[00:47] infinity: kubuntu images good (if not perfect) release page at https://wiki.ubuntu.com/TrustyTahr/Beta2/Kubuntu [01:30] Riddell: Want to email that to me so I don't lose it in backscroll? [01:53] marked Edubuntu as ready, did two test installs for each image, I guess that'll have to do. highvoltage will take care of the paperwork in the morning. === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [06:10] hi release team, who could take a look at FFE bug #1293299? [06:10] Launchpad bug 1293299 in Ubuntu Kylin "[FFE]upload ubuntu-kylin-software-center into archive" [High,Confirmed] https://launchpad.net/bugs/1293299 [06:10] thanks so much:) [07:25] JackYu: When we're in the middle of the beta, that's probably not the best time. [07:25] JackYu: Unless you're implying that you want that in, and in your images, and a respin and test before I wake up... [07:26] infinity, yes, we can:). It's 3pm here. [07:26] JackYu: (Honestly, I'm not sure that's a realistic goal) [07:26] infinity, but there needs time to upload it... [07:27] JackYu: Yeah. I think we'll be happy to review it after the beta freeze, pretty sure that's not a sane goal to accomplish in the next ~8h, which is when I expect all the image testing to be done. [07:27] infinity, yep. We can upload it after this beta. [07:27] infinity, sure, thanks:) [07:28] JackYu: Out of interest, why a whole new software-centre, instead of figuring out a clever way to add your apps to ubuntu-software-centre and maybe score them higher in searches? [07:28] JackYu: That would seem like the better integrated approach. [07:29] JackYu: USC already has the ability to read multiple databases. [07:29] JackYu: (See app-install-data and app-install-data-partner) [07:30] JackYu: An app-install-data-chinese or something would make sense, I'd think. [07:30] infinity, en, you are right. We have discussed with the USC team. Since more and more customized features would be included into UKSC, they suggest to create a new one. [07:31] infinity, and the UI of UKSC would be more easy for Chinese users:) [07:32] A fork is probably going to end up being a lot more work than you want it to be, though. :/ [07:33] Anyhow, bring it up in the channel after beta release, when one of us can dedicate some more time to the matter. [07:33] I should stop arguing with nodejs on powerpc and sleep at some point, so I can do the release tomorrow. [07:33] infinity, yep, we tried. But USC is a big project:(. [07:34] infinity, OK, thanks. Good night! [07:34] JackYu: If you can articulate exactly what you need, and why it needs to be different, we may be able to reopen the conversation with the USC folks. I'd really prefer it be integrated instead of forked, if we can make that happen for you. [07:35] JackYu: (Perhaps in the bug, so it doesn't get lost in IRC backscroll) [07:36] infinity, sure. I will add some information on the bug. [07:36] JackYu: Thanks. [07:36] JackYu: To be clear, I assume you're open to the idea of integration, if USC upstream can be talked into it, right? [07:37] infinity, To be honest, I prefer to create a UKSC, instead of integration. [07:38] infinity, we talked to USC team, but they are much busy... [07:39] infinity, at the beginning Trusty, we have some conf call with USC team. They were busy with Ubuntu Store. [08:22] jibel: If you run into anything that you think absolutely must be release noted for the beta, can you add it to the wiki, or poke me about it while I sleep? [08:22] jibel: Not sure how much timezone overlap we'll have. [08:23] infinity, okay, major issue ATM is wrong keyboard during live session and after installation for some flavors. I'm trying to understand if that's 1 or 2 issues and will upadte the release notes. [08:24] jibel: Ta. [08:25] jibel: just so you are aware this issue pitched up shortly after the ibus update to 1.5.5 but we have had no luck trying to go back to a previous version [08:26] elfy: As in, you've ho luck talking people into reverting, or trying to revert hasn't made the bug go away? [08:27] infinity: as in the people trying to go back to previous version weren't able to [08:27] ended up with dependency issues I