[00:00] alex-abreu: whats in the webbrowser? [00:00] so [00:00] wahts the diff? [00:00] 516 is still optional, it includes Olivier's fix + other extras (ruisky) [00:00] actually, forget that so [00:00] https://code.launchpad.net/~osomon/webbrowser-app/handle-new-view/+merge/215831 [00:00] I don't know why webbrowser-app is needed [00:00] alex-abreu: 516 is broken [00:00] it was abandoned [00:01] asac, the webbrowser bit makes clicking on links in a webpage open those links in tabs, w/o it the opened links are a bit busted [00:01] asac, yes I know [00:01] I'm not familiar with 516. [00:01] robru: can we merge the silos 005 and 001 in theory? [00:02] ToyKeeper: dont worry about that 516 thing [00:02] asac, of course... [00:02] robru: how does that work? [00:02] robru: i mean on the spreadsheet etc.? [00:02] jdstrand, there was an issue in oxide that made fixing new target=_blank & window.open() links (popups) in webpages not properly handled [00:02] so that merge & clean etc. will work in the end? [00:02] asac, i would just copy the info from one into the other, reconfig, and rebuild. [00:02] ToyKeeper, yeah its part of the past [00:02] robru: rebuild? [00:02] robru: we certainly dont want to rebuild those binaries [00:02] asac, i guess we could pocket copy if it was urgent [00:02] that took us 4 hours to get :) [00:03] ok [00:03] sounds scary [00:03] if we loose those binaries we are screwed [00:03] asac, do you want me to merge everything into silo 1? [00:03] i can binary copy [00:03] i dont think so [00:03] i feel uncomfortable with this landing [00:03] I take it we can't land both? (test them as a single unit, approve and land both) [00:04] asac, it's no trouble to binary copy from one silo to another, rebuild is not necessary, it's just the "easiest" (for me, assuming that time is not critical, etc etc) [00:04] i dont htink we need that [00:04] we can just publish both at the same time if we really need both [00:04] just dont get why folks didnt do that [00:04] asac, that is also true [00:05] asac, I cannot help you on that question, but why would it be a problem? [00:05] alex-abreu: ok so you have tested both together? [00:05] alex-abreu: can you align with ToyKeeper the testplan on both together? [00:06] asac, yes on desktop & touch [00:06] and then you guys can work on both as if it was the same? just ensure its properly documented [00:06] well I just confirmed that silo 5 successfully makes it possible for webbrowser-app to open links, so that's good and stuff. [00:06] asac, sure [00:06] activity overview even works [00:06] ToyKeeper, do you need me to setup a specific test plan (whatever the shape of it is? ) [00:07] robru, yes [00:07] alex-abreu: I don't think so... looks like both silos have the same tests. [00:07] ToyKeeper, the gist of it is that w/ the fix you should be able to click on links in e.g. facebook (links that are mean to be opened in a 'new' webpage) and they are opened properly as new tabs in the new webbrowser-app [00:08] ToyKeeper, w/o the fix they are now opened properly (new tabs are created but they dont browse where they should etc.) [00:08] ToyKeeper, basically what is writtent in the TP for silo 005 [00:10] It's kind of funny. Ever since pages got the ability to specify "open link in new window", I've been overriding that in my browser to force it to ignore that particular bit of markup. [00:10] ToyKeeper, well yeah in that instance we follow design's wills :) [00:11] ToyKeeper: alex-abreu: the testplan in the sheet for 005 is super minimal [00:11] i believe oxide potentially might break all webapps etc. [00:11] and autopilots [00:11] did and will you run those to validate this? [00:12] isnt there a testplan on the wiki? [00:12] yes I updated silo 001 w/ those [00:12] asac: I'm assuming that the webbrowser-app and webapp-container tests in silo 001 are there to cover that. [00:12] oxide being under those 2 guys [00:12] ToyKeeper: hmm. i dont see that on the frontpage [00:12] let me look at the detailed page [00:13] ok got it [00:13] sorry [00:13] the main page didnt refresh it seems [00:13] asac, I updated the main page [00:14] ok [00:14] alex-abreu: field looks grey now :) [00:14] not all of the webapp-container TP is relevant (part about install) [00:14] guess copy paste background [00:14] thx [00:14] asac, argh yeah [00:15] hehe [00:15] alex-abreu: i dont know ... the idea is for extra care we dont skip things we believe are not relevant [00:15] because wee might miss something :) [00:15] asac, sure sure [00:15] in particular late at night [00:15] asac, I am saying this w/ extra confidence too [00:16] oxide is not impacting those bits [00:16] but otherwise 100% w/ u on this [00:16] hehe [00:16] not sure :) [00:16] you cannot be otherwise 100%^ [00:17] anyhow, toykeeper probably doesnt be lieve you either and will just go through them [00:17] so i am happy [00:17] byut if they turn out to be broken that would be not so great [00:18] heh [00:18] oh just notice its install [00:18] on desktop [00:18] ToyKeeper cannot test desktop [00:18] Nope, I'm not set up for that at all. [00:19] asac, how do we handle this? afaik robru, dbarth & me did the testing on desktop [00:19] right so you dont worry [00:19] alex-abreu: i asked jfunk if someone else can help [00:20] ok [00:20] asac, yep i tested silo 5 on desktop, worgs grate. [00:20] robru: did you go through extended testplan? [00:20] or just fun testing? [00:20] :) [00:20] hehe [00:20] ... and it only took 8 tries to get the thing to stay online long enough for apt-add-repo to work. [00:21] oh, well I did the steps listed in teh bug, confirmed the bug before the patch and confirmed the patch resolved the bug (I'm talking about webbrowser-app not opening links to new tabs) [00:21] robru: right [00:21] we dont wnat that [00:21] we want the whole image tested and everything that uses the webcontainer [00:21] etc. [00:21] yeah, was just testing on desktop [00:21] so confirming that bug is fixed is step one [00:21] alex-abreu: Just to make sure, no -dbg packages are needed? [00:21] robru: right. but even there we should probably apply more care, no? [00:22] ToyKeeper, right [00:22] robru: the estplan has desktop specific tests in it [00:22] quite a few [00:22] ok [00:22] robru: https://wiki.ubuntu.com/Process/Merges/TestPlan/webapp-container [00:22] alex-abreu: oxideqt-codecs or oxideqt-codecs-extra? They mutually conflict. [00:22] ToyKeeper, oxideqt-codecs is fine [00:22] robru: if you want you could go through that and https://wiki.ubuntu.com/Process/Merges/TestPlan/webbrowser-app for desktop [00:23] ok, in a bit. still working on unity7 super-emergency-security landing. [00:23] then we woul dhav eone problem less [00:23] ah [00:23] robru: yeah, then wait [00:23] take your time on that [00:24] alex-abreu: You sure, no -extra? [00:24] The following packages will be REMOVED: [00:24] oxideqt-codecs-extra ubuntu-touch [00:24] The following NEW packages will be installed: [00:24] oxideqmlscene oxideqt-codecs [00:24] ToyKeeper, yes [00:25] ubuntu-topuch? [00:25] if that gets removed [00:25] Yeah, exactly. [00:25] it seemst hat we