17:02] :) gotcha. [17:04] Is there an expert here or resources on packaging license dependencies? [17:04] I'm not sure where to ask about this [17:05] * pitti waves goodnight, need to runu [17:05] Is no one testing kubuntu? [17:06] mdeslaur, I've got some background, what's up? [17:06] I'm pulling the armel images now so I can whip through them. [17:06] mdeslaur, don't ask to ask just ask I guess, worth case you get ignored :p [17:06] GrueMaster, I've posted in kubuntu-devel channel asking for some help, but no one's stepped forward. [17:07] seb128: pffff :P [17:07] skaet: I'm looking at bug 891192 [17:07] Launchpad bug 891192 in ubuntu "Please sponsor the proposed qt4reactor packaging branch (affects: 1) (heat: 8)" [Undecided,New] https://launchpad.net/bugs/891192 [17:07] RIddell is out due to car accident, and not back on line yet. [17:07] mdeslaur, looking [17:07] skaet: about sponsoring qt4reactor, but the reason it's not bundled with twisted in the first place is because twisted is MIT and pyqt is GPL [17:08] oh [17:08] pitti, skaet: entirely my fault [17:08] fixing now [17:08] (directory structure is missing in my public_html) [17:08] thanks ev [17:09] mdeslaur, do they need to link together in some way (dynamic or static) to work? [17:10] skaet: qt4reactor imports pyqt on a try-except block [17:10] skaet: it's python, which is why it's complicated :P [17:10] skaet: if pyqt is not available (ImportError), it will try to use pyside [17:11] pyside is LGPL [17:11] mdeslaur, heh, yeah, see the issue, LGPL is better license for this sort of thing. [17:13] mdeslaur, let me ping around on this then. imports aren't quite dynamic linking, but ... yeah, good question. [17:13] skaet: fixed [17:14] jibel, ^ [17:14] skaet: thanks. [17:15] skaet, ev thanks. [17:15] skaet, you'll rebuild the desktop images but not the squashfs ? [17:19] jibel, can do, am also ok with release noting this, since some things have crept into the repository during soft-freeze that might have side effects. What's the re-testing bandwidth like? [17:21] skaet, not much. [17:21] skaet, +1 for release noting for me. [17:22] jibel, anything else showing up from that testing that might provoke a respin? [17:22] s/provoke/necessitate/ [17:23] skaet, nothing [17:24] jibel, ok, lets plan on release noting it. [17:40] jibel, just talked to superm1 - they won't be releasing mythbuntu images with A1. I've removed them from the tracker. [17:40] skaet, k [17:41] skaet, netboot images are not on the tracker but are good. [17:42] jibel, thanks for checking them [17:54] stgraber, ^ I'm seeing netboot images as active, any idea why they might not be showing up? something need to be twiddled on nusakan? [17:58] about netboot, precise is not listed at http://cdimage.ubuntu.com/netboot/ [18:01] jibel, yup that would do it. [18:01] jibel: http://archive.ubuntu.com/ubuntu/dists/precise/main/installer-amd64/current/images/netboot/ and installer-i386 I imagine [18:02] jibel: as a temporary fix [18:02] jibel: if memory serves cjwatson does some hocus pocus and it appears in http://cdimage.ubuntu.com/netboot/ but that is the original link [18:03] davidm, thanks, I know where to find it, just mentioning that it is not linked from cdimage.u.c [18:04] jibel: wrong david but I figured you might but always worth checking :) [18:04] :) [18:05] jibel: assumption the mother of all *choose your expletive* ups ;) [18:12] jibel, I've gone in and added precise pangolin to the page, but will need to wait for it to publish, and we'll see if it picks up the right pieces. If not will get cjwatson to look at it tomorrow morning. (he'll be back from vacation then). Since the images are known to test out ok, that's the main thing. [18:14] skaet: netboot is an interesting one, as we don't build them daily but only when building a new d-i, they'll never show up automatically on the tracker, unless cjwatson does some magic to get them