think - it was knome doing it [08:27] infinity: purging ibus fixes the issue [08:28] elfy: Fun. [08:28] loads ... [08:28] anecdotally it's affected other keyboard layouts [08:28] but I've not seen that in bugs [08:29] We'll definitely need to make sure someone prioritized that one post-beta. [08:30] that will be good - up to now this bug hasn't really had much traction outside of xubuntu [08:32] Right, well. If it *is* due to our ibus changes, Canonical engineers shouldn't be breaking flavours gratuitously, so if you have well-documented bugs with clear reproducers, I'll make sure someone stares at it grumpily for a bit. [08:32] thanks :) [08:32] Probably William Hua, he seems to understand ibus. [08:33] At least, more then people like me, who don't understand it at all. :P [08:33] * elfy too [08:39] infinity, elfy bug targeted to trusty and assigned to the desktop team [08:39] jibel: thanks :) [09:07] seb128, salut, bug 1284635 is for your team I guess [09:07] Launchpad bug 1284635 in ibus (Ubuntu Trusty) "Keyboard layout changes after login" [High,Confirmed] https://launchpad.net/bugs/1284635 [09:09] jibel, salut, I saw it, do you have an install in that state by any chance? [09:09] jibel, what is in /etc/default/keyboard? [09:10] seb128, no, I couldn't reproduce on Ubuntu, I'll try on Xubuntu but I've another high/critical issue to confirm before that [09:10] ok, thanks [09:10] jibel, if it doesn't happen on Ubuntu I'm unsure it's for our team... [09:11] seb128: It's for your team if an upload by your team broke other flavours, IMO. [09:11] seb128: But we can argue that when I'm not supposed to be asleep. :P [09:11] seb128, it's a bug introduced by ibus 1.5.5 [09:11] infinity, right, but I don't think we changed anything, that's coming from upstream/debian [09:11] the xfce/ibus integration might be buggy [09:12] anyway, I've no clue about ibus [09:12] I'm going to ask happyaron to have a look though [09:12] thanks [09:12] but I don't think that's going to happen this week [09:12] he's busy dealing with ubuntukyling work afaik [09:12] Well, it's too late for beta (obviously), just want someone to give it a good stare before release. [09:14] seb128: it's not just xfce - it's lxde as well [09:15] elfy, but it's not happening in Unity? [09:15] I didn't see it there [09:17] well, anyway, assigned to happyaron, let's see when he finds a slot for it, as said he's quite busy [09:17] seb128: and in an install that has the issue /etc/default/keyboard looks right for a UK keyboard [09:18] If the bug is reasonably well documented and we're (fairly) sure it's ibus, a bisection might not be too tough. [09:18] It can't have changed THAT much in the last couple of upstream microreleases. [09:18] infinity, ibus doesn't work that way [09:18] their "microreleases" include rewrite of part of codes and features [09:19] they are just new versions [09:19] seb128: That's not comforting. :P [09:19] I know :/ [09:19] well, happyaron and others are going to tell that we should use fcitx, they keep pushing for that [09:19] A little late for trusty, methinks. [09:19] we just never managed to get anyone wanting to look to IMs on our side [09:19] right [09:20] well, UbuntuKylin is using fcitx [09:20] Yeah, but they also only have one (two, I guess) languages they care about. [09:20] right [09:20] Validating it for every current ibus consumer is probably a bit tougher. [09:20] but they are the only people who are doing IM work for us [09:21] or having knowledge on the topic [09:21] Aaaaanyhow. Really going to bed now. Tomorrow morning is going to come way too early and suck way too hard. [09:21] I don't even know how to input stuff with ibus [09:21] infinity, night! [09:21] infinity: good night :) [09:22] seb128: Yeah. I used to care when I ran my desktop in Japanese, but I never cared enough to learn how it all worked. I was just simultaneously amazed that it worked at all and frustrated that it didn't work as well as Windows, and stopped doing it. [09:22] That was almost a decade ago, though. I really hope it sucks less today. [10:05] confirmed bug 1298251 on kubuntu only [10:05] Launchpad bug 1298251 in ubiquity (Ubuntu Trusty) "[kubuntu] Ubiquity crash when starting LiveCD with a chosen language" [High,Confirmed] https://launchpad.net/bugs/1298251 [10:38] hmm, I don't like the look of process-cpp/unity-scopes-api/unity-mir being accepted without dbus-cpp - wonder what's going on there [10:40] ah, it's in the ubuntu-desktop packageset but not images [10:40] reviewing manually then [10:41] no-change rebuild, ok :) [11:23] stgraber, liburcu ftbfs on armhf === alex_abreu is now known as alex-abreu [12:58] apw, there are 2 kernel bugs in the release notes, bug 1054732 and bug 1066883 , they are almost inactive since saucy, should they still be mentioned in the release notes for Trusty? [12:58] Launchpad bug 1054732 in linux (Ubuntu) "[LENOVO 4298R86] suspend/resume failure" [High,Incomplete] https://launchpad.net/bugs/1054732 [12:58] Launchpad bug 1066883 in linux (Ubuntu Saucy) "[Macmini 5,1] Fatal server error: Can not run in framebuffer mode on reboot" [High,Confirmed] https://launchpad.net/bugs/1066883 [12:59] apw, also any new issue that should be added to the release notes? [13:09] jamespage, there is no result for MaaS on the tracker, has it been tested? [13:09] jibel, apw is out until tomorrow. as for bug #1066883, I have a Mac mini that appears to be working fine. I'll reconfirm using today's daily. [13:09] Launchpad bug 1066883 in linux (Ubuntu Saucy) "[Macmini 5,1] Fatal server error: Can not run in framebuffer mode on reboot" [High,Confirmed] https://launchpad.net/bugs/1066883 [13:11] rtg, thanks. Anyone from the kernel team could review the release notes of Beta2? Leann is out too apparently. [13:12] jibel, URL ? [13:12] rtg, https://wiki.ubuntu.com/TrustyTahr/ReleaseNotes [13:12] looks like we have the perfect kernel :) [13:13] right :) [13:13] jibel, matsubara was going todo that - I'll ping him [13:13] jamespage, thanks [13:17] jibel, the Kernel release note about suspend/resume issues is likely still accurate. There are about a zillion suspend/resume bugs against Trusty right now, so I'd just leave it in the notes. [13:18] jibel, I'm getting ppisati to review the OMAP4 installation note. [13:28] rtg, Ta. also add any major kernel issue that you think is worth mentioning. [13:29] jibel, well, like you said, we have the perfect kernel :) [13:30] rtg, I know, we do the same in QA, don't test so there is no bug ;) [13:32] jibel, oh, there are plenty of bugs. the suspend/resume issues are the only problems that I know of that are systemic. [13:50] doko_: I'll look into it, thanks [13:56] doko_: upstream is on it [14:11] doko_: so apparently the problem is gcc 4.8.x on arm producing some invalid code (http://git.lttng.org/?p=userspace-rcu.git;a=commit;h=b52a3dc23521f7a21fb5ab3d2b53dad89c586ee4). I'm trying to get the details of what exactly the bug is, whether they reported it upstream and whether it's been fixed there... [14:12] the obvious problem is that our current liburcu which was rebuilt a few days ago when xnox did some packaging changes is likely currently affected by this... [14:12] # ifdef __ARMEL__ [14:12] then why on hard float? [14:14] stgraber, and if it really is a GCC bug ;p ... [14:15] doko_: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58854 is apparently the bug in question [14:15] gcc.gnu.org bug 58854 in target "[4.8 regression] "sub sp, fp, #40" hoisted above frame accesses" [Normal,Resolved: fixed] [14:16] stgraber, fixed in trusty [14:16] doko_: cool, is that because it made it to a stable release or did we cherry-pick? [14:17] stgraber, gcc patches are updated to the current branches [14:17] will ask for an 4.8.3 sru for 14.04.1 ... [14:21] doko_: good, I'll patch out that chunk of code and re-upload then. Thanks for checking. === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [15:40] FYI not the final report but close to: https://wiki.ubuntu.com/QATeam/ReleaseReports/TrustyBeta2TestingReport === doko_ is now known as doko [15:45] stgraber: please could you force gnat into trusty? the dependency packages don't yet exist [16:16] doko: Why would we need to force it in when the only change is to those deps? [16:16] doko: Raising the uninst count for nothing