also need a meta [00:25] change [00:25] ogra_: ? [00:25] what do think? [00:25] jdstrand: ? [00:26] do we conflict on the -extra now and have seeded that -extra? [00:26] umm [00:26] not sure why ubuntu-touch would get trashed otherwise [00:26] I thought didrocks seeded oxideqt-codecs-extra [00:26] I don't know what anyone did with ubuntu-touch [00:26] jdstrand: yes, but the upgrade to latest oxide seems to remove that [00:26] smells as if a conflict was added [00:26] jdstrand: i am talking about oxide [00:26] which latest? ubuntu2? [00:27] ubuntu2 [00:27] yes [00:27] what silo is that in? [00:27] That's in silo 001. [00:27] jdstrand: its awfully spread across 001 and 005 [00:27] there are no packaging changes over ubuntu1 [00:27] ToyKeeper: did you get that by adding both ppas and dist-upgradding? [00:27] https://launchpadlibrarian.net/172964547/oxide-qt_1.0.0~bzr501-0ubuntu1_1.0.0~bzr501-0ubuntu2.diff.gz [00:27] ToyKeeper: or how? [00:27] a single patch is added [00:28] right [00:28] ToyKeeper: so thats not right what you do trhere, thise shouldnt get removed [00:28] asac: No, I've found that dist-upgrading usually pulls in a bunch of extras. I told apt to specifically install individual packages based on the contents of the silo. [00:28] ToyKeeper: add ppas and dist-upgrade [00:28] ToyKeeper: try if it does now [00:28] i dont think so [00:28] because archive is calm [00:28] ToyKeeper: but in general you are right, but then the list of packages you choose is not correct [00:28] ToyKeeper: we have seeded -extra so we certainly want that [00:29] oh [00:29] jdstrand: ? [00:29] maybe you tried to install oxideqt-codecs and oxideqt-codecs-extra at the same time? [00:29] they conflict with each other [00:29] she did [00:29] Yeah, that's an ongoing issue, actually. Most silos have extra packages which shouldn't be installed, and dist-upgrade pulls in too much, so I have to guess which packages are appropriate. [00:29] ah, well this is probably fallout from all of that [00:29] because alex said we should use -codecs and not -extra :) [00:29] which is not what is in our image [00:30] * asac scared [00:30] ToyKeeper: in this case, use oxideqt-codecs for desktop and oxideqt-codecs-extra for touch [00:30] right [00:30] thx [00:30] you can use just codecs on touch [00:30] she is not doing desktop :) [00:30] it just isn't representative of what is on the image [00:30] right dont do that [00:30] hehe [00:30] and vice versa for desktop [00:30] yeah right, sorry about the confusion there [00:31] well, at least we know now [00:31] hehe [00:31] In most cases, it's safe to install everything without -dev, -dbg, -autopilot, or other similar suffixes. [00:32] (-extra is part of the fix for the grooveshark bug) [00:32] right. well, upgrade the right packages and get on the testing :) [00:32] Yeah, testing it now. [00:33] rsalveti: ok i guess i will drop off in 15 or so ... will you still be around to helkp this oxide landing? [00:43] ToyKeeper, alex-abreu: hey what's the deal with webbrowser app on the phone? doesn't seem to be opening links like it does on the desktop. just click on them and it doesn't open them. [00:43] i double checked that i had the silo version, even rebooted [00:43] robru: Before or after adding silos 1+5? [00:44] ToyKeeper, oh, just silo 5 [00:44] robru, w/ silo 001 & 5? [00:44] robru, you need 1 that's what it fixes [00:44] alex-abreu, odd, why does just silo 5 fix it on the desktop then? [00:45] robru, you didn't test w/ the security team ppa or something? which oxide version do you have? [00:45] alex-abreu, nope, didn't test security team anything. desktop oxide is from the archive [00:46] 1.0.0~bzr501-0ubuntu1 [00:46] silo 5 alone was enough to make links start opening on the desktop [00:46] robru, the fix was mostly for touch [00:46] heh [00:49] It's looking mostly good... I'm just sometimes getting a blank white page instead of whatever it was supposed to load. [00:50] The tab shows the correct title in the activity widget, but no content. [00:50] The following extra packages will be installed: [00:50] oxideqt-codecs [00:50] The following packages will be REMOVED: [00:50] oxideqt-codecs-extra ubuntu-touch [00:50] The following NEW packages will be installed: [00:50] oxideqt-codecs [00:50] The following packages will be upgraded: [00:50] liboxideqtcore0 [00:50] robru: Use -extra. (discussed in the scrollback) [00:50] I guess we need to update the seed or something? [00:50] hm [01:00] ToyKeeper, alex-abreu : ok it's looking good to me on the phone [01:00] robru, ok good [01:00] Yes, so far so good. [01:01] The webapp-container plan seems to be for the desktop though. [01:02] ToyKeeper, it is mostly for desktop, I dont want to influence you :), but on touch the webapps (gmail, facebook, twitter etc) are using the webapp container and one should make sure that clicking on *external* links fw the new page opened to the webbrowser-app [01:03] ToyKeeper, we dont have tabs in the container and foward all external links & popups (window open, etc.) to the main browser [01:03] this should changed, silo 005 being only concerned w/ the browser [01:04] alex-abreu, yep, confirmed. when I'm in the Twitter *app*, and I click a link, it opens in the browser app, and unity8 shows twitter and browser apps running independently. [01:04] but only with both silos 1 and 5 [01:06] So, to test target="_blank", I just made a quick page. It was easier than finding one online. [01:06] http://toykeeper.net/tmp/newtab.html [01:06] ToyKeeper, sure you can also check e.g. twitter [01:06] I'm checking the facebook app now. I don't have a twitter account though, and their site apparently won't let me do anything except create an account. [01:07] (weird, since I can view tweets anonymously on my desktop) [01:09] robru: on desktop use codecs... on touch use -extra i was told yes. [01:09] ah ok [01:10] robru: didnt you want to test on desktop :)? [01:10] * asac sees REMOVED ... -touch [01:10] In any case, the facebook app seems to behave properly. The xkcd viewer also works nicely, though it doesn't really have external links to check. [01:10] ok [01:10] well, once you feel its good to go and you have tested it to all extend [01:11] just coordinte with robru that desktop also has been tested according to testplans and then give an ack on the landing sheet for those two silos [01:11] ... except if you try to click a link from mobile facebook to regular facebook, and then it goes to the full browser and asks the user to log in again. [01:12] Not exactly our problem though. It does the same on other mobile clients. [01:12] ToyKeeper, yeah the cookies are not shared, webapps run w/ specific "confined" cookie stores [01:12] Even with shared cookies, I don't think m.facebook cookies work on www.facebook. [01:12] ToyKeeper, this might change later next cycle [01:13] I've occasionally used m.facebook + dillo as a primary interface for it. So much faster! So much less bloated! So much less likely to crash my main browser! [01:13] mmh it should from what we tested [01:13] (though on Chrome, it hasn't been as much of an issue) [01:14] m.