update when a new d-i is uploaded [18:15] stgraber, interesting, lets find out how he wants to handle it then when he's back tomorrow. [19:15] I test omap4 netinstaller daily, and it works fine here. After I run through the kubuntu armel images, I'll test the omap netinstaller. [19:17] skaet, jibel: LTSP failed on alternate, apparently something changed in ssh ... looking into that now [19:18] stgraber, ack. thanks [19:19] skaet: I'll likely have a fix in the next 30 minutes (as in, I have a fix, it takes me 30 minutes to validate). Not sure if it's worth rebuilding though, your call. [19:21] * skaet pondering [19:50] skaet: looks like the fix works (at least it didn't fail at the same place as before) [19:50] skaet: I should be able to check that a thin client can actually boot from that server in the next 10min or so [19:51] stgraber, ok. [19:55] skaet: confirmed to work after reboot [19:55] skaet: should I upload a new LTSP with the fix? [19:55] stgraber, yes go ahead and upload. [19:56] (btw, the problem is Precise's ssh being picky about the permissions of /var/run/sshd) [19:58] stgraber, I think we can live with it being release noted, but want to get jibel's input. [19:59] skaet: right, only thing to be aware of is that I can't provide a workaround. Install will just fail, so workaround is to use a more recent daily. [20:00] stgraber, fair enough. [20:01] skaet: I don't feel like explaining how to modify the postinst script by hand at run time in the release notes ;) [20:02] * skaet interprets that stgraber recommends a respin. [20:03] skaet: if we want/care about LTSP for alpha-1, then we need a respin [20:03] as usual it depends on how quickly we can get it re-tested [20:03] stgraber, yes, retesting is the concern at this point. [20:03] jibel, ^^ [20:04] at least it's limited to Ubuntu Alternate (i386 and amd64), Edubuntu's LTSP isn't affected [20:04] thanks for confirming Edubuntu's LTSP isn't affected. [20:05] * skaet goes and kicks off the Ubuntu Alternate respins [20:08] at least the current ones are fully tested, so we can always go with that [20:10] * skaet nods [20:15] * GrueMaster hears "respin", gets as nervous as a strung out long-tailed cat in a room full of mouse traps. [20:17] * skaet reassures GrueMaster with phrase "Ubuntu Alternate" at this pont. [20:18] Yea, saw that. [20:18] :) [20:22] jibel: I'm sure you'll also be interested by http://91.189.93.73/qatracker/milestones/205/history (there's no link to /history in the UI at the moment as it's still fairly new) [20:23] skaet: I guess that between that page and jibel's report you should have a pretty good overview of what happened for a given milestone [20:24] stgraber, yes, this is great!!! Thank you very much! [20:25] skaet: that's for Oneiric Final: http://91.189.93.73/qatracker/milestones/202/history [20:28] * skaet *hugs* stgraber. [20:28] sweet!! [20:39] * skaet waiting for LTSP to actually publish before trying again. [20:51] skaet, where should release notes be going ? [20:54] smoser, https://wiki.ubuntu.com/PrecisePangolin/TechnicalOverview [20:54] yeah, jst found thank you. [20:56] :) [21:52] LTSP published now. [21:52] started respin of ubuntu alternate. [21:53] pitti, jibel - xubuntu won't have time to retest, so leaving their images alone.. Haven't heard backfrom kubuntu or lubuntu yet on whether they want a respin. [21:55] lubuntu just responded, they'll stay with their current images. [22:12] kubuntu responsed they'll stay with their current images in #kubuntu-devel channel. [22:12] new ubuntu alternate images with LTSP posted. [23:21] latest alternate looks good. [23:22] I'll test wubi tomorrow morning when it's published somewhere. [23:22] good night [23:29] skaet: Kubuntu armel images tested ok. [23:29] Thanks Gruemaster [23:31] jibel, thanks for confirming alternates look good. [23:32] jibel, unless pitti feels otherwise, i recommend release noting the WUBI issue.