doesn't make sense. [16:16] infinity, it's not raised. it's the same before and after the change [16:17] doko: Okay, changing the uninst profile, then? Still doesn't make sense. If the packages are going to exist, it'll migrate when they do. [16:17] "Migration at any cost" shouldn't be the goal. [16:24] zequence: There don't seem to be any results for studio/amd64. Has anyone smoketested it at all? === shadeslayer is now known as shadeslayer_ === shadeslayer_ is now known as shadeslayer [16:25] * lamont has a question about postgresql SRU -- when do we expect that to come from -proposed into -updates, assuing all goes well? [16:26] lamont: Monday. [16:26] lamont: Though, we could release it today. [16:27] infinity: Just about to [16:28] zequence: Ta. [16:28] infinity: if it's going to be "by monday", today would make my life simipler [16:29] but "make lamont's life simpler" is not a valid release reason [16:29] lamont: Sure. I've pinged pitti to get his take on it. If he thinks it's well-tested, I'm fine with releasing it a day early. [16:29] ta [16:44] highvoltage: is the paperwork for Edubuntu 14.04 beta 2 ready? [16:58] stgraber: no :( (I was just dreading when you'll ask me but I'll check it in a minute) [16:59] jibel, ppisati says the Installation release not regarding ARM OMAP4 is no longer relevant. That bullet item can be removed. [16:59] note* [17:01] jibel, the release note "On a mac with an external display" can also be removed. I've installed with Trusty desktop daily on a mac-mini with no issues. [17:03] stgraber: what will the urls be for the iso images? http://cdimage.ubuntu.com/edubuntu/releases/trusty/beta-2 ? [17:03] highvoltage: yep [17:04] rtg, thanks for the confirmation [17:06] stgraber: any noteworthy bugs found during testing? I see none are mentioned on the tracker and the previous bugs that were listed during alpha 2 were resolved [17:06] infinity, I added the most important issues found during testing to the release notes [17:07] highvoltage: nope, just the usual list of bugs from Ubuntu [17:07] jibel: Thanks, you're a champion. [17:07] highvoltage: I'm not aware of any Edubuntu-specific bug at this tage (besides us sucking at putting new stuff into that release :)) [17:08] stgraber: *nod* [17:13] stgraber: ok seems in order, will probably make some small edits still after proof-reading again [17:13] (kubuntu's release notes are really nice I'm going to steal some ideas there next time) [17:16] highvoltage: cool, are we doing wiki-only like the past few milestones or are you also doing the whole blogpost thing? [17:16] highvoltage: in either case, you'll want to give the URL to infinity for inclusion in the announcement [17:16] infinity: Link to our release announcement http://ubuntustudio.org/2014/03/beta-2-is-out/ [17:16] stgraber: wiki only, but I think we should do some form of blog post soon anyway as a status update (doesn't have to be directly coupled to the releast though) [17:16] *release [17:17] * infinity wishes we hadn't gotten into this linking to separate announcements in the release announce habit, it's a monumental pain to collect them all. [17:17] infinity: you can also do it the lazy way as I've been doing for the past few milestones, copy/paste the previous announcement to etherpad, post link on IRC, let everyone edit [17:17] at release time, copy/paste in mail client + send, done :) [17:21] I'm stuck with burning a DVD (which so often ends up with failure, at least for me), since my computer at times doesn't like my usb sticks. While I wait for it to finish, I'm going to the store. If the gods approve, I'll finish testing our amd64 image within an hour [17:22] zequence: Yeah, that's fine. I need to clean up paperwork and such, we're not releasing in the next half hour. :P [17:25] infinity, I'll sent the testing report and I think Beta 2 is as good as it can get. [17:25] *send [17:25] jibel: Many thanks. [17:27] infinity: ETA for release? [17:28] Riddell: I think I said mid-to-late-afternoon before. Let's say ~4h. [17:29] groovy [17:32] infinity: Can we start flushing out unapproved? I know the CI folks would love to get some silos back [17:33] cjwatson: Yeah, I think we're mostly green, and anyone who's late to the party isn't getting a respin now, they're getting a not ship. [17:34] That's what I figured === shadeslayer_ is now known as shadeslayer [18:00] Somebody who isn't about to EOD might want to look at the FFe associated with the webapps landing currently in unapproved, and see if it can be approved [18:00] I'd rather not accept it before that's done [18:13] I think I've run out of steam for today, somebody else can look at the rest of the queue :) [18:14] I'll do some tomorrow [18:14] FFe queue needs some attention too [18:14] Cheers for the reviews thus far [18:15] infinity: Done [18:28] hi, can we please have unity8 pushed through from proposed? [18:29] Saviq: We'll get there. [18:29] infinity, ok thanks, I know you're swamped [18:37] Sonofa... [18:37] stgraber: How did we completely miss adding mythbuntu to the manifest? [18:37] superm1: And how did you miss telling us? :P [18:41] infinity: that's a good question, my best guess is that since they are LTS-only and didn't participate in any of the trusty milestones, we just never had them in the list to begin with. [18:41] whereas all the others did either participate in the saucy release or did a milestone release for trusty already [18:42] stgraber: Yeah. And I'm thinking the LTS-only thing is also why THEY didn't notice, since they haven't done a milestone in 2 years, and fell out of the habit. :/ [18:42] Anyhow, if they step up tomorrow and say "whoa, wait, halp, we want to do our release, honest", I don't think I'll disqualify them for missing the beta. [18:42] Since it seems like no one remembered them on either side. :P [19:20] infinity: we didn't notice because i've been really busy with other stuff :( [19:21] superm1: Well, as pointed out, it seems NO ONE noticed. So, I'm waiving the usual "if you don't do beta, you don't get to do release" stipulation. [19:21] phew :) [19:21] superm1: But, it might be in your best interest to get testing dailies ASAP, so you know what you'll need to fix for final. [19:22] yeah i know we have some problems right now [19:22] i'm going to try this weekend to help sort them out [19:36] * infinity lunches while he waits on mirror propagation. === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [20:59] zequence: Your studio announce is a 404, is that going to be not true soon? [21:02] infinity: whats the eta for release? [21:03] infinity: wondering if I should disappear for a hour or so [21:05] utlemming: It's basically all ready, except for a website hiccup. We'll see how long that takes to get resolved and, if it doesn't, I'll just use a different URL in the announce mail. [21:05] utlemming: But I'm in no massive rush. I'll wait another couple of hours before pressing the button, probably. [21:05] infinity: ack...okay, can you ping me via email and I'll make the cloud images public then [21:06] infinity: Published now, thanks [21:15] zequence: I beg to differ. Or you gave me to wrong URL. :P [21:16] zequence: Good thing I didn't send yet. [21:16] * infinity fixes his reference. [21:16] infinity: hey, I have not assigned ubuntu-release yet, but can you look over bug #1298611 and comment? [21:16] Launchpad bug 1298611 in linux (Ubuntu) "[FFe] apparmor signal and ptrace mediation" [Undecided,New] https://launchpad.net/bugs/1298611 [21:17] infinity: note, we've already discussed with the kernel team and they said they would perform the pull on monday once we were satisfied with our testing [21:17] jdstrand: In a bit. [21:17] infinity: Ah, the url was changed after I edited the post. Thought it wouldn't, sorry [21:17] infinity: thanks. I don't need the full review now-- just want to alert you to it and hear your feedback [21:17] (preliminary feedback) [21:32] utlemming: Is 'http://cloud-images.ubuntu.com/releases/14.04/beta-2/' expected to exist at that location soon? [21:35] infinity: how about now? [21:35] infinity: the sync is running now, so images will appear in a few minutes [21:36] utlemming: Cool, was just link validating my emails and such. Didn't mean to make you do it Right Now. [21:36] (Not that it hurts to have