* is definitely lighter & easier on the eyes [01:15] ... and dillo is faster/smaller than lynx or w3m, despite having a GUI. [01:16] So, slight issue in the facebook app... can't actually write anything. Tapping a comment or message entry doesn't invoke the OSK. [01:16] that is a known bug [01:17] existed for ages [01:17] ok i have to crash; i see touch testing is going on and robru will rerun the desktop testplan to be sure [01:17] then you guys land if nothing breaks :) [01:17] 'k. [01:17] (and annoying) [01:17] thx asac [01:17] asac, sure thing, thanks. goodnight [01:17] For touch, I see no reason not to land it. [01:18] ToyKeeper: then once robru has finished his desktop testplan thingy [01:18] I haven't done AP tests on it though; anyone have a link to AP test results? [01:18] give it an ack in the sheet and land both silos [01:18] ToyKeeper: we dont have those results [01:18] thats part of testing [01:18] at least run the webbrowser etc. things [01:18] Yeah, just need to figure out how. [01:19] how to run APs? [01:19] never done? [01:19] robru can explain/teach for sure, but thought you had done that before [01:19] Nope, not yet. I wasn't really involved in the QA team's main tasks before last week. [01:19] alex-abreu: did you run APs? [01:20] asac, ToyKeeper I know how, one sec [01:20] anyhow, we need to run the ones for this thing [01:20] thanks [01:20] so do that and desktop testing, then go for landing i guess [01:20] cu in a couple hours [01:21] ToyKeeper, https://wiki.ubuntu.com/Touch/Testing#Running_Deb_tests [01:21] and remmeber to kick off an image once its in :) [01:21] asac, for the webbrowser-app & this silo no [01:21] hehe [01:21] safes time [01:21] alex-abreu: its part of the testplans [01:21] isnt it? [01:21] anyway, needs to be done on all silos [01:21] ToyKeeper, so, uh, just make sure the PPA is enabled on your phone first, then do 'phablet-test-run -p webbrowser-app-autopilot webbrowser_app' [01:21] asac, yes [01:21] never land without running APs [01:22] ok do that then still i guess [01:22] before landing [01:22] thx [01:22] Thanks. [01:22] ToyKeeper, the above command being run from the host machine [01:26] Quick relocation, then I'll be running those. :) [01:26] rsalveti: you around? The testing has finished for the autopilot silo 007: http://q-jenkins:8080/job/autopilot-release-gatekeeper/112/testReport/ [01:26] I was talk to ping you when it had :-) [01:27] told* [01:28] veebers: set your landing to testing done i guess [01:28] and reply to the mali i sent [01:28] folks are running qa sign off on oxide still [01:28] after that and afte rwe have an image we could try that [01:28] asac: I'll check the results now [01:28] but didrocks will have to do that [01:28] ah opk [01:28] yeah [01:29] i am out for couple hours catching some sleep; jdstrand probably also can help a bit [01:29] if rsalveti is off [01:29] cu [01:29] * rsalveti waves [01:29] just got back from dinner [01:30] veebers: alright, can be landed I guess [01:30] asac, rsalveti: fyi there were 2 failures both with ubuntu_weather_app. They look like test failures not related to the autopilot change [01:30] normally I would re-run the gatekeeper job with just that test suite to be certain [01:31] yeah, they shouldn't be related [01:31] veebers: alright, will land it [01:31] rsalveti: It's really up to you to decide if that's required I guess :-) [01:31] rsalveti: awesome [01:33] veebers: done, should be in release soon [01:33] * rsalveti checking backlog [01:34] rsalveti: sweet, what else is left to do wrt to that? I guess just respond to the mail asac sent right? [01:34] veebers: I can reply it [01:34] rsalveti: ack, thanks' [01:40] ToyKeeper, sorry, that page was slightly out of date (predating CI Train). I've updated it, you might want to give it a glance: https://wiki.ubuntu.com/Touch/Testing#Preparation [01:40] ToyKeeper: can you guys summarize what was decided regarding silo 1 and 5? [01:40] ToyKeeper, alex-abreu : so I've run AP tests for webbrowser app, it's all looking really good to me, what do you guys think? [01:41] rsalveti, asac decided to go ahead with those pending our successful testing of them [01:42] robru: right, so we're landing both I guess [01:42] robru, I just ran them too, all passed [01:42] yes [01:42] great [01:42] oxide-qt is seeded though, would need an archive admin to approve it [01:42] rsalveti, do I have to do some paperwork or are you the one to handle it ? [01:43] rsalveti, jdstrand ^ [01:43] and at this point I'm not sure they will accept it [01:43] did we raise this with the release team already? [01:43] rsalveti, I think so yes, jdstrand can comment further on this [01:43] he is archive admin I think [01:44] same for webbrowser-app [01:44] robru, can I update the 'testing done' ? [01:44] alex-abreu, yep [01:44] he's not part of the release team [01:44] https://launchpad.net/~ubuntu-release/+members#active [01:45] ok I thought he was too [01:45] alex-abreu, wait, what's the plan here? the goal was to land this stuff in the trusty desktop release? [01:46] robru, afaik ideally yes, not sure what was specifically discussed between David & didrocks [01:46] alex-abreu, well I just did this super-security-emergency unity landing, I'm not sure it's a good idea to land anything else, ultimately it's up to the release team [01:47] robru: and it seems they just published a possible final image [01:47] alex-abreu, we might hope to land in trusty-updates so that people get the update after installation, but yeah, I think the ship sailed on getting this in the final image. [01:47] so I wonder if we'll be able to release the oxide-qt and webbrowser-app fixes [01:48] right .. [01:48] infinity: guess we're done with the archive already right? (for seeded packages) [01:48] robru, rsalveti can we have them for touch ? [01:48] we got two important fixes for touch, but both packages are seeded in a few other images [01:48] alex-abreu, that's a good question. [01:48] not so sure [01:49] we would need someone from the release team to put the final words on the subject [01:49] ok [01:49] wonder why this wasn't discussed before [01:50] I thought it was & has been [01:52] we really need to land this [01:52] well I just asked in -release [01:52] anyone would be avail from the release team? [01:52] ok [01:53] ToyKeeper, when you get back let me know how you feel about oxide, need your final +1 on that [02:02] rsalveti, thx for the email === Ursinha is now known as Ursinha-afk [02:09] alex-abreu, rsalveti : ok, so basically it sounds like I'm gonna hit publish, that stuff is gonna get stuck in proposed, and then we have to poke them to get it accepted to -updates, that sound ok to you guys? [02:10] robru: yup, looks like the way to go [02:10] robru, yes [02:10] robru, no paperwork needed? [02:10] yup, we have bugs already, don't we? [02:10] rsalveti, yes [02:10] then the usual SRU paperwork [02:10] alex-abreu, not sure exactly if we have to do the full SRU rigamarole for this... i hope not [02:10] https://wiki.ubuntu.com/StableReleaseUpdates [02:11] robru: at least the updated description in the bug would be good [02:11] rsalveti, we need the paperwork for this? [02:11] ok [02:11] impact, test case and potential regression [02:11] I know the drill [02:11] great [02:13] rsalveti, alex-abreu : ok guys both silos 1 and 5 are published [02:14] robru, I am updating the description SRU style [02:17] robru: +1 except that I'm just starting the AP tests. [02:17] ToyKeeper, ok great. keep us posted ;-) === Ursinha-afk is now known as Ursinha [02:20] Sounds like it's already published? [02:20] robru: can you please reply that email saying we decided to go with SRUs? [02:21] ok [02:22] thanks :-) [02:22] oh I was about to [02:22] but better someone from ci-eng to do it [02:22] I updated the description [02:25] alex-abreu, ok, I'm going for dinner shortly, can you ping #ubuntu-release with that bug number and ask them to accept those packages into trusty-updates? [02:27] robru, except that for oxide its not a branch perse, would it be fine? [02:27] alex-abreu, I'm not sure what you mean. both packages are in UNAPPROVED, release team needs to shepherd them through proposed and into updates [02:28] brb [02:28] forget about it they are already in proposed [02:29] forget what I said [02:31] alex-abreu, what? no they're in UNAPPROVED. they won't go to proposed until somebody from release team ACKs them [02:32] well quoting you above [02:32] "that stuff is gonna get stuck in proposed, and then we have to poke them to get it accepted to -updates" [02:33] anyway [02:33] Sorry for the delays, it's a little odd diving into this for the first time right before release. Weird timing. [02:35] ToyKeeper, yeah no worries, the whole thing is bad timing [02:40] Hmm, 37 tests run for webbrowser-app, 4 exceptions, but the test harness seems to think that's okay. [02:41] ToyKeeper: that also is normal [02:41] ToyKeeper: that is the AP tests, right? [02:41] Yes. [02:41] yeah, that's normal [02:42] I feel like I should just leave it running AP tests whenever I'm not using it, simply because it looks neat when the phone is possessed. [02:42] at least that was the case the last few times I ran them, and I asked and someone else said they saw the same thing [02:42] hehe, it is kinda wild [02:43] jdstrand, hi did browser and oxide land ok [02:43] pmcgowan: I'm going to point you at alex-abreu [02:43] pmcgowan, not yet, ... we are passed the freeze [02:43] he has been dealing with it this evening [02:43] pmcgowan, not possible to land at this point from what I have been told [02:43] pmcgowan, we are tying to SRU them [02:44] ah too bad [02:44] ok [02:44] yeah [02:44] pmcgowan, I'll forward you the emails for the evening [02:44] thanks [02:45] pmcgowan: if we can get them in -updates in time, we can spin a new touch image with -updates enabled by default [02:45] but we'll see [02:46] still trying to find someone from release to ack then out of the unapproved queue [02:46] up to the release management team [02:46] rsalveti, ok [02:48] alex-abreu, I was more concerned with desktop but a 0-day SRU will be fine [02:48] and touch images march on anyway [02:49] yeah I guess [02:49] I still had hopes [02:49] shame they were ready seemingly early enough [02:49] pmcgowan, the oxide build took a long time to get there in armhf so it delayed the checks & validations [02:49] right [02:57] alex-abreu, ok, I'm heading out for dinner now. will be back a bit later. [02:58] robru, did not get any response yet from release, not sure what's next & if I can do more [03:07] It doesn't sound like I can help with any of the current issues, but if I'm wrong, let me know. [03:45] robru, ok I'll be heading out, I'll send an email to dbarth so that it can pick it up in a few hours for him [04:38] hello folks [04:38] I do not know who is active :) maybe robru or rsalveti [04:38] robru: rsalveti: The UITK in the silo9 is good to go [04:40] bzoltan, looks like it needs QA signoff? I guess om26er for that... [04:49] bzoltan: yep, QA signoff is needed there [04:50] bzoltan: and if you want it to trusty you need SRU it https://wiki.ubuntu.com/StableReleaseUpdates [04:51] Mirv: I need it on the image... does it mean trusty: [04:51] ? [04:52] Mirv: is this SRU thingy for real??? D I really need an other step of paperwork for each and every simple release? [04:52] bzoltan: well the release is today, and trusty is done, there is nothing that can go into the non-updates archives at this point [04:52] I don't know the plan for divide between trusty SRU:s (maybe in the beginning?) and u-series [04:53] the fact that Touch lives in its own world doesn't mean desktop doesn't do releases :) [04:54] Mirv: we have convergence specific tasks and targets [04:54] Mirv: it is a joke [04:54] bzoltan: if they are not targeted to trusty, then I'd guess it's u-series. [04:55] bzoltan: the only problem with u-series I can see is that it doesn't currently exist [04:55] Mirv: I have no idea ... [04:55] so that's why I gather some people will do a couple of SRU uploads to trusty instead meanwhile [04:55] Mirv: so in fact we stop releasing anything? [04:55] bzoltan: trusty images will be spinned until u-series is open [04:55] but since trusty is released, it'll need to go through SRU:s, especially for components that are shared on desktop too like UITK [04:57] Mirv: what is the point of the whole CI and all the AP tests, silos and multiply manual signoffs if we still need yet an other round of bureaucracy round to push things out of the door? [04:57] bzoltan: I'd have thought there had been some good meeting between team leads about what happens at this point, but I guess not :S [04:57] bzoltan: maybe there is no point since the last minutes' procedure was finalizing the trusty release, and it's now done [04:58] bzoltan: the only thing is that maybe trusty touch image won't be released today yet and they want a few bits like the Qt fix in still [04:58] in that case a couple of SRU:s and respins from trusty [04:58] Mirv: Now that you say... it all sounds logical and sensible... for the good old desktop Ubuntu. But with the Touch our QA machinery and release process should be good enough for landing anything on Trusty even beyond the release date [04:59] bzoltan: yes, I agree, but note that in your specific case you're on the desktop too so it does not apply [04:59] Mirv: for me the question is how to land stuff on the phone image and how to merge MRs back to trunk [04:59] for touch-only stuff I agree that either the current bureacracy goes away or the SRU one isn't added on top of it in whole [05:00] bzoltan: yep, and to that question I'd have hoped you have an answer, but we need to wait for alexander & friends probably at this point for that [05:00] bzoltan: Silo 9 is ready for QA? [05:01] ToyKeeper: yes it is [05:03] Mirv: I have more questions than answers :) and... just for the record. I am not challenging the present model of releasing stuff on Trusty. All I wish to know, how to land MRs on our very-very own trunk [05:04] Mirv: this new measure will drive us away even more from th trunk and we will use the staging for real. So the trunk will just age out and our release cycles get longer and