[08:32] jamespage, ping [08:32] babyface_, morning [08:32] jamespage, good morning! [08:33] jamespage, I just found that the job http://10.189.74.2:8080/view/Quantal/view/ISO%20Testing/job/quantal-server-amd64_tomcat-server/70/testReport/junit/test/TomcatServerTest/testTomcatDaemon/ failed again. but in this bug: https://bugs.launchpad.net/ubuntu/+source/tomcat7/+bug/1025595 you said you already increased the time [08:33] Ubuntu bug 1025595 in tomcat7 "tomcat7 not running on port 8005 after installation" [Undecided,Invalid] [08:34] jamespage, could you have a look on it? [08:35] babyface_, I have - but it is possible that on a really slow VM that it still does not complete in time. [08:36] babyface_, I am considering an alternative fix to this - to make randomness a bit quicker in the test VM for this test [08:36] jamespage, yes, please. [08:37] jamespage, we can not just let it be unstable, that looks ugly ;) [08:37] babyface_, yes - but it is a test failure rather than a bug... I will fix it but not right now... [08:38] babyface_, if you want a green ball just run it again - in all likelyhood it will pass.. [08:38] jamespage: Assuming this is a /dev/random entropy issue, just make the VMs use urandom for random? [08:38] infinity, yeah - that was my thinking [08:38] jamespage, ack, I know, it's not a big deal, and not a real bug in code, so take you time and fix it when you are free [08:38] i've just not got to it yet [08:38] (That said, I could see it as a package bug too, if the package sometimes forks the daemon but then doesn't *actually* start for a minute, that's a bit unintuitive) [08:40] jamespage: But it's certainly a bug/misfeature in the test framework to ever use a real /dev/random, unless the point of the test is actually to test quality of randomness (which would be a rather rare testcase, I suspect). [08:51] infinity, certainly not in this case [09:51] !package haveged [09:51] Factoid 'package haveged' not found [09:51] !deb haveged [09:51] Factoid 'deb haveged' not found [09:51] jamespage: infinity: $ sudo apt-get install haveged [09:51] this will give you more entropy without using urandom. I do this a lot on VMs. [09:51] xnox, that of course is another alternative... [09:52] * jamespage wonders what that package actuall does... [09:52] Linux entropy source using the HAVEGE algorithm [09:52] french wrote it, it must be good =) [11:17] stgraber: I've confirmed that all those duplicate ArchivePermission rows are gone now. [11:18] stgraber: (P.S. please use edit-acl rather than edit_acl.py - at some point I'll drop that compat symlink) [11:48] erm... [11:49] Riddell: you're not in backporters; why are you uploading and accepting backports? [11:56] Laney: it's not good enough to be a core-dev? [11:56] no, just as it isn't for SRUs [12:30] Laney: sorry about that === shadeslayer is now known as shadeslayer_ === shadeslayer_ is now known as shadeslayer === yofel_ is now known as yofel [13:09] cjwatson: yay, thanks for cleaning that up! === shadeslayer is now known as shadeslayer_ === shadeslayer_ is now known as shadeslayer === ogra is now known as ogra_ === barry` is now known as barry_ === barry_ is now known as barry [14:34] ogra_, Daviey - I'm seeing we're still building arm server armhf+omap - who's using it/testing it? and can it be dropped? [14:49] mvo, we're starting to ramp into A3 now. Could you please do the GnomeAppInstallDatabaseUpdate now? [15:11] I just realized we're still showing 'quantal daily' and not alpha 3 as our milestone [15:14] skaet, well, i switched armhf to alternate over the weekend and would greastly prefer to drop server preinstalled but i'm not sure if QA relies on preinstalled [15:14] skaet, alternate is only built for omap4 atm [15:14] balloons, we'll be switching it over today [15:14] skaet, and server preinstalled is still broken [15:15] ogra_, why do we care about armhf+omap? [15:15] for server [15:16] skaet, because its the easiest way to install adev setup on a beagle [15:16] *a dev [15:16] using ubuntu-core is painful to set up, the alternative is the netboot images but these indeed require a network connection [15:17] but i wouldnt cry if we lose them [15:18] skaet, as for all omap3 images, we only build them because it has such a big community ... but we're lacking testers here [15:20] ogra_, if we're lacking testers, not much point in building them while we're limited for capacity. [15:20] * Riddell wonders if he has time to slip in one extra new package into the archive for alpha 3 [15:21] skaet, well, we get the kernel and bootloader for free (close to zero work) and we have a ton of spare beagleboards in the datacenter (former builders) [15:21] we should just send them out to some people :) [15:22] Riddell: mind if I rebuild calligra-l10n today for bug #1027215 [15:22] Launchpad bug 1027215 in calligra-l10n "calligra-l10n-xx should enhance, not recommend calligra" [High,Triaged] https://launchpad.net/bugs/1027215 [15:22] * micahg doesn't want to disrupt kubuntu alpha 3 [15:23] Riddell, how many hours is it going to take to build? [15:23] micahg: yes please go ahead [15:23] (and dependencies...?) [15:25] skaet: 5 minutes says launchpad, but it's actually not on the images so it shouldn't be an issue for alpha I think [15:32] Riddell, if its not on the image no worries, even if it is, we're still fine for something that small. [15:33] * skaet more worried about kernels, unity, firefox, launchpad, etc. [15:42] * cjwatson is still hoping to get the server squashfs work in today [15:43] just fixing tasksel at the moment [15:43] almost everything else is done [15:47] skaet: what's the worry about launchpad? [15:52] cjwatson, no worry about launchpad, just that "package" can be rather an ambiguous term in terms of build time. [15:53] was just trying to figure if it was something smallish, or not [15:55] ok [16:07] infinity: What are the chances you get around to fixing the fact that soyuz makes packages with uninstallable build-deps FTBFS instead of putting them in dep wait anytime soon? [16:07] In Kubuntu there's discussion of rewickering the packages to work around this, but it'll be work that could be avoided if that's going to happen soonish. [17:59] Laney, slangasek https://wiki.ubuntu.com/QuantalQuetzal/TechnicalOverview/Alpha3 is ready for data to be added from the various teams now. [18:00] great, perhaps link it from the freeze announcement [18:04] Laney, yup, just posting here to let you know that step is done. [19:22] infinity, until the new arm build infrastructure comes online, can we drop arm server armhf+omap? I don't think we have any testing lined up for, it, and we're getting pretty close to capacitity with arm builds in the 24 period. [19:23] skaet: I'm happy dropping it forever and just having netboot. [19:24] (Same goes for all armhf+omap images) [19:25] We discussed this last Alpha, but I think failed to come to action. [19:25] infinity, ok. for all the ones you're the signoff for, consider removed from the manifest then. [19:26] Riddell / ScottK: Any qualms about dropping armhf+omap from Kubutu (not omap4, just omap) [19:26] skaet: Which ones am I signed off for? :P [19:27] infinity, https://wiki.ubuntu.com/QuantalQuetzal/ReleaseManifest [19:28] ubuntu server pre-installed armhf+omap, ubuntu preinstalled armhf+omap are the ones I'm spotting with your name beside [19:28] Right. And Kubuntu doesn't list it in the manifest anyway, so dropping that works too. [19:28] infinity: I'm fine with it but is there a way I can justify that to my community? [19:28] Riddell: I don't think we have any testers. [19:29] Riddell: You apparently don't list it as a release image anyway... Unless that was a mistake. [19:29] Riddell: And you so don't have testers. I ended up testing that image for precise. [19:29] (And it sucked... Beagleboard + KDE = lolz) [19:29] And the Canonical QA person who used to do it in his personal time doesn't work for Canonical anymore. [19:29] So no testing. [19:30] that'll do then, fine to drop it [19:34] infinity, ok, dropped from the manifest. nusakan need editing to reflect, but I need to go to dentist now. If you get to it before I return, post here please. [19:34] Erk. [19:34] Laney: We dropped all precise arm images? [19:35] infinity: Apparently that was the decision. [19:35] infinity Core should still be there. [19:35] I missed that memo. [19:35] Well, yes. I don't count core when I talk about images. [19:35] Cause it's not one. [19:36] Not an installer, anyway. :P [19:36] Laney: What was the driver behind ARM not being cool enough for point releases? [19:36] https://wiki.ubuntu.com/PrecisePangolin/ReleaseManifest/12.04.1 [19:36] infinity, ^ this is for daily builds prior to LTS release. [19:37] skaet: Kay, that doesn't give me rationale. [19:37] and follow on from what was indicated as LTS vs. not in: https://wiki.ubuntu.com/PrecisePangolin/ReleaseManifest [19:37] all the arm ones were 18 month support [19:38] Ahh, fair enough. [19:39] Laney: That probably didn't need default-arches mangling, mind you, since "not doing daily-preinstalled builds" would have a similar effect. [19:39] Anyhow. No big deal. Just conflicted with my changes. :P [19:39] infinity, what changes???? [19:39] ... [19:40] hahaha [19:40] skaet: The ones we discussed 6 minutes