it up now) [21:37] infinity: meh, I was doing it anyway actually :) [22:39] jdstrand: Pointers to the kernel and userspace diffs, with a bit more explanation than "yo, this is a feature" might be nice. ;) [22:40] jdstrand: The promise of backward compat from both sides is good, but it would be nice to see how isolated or invasive this looks. [22:40] infinity: right, that is why I didn't subscribe the release team yet [22:40] I more wanted a 'conceptually, this is ok to proceed' kind of thing [22:41] infinity: I can say for the kernel, it should be isolated to the apparmor LSM, but I'll let jjohansen comment [22:41] jdstrand: Also, do be very aware that while you don't have to get the feature correct on the first upload, you have to make damn sure it doesn't break anything else, cause this next kernel upload is likely to be the last, barring some massive bug we need to rewind for. [22:41] infinity: for the userspace, we need parser updates and a few other things, including policy updates [22:42] (Well, you should try to get it correct on the first go, but you know what I mean, pay special attention to regressions, even moreso than testing the feature itself... Broken new feature fine, broken old feature bad) [22:42] infinity: yes, we are keenly aware of that point [22:43] infinity: the packages that ship policy that need to be adjusted need to still be enumerated. for sure libvirt and lxc. others tyhicks will be investigating. those debdiffs will but a few lines (at most) of apparmor policy [22:45] jdstrand: If it's any consolation we're still making changes to the ppc64el kernel too. This "last upload" will be a lot more changeful than it should be. :/ [22:45] jjohansen: can you give infinity the location of the pull request you expect to have ready soon? [22:45] jjohansen: Just toss it in the bug. [22:45] infinity: this should be quite self-contained aiui, but I'll direct you to jjohansen for specifics [22:46] (the kernel bit) [22:46] infinity: fyi, we did the recent 2.8.95 userspace upload recently to make this FFe easier to manage for you guys [22:47] tyhicks is working on a bug fix upload for that now. next week we'll have the debdiff for signals/ptrace on top of that [22:47] Argh. [22:48] oh? [22:48] * jdstrand is talking about the userspace, not kernel [22:48] jdstrand: Sorry, that argh wasn't for you, it was for pycxx being accepted. [22:48] (and as mentioned in the bug, they can be done at different times) [22:49] ah [22:49] :) [22:49] (Who accepted pycxx?) [22:49] * jdstrand did not [22:49] :) [22:49] Was a human, not a bot. [22:49] If only we had queue auditing. [22:50] infinity: so, I think that is the big picture. jjohansen will git you the kernel git tree for review and he can answer questions re that [22:50] jdstrand: Is git a verb now? :) [22:50] we might tinker a little before I ask you to look at the FFe more officially (and before the kt pull request) [22:51] haha [22:51] that was a funny typo [22:52] and I/we'll add the MPs/debdiffs for the userspace changes when we are ready for review [22:52] infinity: thanks for the pre-review :) [22:52] jdstrand: Alrighty. Here's hoping it all goes smoothly. [22:53] yes :) [22:53] jdstrand: Conceptually, I'm fine with it, the reality could be scary. [22:53] right, that is why we'll prove it is good in testing [22:54] jdstrand: *nod* [22:55] jdstrand: Just be prepared for the let-down of a possible rejection and demand for a month of testing before landing it into the 14.04.1 kernel. [22:55] jdstrand: (But let's hope we can do better) [22:55] jdstrand: Well-documented testing will go a long way. [22:55] that wouldn't be the worst thing [22:55] but yeah, we'll get that to you === infinity changed the topic of #ubuntu-release to: Released: Trusty Final Beta | Archive: Gated Review | Trusty Tahr Release Coordination. Please don't upload things during freezes where you shouldn't, or be prepared to apologise to the release team | we accept payment in cash, check or beer | melior malum quod cognoscis [23:40] infinity: awooga! [23:41] Riddell: Thanks for being on the ball with nice release notes and upgrade pages. You were one of the few. :)