longer. [05:04] bzoltan: Sorry, I didn't see your messages at first. [05:04] ToyKeeper: no worries :) [05:05] bzoltan: yeah, I'd be very interested to know as well :) I see some other similar wondering in the backlog. [05:05] bzoltan: note that the period of unknown only lasts until we're able to land and release from u-series, probably on Tuesday or so [05:05] so essentially I'd guess only what to do today is the question mark [05:05] ToyKeeper: I have been running the AP tests for apps in the last hour.. . all seem green, but an other eye on it will be good for sure :) [05:05] bzoltan: Got a link to a test plan for this? The landing proposal just links to a generic template. [05:06] Mirv: if it is about a day or two... then I do not care much [05:08] bzoltan: yep, I don't think there's any reason to worry. worry only when you start hearing that we'll continue polishing trusty for a while :) [05:08] Mirv: LOL :) yes, I know what it would mean... [05:09] Mirv: where the risk is not little, 14.04 is our LTS, I could imagine that the quality-stability-security of our LTS is important. [05:15] bzoltan1: If I understand correctly, this isn't an urgent landing? I get the impression that absolutely nothing non-critical can land for now, due to the trusty release coming up later today. [05:24] ToyKeeper: I will not fill an SRU for it ... but it would be good to land it. === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [06:22] Mirv: can you look at the 6 failing tests on the dashboard meanwhile? [06:22] (hey btw ;)) [06:24] didrocks: hey! yes I'm running the tests already [06:24] thanks! [06:24] * didrocks waits impatiently on a release team member to be around… [06:43] bzoltan: Two of the bugs fixed in silo 009 seem to check out. Two others I'm not sure how to test, and I'm having some difficulty getting AP tests to run on it. [06:44] ToyKeeper: OK. Is there anything I could help? [06:53] bzoltan: I'm checking the base image, but I'm not sure yet why the AP tests aren't running. [06:55] ToyKeeper: timp was complaining about similar issue yesterday. I u-d-f --channel=devel-proposed flashed my device today, made it writable, cloned my network setting, enabled the Silo9 PPA and the tests were running fine. [07:01] Saviq: awake already :)? [07:01] Saviq: we wanted to start picking up your ted patch [07:03] bzoltan: I'm not sure what the exact issue is. The same tests worked just fine a few hours ago with a different silo. [07:04] ToyKeeper: yes, that is a common black magic stuff... I would reflash with --boostrap and --wipe that device and start it from scratch [07:05] Yeah, just tried it. [07:07] ogra_: tell us when you are around, you will have maybe a special mission! [07:09] asac, yeah, here [07:11] Saviq: we startd building it in a silo proactively fr now [07:12] asac, saw that [07:12] guess you needed to do some review still [07:12] but we have like < 1h time [07:12] :) [07:13] sil2100: didrocks: I can reproduce none of the 301 image test problems, so the tests seem flaky not totally failing [07:13] Mirv: ok, already some "good" news [07:14] weird, though, where's the extra flakiness coming from [07:14] Mirv: do you see some unity8 crashes? [07:14] Mirv: or like if the shell was going away? [07:14] didrocks: no, it does not look so. so I've run dialer-app, weather app, sudoku app and unity8 AP:s [07:15] no crashes with those [07:16] Mirv: I meant, on the dashboard? [07:16] like if you look at the failure [07:16] does it seem the shell disappeared? [07:16] bzoltan: I'm probably missing something obvious. I'm getting mostly-bailed tests even with just the base image. [07:18] asac, so that < 1h deadline is for what? for 14.04? didrocks, could we still make the desktop session fix for that, or is it too late? [07:18] Saviq: <1h is for touch things [07:19] desktop i dont know [07:19] unit8 doesnt need an image [07:19] on desktop [07:19] Saviq: unity8 can wait, it will be in -updates [07:19] so you dont need time [07:19] Saviq: not installed by default [07:19] didrocks, ah right [07:19] * Saviq files a SRU bug already [07:19] yeah ;) [07:20] Saviq: can be a 0-day SRU [07:20] Saviq: but please focus on ted's one first [07:20] didrocks, I have [07:20] thanks! [07:21] didrocks: no unity8 crashes in the dashboard either. only dialer-app and system-image-dbus crashes on the latest image (2 crashes), and additionally lists-modems crash on #300 [07:22] looking at the logs a bit more regarding the failing tests [07:23] asac, hmm wait, your email suggests that *apps* not starting is what the bug will fix, which is not the case [07:23] Mirv: ok, I was wondering if this would be a consequence [07:23] asac, the bugfix is for a rather random unity8 crash [07:23] Saviq: i said that? [07:23] i was saying app starting can creawte random crasehs [07:23] asac, ok wait [07:23] didrocks, Mirv: dialer-app fails in the same way before :/ I wonder if boiko made some progress there [07:23] in unity8 [07:23] asac, yeah, I should open my eyes [07:23] Saviq: i am sure if you use free strings to then later start stuff it can also caus apps to fail though [07:24] :) [07:24] e.g. just get garbage [07:24] and send that to command line [07:24] doesnt sound too far away :) [07:24] o/ for oxide, everything good? [07:24] no [07:25] what's up? [07:25] it took the whole night to validate [07:25] at which point release team wasnt available anymore [07:25] didrocks: no sign of unity8 disappearing either. the next tests after failing ones succeed normally. there's eg. a tab switch / index errors (UITK usage, probably something wrong clicked or some delay), a dbus timeout [07:25] dbarth: lots of confusion about you not using hte sam silo for weebrowser and oxide as well [07:25] noone - neither me nor jdstrand - remembered that it was more than oxide [07:25] we did it anyway [07:25] Mirv: ok, thanks for checking! [07:26] dbarth: yeah, yesterday, it was clearly only oxide [07:26] dbarth: and the other fixes were "nice to have" but not related to the other failure [07:26] asac: didrocks himself advised not to, you were inthe call! ;) [07:26] hence the decision to separate [07:26] right [07:26] dbarth: so, they are not linked? [07:26] we don't need webbrowser-app [07:26] so anyway, the oxide build is tested / etc. [07:26] didrocks: oxide fixes links in webapps, not in browser [07:27] didrocks: the other fix is for links in the browser itself [07:27] so i was told olnyu oxide [07:27] don't ask why, it's insane [07:27] then alex said without webbrowser it will not work [07:27] asac: i thought it was only oxide as well initially [07:27] ah, so that explains [07:27] and that's why we separated [07:27] then digging into the 2 different fixes we found we needed the 2 [07:27] then i harassed people on this channel for hours to get a silo, remember ;) [07:28] i didnt [07:28] noone remembere by alex [07:28] but doesnt matter [07:28] my brain was fried at 3:30am [07:30] i think you mostly need webbrowser-app out of unapproved for the touch image; for the desktop, a 0-day sru seems to be the accepted solution [07:31] asac, didrocks: should i ping #ubuntu-release as well, or