ago? [19:40] heh, ok. :) [19:41] sorry, Laney's were done earlier and I was worried it was something else. [19:41] * skaet --> dental appt. biab [20:08] I updated the notice board for the ETA on when the milestone builds will hit. Can we please make sure it stays up to date as we have respins, etc? Thanks! [20:08] it's currently listed as 0600 on July 24th -- update if that day/time changes [20:10] I'm switching the server images over to squashfs-base style now [20:10] I guess there's some kind of fiddly pad thing I have to update? [20:11] for the image build list, there's http://pad.ubuntu.com/ubuntu-release [20:13] http://greenwellplumber.com/blog/wp-content/uploads/2011/12/Bathroom_humor_by_charfade1.jpg [20:18] OK, server switched to squashfs-base, amended pad accordingly [20:19] running an experimental build now, since all is quiet [20:34] Oops, build failure with squashfs-base. Fixing ... [22:23] Is the highbank stuff going into Quantal as well? Precise-proposed before Quantal doesn't fit the SRU policy. [22:28] NCommander: ^ [22:28] ScottK: its already in Quantal, but flash-kernel had a rewrite so its a different patch [22:28] OK. [22:28] Thanks. [22:29] I'm waiting for my PPA to finish building d-i to make sure I got everything sane then that gets uploaded and can be fully tested [22:30] Good to hear it's (at least not in the short term) another poulsbo. [22:30] * NCommander grumbles, armhf/powerpc seems tohave dropped off my PPA [22:30] (which you could only ever run on one Ubuntu release) [22:31] ScottK: its been a bit of a pain TBH. d-i itself requires a command line option to run while everything is in proposed. There is a chance everything might work, and the moment it gets pocket copied to updates it will break [22:31] (I'm pretty sure it won't, but we've never built a new image on a point release before) [22:31] * skaet back [22:32] The good news is that whatever it does, it's not a regression .... [22:48] slangasek, Laney - email sent for the freeze. [22:49] NCommander: Eh? d-i requires no such thing. [22:49] infinity: ? [22:50] NCommander: We do d-i in proposed all the time, how is that broken for you? [22:51] infinity: no, I'm not expecting it to break. However, because highbank's enablement bits are in proposed, we need to set apt-setup/proposed=true on the command line for testing [22:51] slangasek, will you be doing the debian-installer upload for the new kernel? [22:51] Oh, sure. Kay, I thought you meant something sketchier for building. [22:51] skaet: I will. [22:51] skaet: Once they all roll in. [22:52] I know d-i will pull udebs from updates, but a voice in the back of my head tells me to revalidate the instant highbank migrates to updates [22:52] infinity, what's still left to land? [22:52] infinity: when do you plan to upload d-i? I'm about to upload it to add highbank (I can depwaiton your patches) [22:52] skaet: Two ARMs and a PowerPC. [22:52] * skaet should have known. [22:52] NCommander: Talking about quantal here. [22:53] oh [22:53] NCommander: But I do need to commit a kernel ABI bump for precise-proposed too. No reason those can't be in the same upload as your fiddling, however. [22:54] infinity: I committed my highbank d-i patches onto the proposed branch, and testbuilt on armhf/i386 and looks good, so if you just want to upload it tonight [22:54] slangasek, infinity - I'm going to turn off the cron now, and point the next set of builds to start populating alpha 3 on the iso tracker. [22:54] * skaet figures Laney's EODed by now... === bjf is now known as bjf[afk] [22:55] infinity: if you plan on doing your ABI bump tonight, I'll leave it as is, else I rather upload ASAP as mahmoh and mreed need to testthis [22:56] NCommander: I can do it tonight. Gives me a chance to quickly review your bits too. [22:56] infinity: thanks [23:07] cron for quantal disabled, builds targetted to Quantal Alpha 3 [23:24] infinity, am looking at the changes made to the server rebuild, and its not synching up with what's in the crontab for server at the moment. Do you know or cjwatson make the changes to the pad? [23:24] (pad.ubuntu.com/ubuntu-release) [23:27] skaet: They seem to reconcile fine for me, what's the issue? [23:29] skaet: What infinity said. [23:29] I made those changes. [23:31] cjwatson, seeing buildlive with for-project $p cron.daily looked a bit odd [23:31] It's correct, and it matches the crontab. [23:32] It's not really a "live" build; we're using buildlive because it knows how to build squashfses, but it's just for the base system. [23:32] The images still use d-i. [23:33] thanks. that makes a bit more sense. [23:33] skaet: cron> sounds good, thanks