are they just tired of hearing about that browser thing? ;) [07:31] dbarth: they are sleepiung it seems [07:31] they are pinged [07:31] ok, so i won't add to the stress then [07:33] sil2100: btw I didn't run dialer-app tests with SIM card, so those might be worth rerunning with SIM inserted [07:35] Mirv: ok, doing that, but from the looks of it in the logs it's the very same failures we're having since some time ;/ [07:35] sil2100: yeah, it looks so [07:36] Mirv: I don't see any updates on the bug, so I'll just have to wait for boiko to get online [07:40] ok [07:42] Since I'm suspecting that the reason for this failure is a crash actually [07:42] Mirv: all pass here with sim inserted... [07:43] sil2100: ok, good [07:43] btw I did have one weather app test failing on one run, but it was a different one from what's in dashboard. the next run succeeded with all passing. [07:44] sil2100: you think you could try installing the python-evdev on device and do a quick smoke test? [07:44] it is 99.99% safe, but since we are waiting anyway and the landing will install that by default i think would be good [07:44] Mirv: and on your side, can you add silo 004? [07:44] Mirv: and install/try the new unity-mir [07:44] sil2100: maybe you can add it as well ^ [07:45] didrocks: ok [07:45] Saviq: you are running the AP tests on it? [07:45] yeah, just put all we want to punt in on the image, test everything together [07:45] didrocks, yes [07:46] Mirv: maybe we should double check meanwhile ^ [07:46] didrocks, and unity8's TestPlan [07:46] yeah ;) [07:46] should be done in 15 mins [07:46] tops [07:46] great! [07:46] ToyKeeper: can you start testing that silo meanwhile? (landing-004) ^ [07:46] ToyKeeper: it's the top priority for now [07:46] asac: sure, what do you want me to smoke-test? [07:47] Oh, hey, that's the bug I ran into earlier. [07:47] asac: like, everything? [07:47] sil2100: the silo where we did the python-evdev [07:47] Ah, k [07:47] ToyKeeper: you are all suited to test the fix then! :) [07:47] sil2100: but maybe just punt the rest we want to land on the image as well for the test [07:47] didrocks, asac: ok! [07:48] didrocks: I'm a little fried, but I'll see what I can do. :) [07:48] ToyKeeper: thanks :) [07:48] ToyKeeper: guess thats your last thing to do today :) [07:48] davmor2 will be around soon [07:48] between you, Saviq and Mirv, we'll have good coverage === psivaa changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: psivaa | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: - [08:15] happy release day ! [08:15] * ogra_ puts on his missionary dress ... [08:15] didrocks, what can i do for you monsieur [08:16] ogra_: hey! imagine we want to do one image build from a ppa [08:16] is it possible/hard/impossible? [08:16] ;) [08:17] hmm, not sure, i know cjwatson knows though ... i think technically it should be possible, practically i have never done that [08:18] (i know we had it for moblin image but i dont know if i.e. the PPA had to be mirrored for that to an internal machine etc, that was long ago) [08:18] *images [08:18] ok, let's see, that would be for unity-mir [08:18] now that the rest is handled [08:18] ah, wait, heh [08:18] we *did* build from a PPA initially [08:18] silly me [08:19] *but* ... that needs a change to the build system (livecd-rootfs needs PPA sources.list and key added) [08:19] not sure how we could get that into the main archive now [08:20] asac, didrocks, ToyKeeper, +1 on silo 008 from me [08:20] erm 004 [08:20] sure? :) [08:20] lol [08:20] ;) [08:20] * sil2100 is still running tests with silo 004 and 007 [08:20] didrocks, i'm not sure this is possible at all at this point of the release [08:20] The bad news is I managed to repro the issue. The good news is I forgot to install the silo first. (or is that the bad news?) [08:20] Anyway, trying to repro w/ silo. [08:20] ToyKeeper: ;p [08:20] ogra_: right, requires archive upload, right? [08:20] sure, landing-004, I've ran autopilot on mako and manta, and were good, launched 6 different webapps, no crash in sight [08:21] think its fine, colin etc. are now there [08:21] Mirv: what's your status with the silo? [08:21] so the engine is moving [08:21] asac, yes and i'm not sure the buildds would pull it from -updates [08:21] right [08:21] they might [08:21] lets not do it unless release team disappears :) [08:21] hehe [08:21] heh [08:21] then we bringup our own image build infra quickly :{P [08:21] j.k. [08:21] kind of unlikely on release day :) [08:21] i heard we have everything charmed now [08:22] yes, thats why i was saying that [08:22] hehe [08:22] :) [08:22] soo [08:22] but you never know... UK might get global power outage next [08:22] my phone was immediately responsive when i woke it up today [08:22] \o/ [08:22] havent had that in weeks [08:22] nice [08:22] more good news [08:22] usually it took about a minute until it updated the clock on the greeter [08:24] * ogra_ goes to check #ubuntu-release-party [08:24] didrocks: it will be possible once I get out from under release and finish the livefs-in-LP thing; right now it would be a lot of manual effort [08:25] cjwatson: ok, no worry, I was looking for a one time shortcut while nobody was around [08:25] didrocks: one unity8 AP run done, 100% pass [08:26] Mirv: did you try to dogfood/play with it? [08:26] cjwatson, just for academic purpose, do you know if the buildds would have picked up livecd-rootfs from -updates ? [08:27] Mirv, didrocks: unity8 run done here as well, 100% pass [08:28] didrocks: a bit, running apps, testing indicators etc [08:28] it's just so slow running unity8 AP:s for me (> 30min), that's why it only just finished [08:28] It's the same here... it's been like that since the introduction of the new scopes === popey_ is now known as popey [08:29] Morning all [08:29] Morning [08:30] hey davmor2 [08:30] ogra_: hard to be completely sure, but my reading of the IS make-chroot.sh used to create the relevant chroots is that their sources.list includes -updates, so I think yes [08:30] thanks :) [08:33] I think this must be the longest I've used UTouch without anything breaking. [08:33] \o/ [08:35] sounds like a winner :D [08:35] hmm [08:35] did someone check the dashboard lately ? [08:35] AP tests surely did look better before [08:35] didrocks, do we have a process for SRU with silos? [08:36] Saviq: nothing special, apart from opening SRU bugs and referencing it [08:36] Saviq: CI Train looks at -updates [08:36] didrocks, yeah, but then... SRU process says it should be fixed in current release before asking for SRU... [08:37] cool [08:37] Saviq: ah, you just need to request 2 silos [08:37] Saviq: but at first, we'll be in sync between -updates and U [08:37] as usual [08:38] and the copy will be handled by the release team [08:38] as long as we are in sync [08:38] * Saviq loves it that "SRU" is onomatopoeia in polish for something crashing down... [08:39] didrocks, asac: now that packages are published, let us know when an image is ready for re-testing [08:39] i can hop on the call if you need me btw [08:40] ToyKeeper: did you install it now and test? [08:41] asac: Yes, I can't get it to break now. [08:41] ToyKeeper: with the unity-mir fix installed? [08:41] didrocks: ^^ [08:41] asac: Yes. [08:45] Saviq: are you still happy with the unity-mir aptch? [08:45] Saviq: because this might go in now [08:51] +1 [08:52] ... and I think it's time to sleep now. I wasn't really planning to be on call for 15 hours today. [08:52] ToyKeeper: good night [08:54] good night ToyKeeper [08:56] asac, yes, I am [08:56] Saviq: awesome... we land that now in the same image shot [08:56] lets hope all three things dont cause issues so we get allt his out [08:56] it will be a good shot! [08:56] yeah [08:56] the golden shot [08:56] hehe [08:57] heh [08:59] ok, autopilot published in the release pocket now [08:59] 3 remains flying [08:59] didrocks: relase pocket == updates? [08:59] asac: no, release pocket as == release :) [08:59] I'm disabling auto-accept now - I think we want manual control today [08:59] err real release pocket? so it would be available for image? [09:00] asac: yes [09:00] cjwatson: sounds good [09:00] asac: autopilot isn't on non-touch images so it's not a problem [09:00] didrocks: ok we just wait for the unity-mir thing then? [09:00] and that takes another hour? [09:00] cjwatson: we don't plan to publish anything anyway, but better to disable auto-accept :) [09:00] cjwatson: aye. thx [09:00] unity-mir shouldn't take another hour [09:00] oh that w ould be fantastic [09:00] asac: and oxide-qt and webbrowser-app to be published in -updates (the publication didn't finish) [09:01] it's publishing to -proposed now, then the next publisher run should copy it to release [09:01] so 10 minutes :)? [09:01] hehe [09:01] guess if its manual publishing, we can even be more precise when to run the job [09:01] oxide-qt and webbrowser-app are publishing to -updates now, indeed [09:02] no, we can't, and it wouldn't help :) [09:02] I think maybe 40mins [09:02] hopefully after release we can land the apt-ftparchive source caching stuff that Michael did for us [09:02] that should make a non-trivial difference to release pocket publishing times [09:03] Just killed it [09:03] a first big mvo goodie :) [09:03] nice [09:03] davmor2: killed? [09:04] asac: Yeap just kept opening apps with the music player playing in the background, I'm assuming that apport is going crazy in the background currently [09:04] well, three minutes or so, but you really notice it when you're 'tail -f'ing the log [09:04] davmor2: with that patch? [09:04] asac: yeap [09:04] davmor2: unity8 crashing? [09:05] and unity8 has crashe [09:05] d [09:05] davmor2: i think there might be lifecycle thing [09:05] as well [09:05] davmor2: can you kill apps and open? [09:05] yay, the first time i see app updates [09:05] intead of opening more apps [09:05] wgrant: "Calculating binary filelist" in the publisher seemed unusually slow right now [09:05] s/right now/just now/ [09:05] davmor2: in any case, extract the crash file [09:05] so saviq can check that [09:05] ToyKeeper, Hi! are you around ? [09:05] and see if its a ddifferent one [09:06] om26er: no [09:06] cjwatson: Hm. How slow? [09:06] om26er: she went to bed a few minutes ... what do you need? [09:06] asac, just wanted to know if she was testing anything [09:06] om26er: no, but you can help i guess [09:06] check with didrocks and davmor2 [09:06] wgrant: about three minutes [09:06] other folksa re testing :) [09:06] Odd. [09:07] it's usually more like a minute for the release pocket [09:07] om26er: she was talking with bzoltan [09:07] om26er: so I guess line 18 [09:07] om26er: however, we'll have a new image in the next couple of hours [09:07] om26er: all testing focus on it would be appreciated [09:07] didrocks: is that line relevant for release? [09:07] asac: nope [09:07] hence my big "hint" :) [09:07] guess we shoudl rather have omer help on our image then [09:08] right [09:08] or popeys apps :) [09:08] so I just flash the latest image, and test the hell out of it ? [09:08] om26er: can you add more packages to it? [09:08] om26er: its not the latest, you need to take latest image and install afew inflight things [09:08] om26er: enables -updates if not [09:08] so you can test the future :P [09:08] om26er: tell me once you are there, depending if things are published or not, I'll tell you what to install [09:09] om26er: just flash latest and keep me posted [09:09] didrocks, ok will do [09:09] asac, didrocks: this is what crashed _usr_bin_unity8.32011.crash [09:09] _usr_lib_arm-linux-gnueabihf_qt5_bin_qmlscene.32011.crash [09:09] Image looking good so far [09:09] davmor2: mind retracing/giving to Saviq? [09:09] please do [09:09] but apport-cli it first [09:09] didrocks: yeah that's the next job [09:09] Saviq: will be denied because of new unity-mir I'm afraid [09:10] didrocks, what will? [09:10] Saviq: unity-mir isn't the latest in release/updates yet, so retracing will fail for now (until it's copied over) [09:11] didrocks, yeah that's fine [09:11] didrocks, but apport-cli adds some fields to the crash file [09:11] didrocks, like package version and such [09:11] didrocks, without which I can't even begin to retrace [09:11] Saviq: if you want to reproduce, open the music player and have it play, that keeps it open in the background then just keep opening apps I got 34 apps open before the crash :) [09:11] Saviq: yep [09:11] davmor2: seems oom related to me [09:12] indeed [09:12] davmor2: and we are not at that level of polish yet [09:12] so doesn't seem valid for what we try to do [09:13] didrocks, Saviq: Indeed I was just trying various ways of crashing it, I'm going to just try opening and closing apps with music playing next and see what that does [09:13] yeah [09:13] its good [09:13] we have a new crash [09:13] most likely not critical [09:13] davmor2, still, if you can apport-cli the .crash, and send it up for me [09:13] but important to check [09:14] davmor2: yeah, don't do exploratory testing, keep things in the known territory [09:14] if its sitll the same that we wanted to fix with unity-miore [09:14] didrocks: That was just testing opening apps [09:15] davmor2: you do the right thing [09:15] just start apps and close them for 30 minuntes :) [09:15] there shouldnt be a single unity8 crash now with the fix if doing that [09:17] Saviq, didrocks, asac: that could be a problem http://paste.ubuntu.com/7266701/ [09:18] * asac checking [09:18] clearly oom killer related, did you close the opened apps? [09:18] riht get phone into clean state i guess [09:18] didrocks, nah, that's apport-cli complaining [09:18] reboots dont loose? [09:18] er reboots dont wipe crashes? [09:18] asac, they don't [09:18] didrocks: no unity8 crashed all the apps have gone [09:18] davmor2: guess make a clean reboot and use it again [09:19] who knows what this crash left behind [09:19] on shared mem etc. [09:19] that still might not allow for the crash to be processed... I've seen that a few times :/ [09:19] how can one stop unity8 to get more mem? [09:19] service lightdm stop? [09:19] sudo -u phablet -i stop unity8 [09:20] davmor2: try runb that first [09:20] and then process [09:20] Saviq: indeed [09:20] dont stop lightdm [09:20] (with unity down) [09:20] so rebooted same issue [09:21] so the collect may have been blocked [09:21] davmor2, please send it up anyway, I'll see what I can do [09:22] Saviq: give me a second I'm going to try the other crash file and see if that unblocks apport collect [09:23] om26er_: ping [09:24] bzoltan, Hi! I just wanted to know if the uitoolkit silo needed testing, but now I am told that's not critical for the release, so was asked to do the image testing [09:25] oxide-qt and webbrowser-app published [09:25] in -updates [09:25] all is on unity-mir now :) [09:25] om26er_: yes, she had some problems with application APs [09:25] (published in -proposed) [09:25] om26er_: I would need your help in the #sdk if you have few minutes [09:26] valide candidate in britney [09:26] and accepted [09:26] bzoltan, sure, my device is being flashed so I can probably help [09:26] we are a publisher cycle away from kicking the image! [09:26] * asac sends some luck to unity-mir to get a perfect publisher pipeline :) [09:28] Saviq: http://people.canonical.com/~davmor2/crash/ it wouldn't let me [09:28] bzoltan: so your landing will not make the image, thats why we prioritize the testing of our imge business for the next couple hours. whats your landing about? [09:29] \o/ [09:29] sil2100: whats the source of happiness :)? [09:30] davmor2, so a crash for unity8 that's 120K is not something that's of any use... [09:30] ack [09:30] Saviq: you dropped all the code? :p [09:30] asac: it is business as usual ... we would like to continue working the same way as yesterday and before. We have ~40 pending MRs, bug fixes, improvements, desktop specific goodies. [09:30] didrocks, and Qt, then [09:30] hehe, yeah ;) [09:30] davmor2, probably means apport failed to collect the crash at all due to memory starving [09:30] bzoltan: right. but please remember that today is a very special day :) [09:31] asac: I know [09:31] so there might be that small delay. after we are done we are back to normal [09:31] ok [09:31] thanks [09:31] for understanding in case you dont get the same service today :_) [09:31] asac: I will survive :) over easter without a release [09:31] * asac is upgrading desktop to very latest trusty [09:31] davmor2, the qmlscene crash is SIGABRT, could you see in music-app's upstart log for a crash message [09:31] yeah thats what i feel [09:32] asac: you will get the unity7 security fix :) [09:32] :) [09:33] not that someone could get to my machine and hit enter on screensaver anyway :) [09:33] (without knocking me out first) [09:33] * asac sits constantly in front [09:33] ;) [09:36] Chipaca: How quickly does the Ubuntu-push log grow? [09:37] davmor2: when everything works, and it is in debug mode, about 80 lines an hour i'd say [09:37] davmor2: mine is 260K after 24 hours [09:38] davmor2: now, there are people whose whoopsie id seems to be "", and that might be messing things up [09:38] (the mind boggles) [09:39] didrocks: am I allowed to upload to the store, but not approve, so when I get the nod from you I can a pprove them and they will be in the next image? [09:39] davmor2: why? [09:39] Chipaca, popey: http://paste.ubuntu.com/7266817/ [09:40] davmor2: WANT! [09:40] popey: after we build the current image, please, yeah [09:40] popey: will keep you posted [09:40] davmor2: gimme gimme gimme gimme gimme gimme gimme gimme gimme [09:40] ogra_: !!! [09:40] ogra_: do you know what time is it? [09:40] ogra_: it's time to start building THE image :) [09:40] unity-mir | 0.3+14.04.20140417-0ubuntu1 | trusty/universe | source [09:40] \o/ [09:41] Chipaca: so that is a little over 24hours [09:41] does it mean its in? [09:41] asac: it does! [09:41] and we can kick image? [09:41] or I wouldn't say it [09:41] ogra_: !! [09:41] yeah [09:41] didrocks: thanks [09:41] It's IN [09:41] dont sleep [09:41] ogra_: where are you? :) [09:41] * sil2100 slaps om26er_ around a bit with a large trout [09:41] mmm trout [09:41] http://drool.popey.com/ [09:41] davmor2: i'm serious, get me those logs [09:41] * didrocks is going to use the web UI it seems [09:41] ogra_: dont be lazy again... we talked just yesterday about this :) [09:41] Good ol' mIRC slap emotes [09:42] didrocks: ack [09:42] Chipaca: no worries [09:42] 10:02 I think maybe 40mins [09:42] 10:40 unity-mir | 0.3+14.04.20140417-0ubuntu1 | trusty/universe | source [09:42] asac: ok, trying through the web, lets hope it's working this time [09:42] ogra seems to be asleep not responding for 20 seeconds :) [09:42] how's that for a guesstimate [09:42] *whip* [09:42] cjwatson: can't deny it :) [09:42] cjwatson: you won the bet [09:42] \o/ for THE image [09:42] cjwatson: well done! [09:43] *clicked* [09:43] (though i think you have ways to fake this :) [09:43] didrocks: both archs? [09:43] asac: I try to keep my manual interference to a minimum on release day :) [09:44] asac: yeah, double build for better goodness! [09:44] hehe [09:44] on it [09:44] - Ubuntu Touch armhf for Trusty (requested by didrocks) => ('ubuntu-touch', 'daily-preinstalled', 'preinstalled-touch', 'armhf') [09:44] ogra_: tooo late [09:44] - Ubuntu Touch i386 for Trusty (requested by didrocks) => ('ubuntu-touch', 'daily-preinstalled', 'preinstalled-touch', 'i386') [09:44] ogra_: too late [09:44] so yes [09:44] dont do it :) [09:44] cjwatson: and great, so working :) thanks for confirming! [09:44] cool all is taking its course. nice one [09:44] great [09:45] ogra_: I got the last one then! :) [09:45] congrats [09:45] :) [09:45] right now i feel nothing can stop us now \o/ [09:45] and building [09:45] ahah, not sure how much it is to celebrate to click a button TBH :p [09:45] lets not test more so we dont hear about bugs :) [09:45] asac: +1 [09:45] heh [09:45] ;) [09:46] I hope the security fix in unity7 isn't going to strike us through unity-service :p [09:46] Saviq, the long left to right swipe behavior got quite flashy ... [09:46] didrocks: please stop creating these scary scenarios [09:46] * didrocks shuts up :) [09:46] didrocks: didnt we dis-upgrade to test? [09:46] so it was in our testing? [09:47] asac: yeah, should be if everyone followed the procedure [09:47] i am sure some did a dist-upgrade [09:47] while others an install [09:47] so guess we are fine in almost all cases [09:48] ok, will visit mr. shower now [09:48] bbiab [09:49] * sil2100 actually upgrades his desktop as well [09:49] ooh, good idea ! [09:49] om26er_: so, flashed? [09:49] didrocks, flashed, apt-get update is happening which seems a bit slow [09:49] om26er_: ok, and then, just dist-upgrade [09:49] === trainguard: IMAGE 302 building (started: 20140417 09:50) === [09:49] you will get everything! [09:50] reboot… and enjoy ;) [09:51] * didrocks reflashed on 301 to confirm and gets after dist-upgrade: http://paste.ubuntu.com/7266860/ [09:51] so really only what we want [09:51] (with unity7 of course) [09:52] indeed [09:52] nothing else entered the archive [09:52] well, nothing on our image [09:52] Saviq: nothing in Music player about a crash let me try gallery which I think was the app that was opening [09:53] ogra_, "flashy"? [09:53] yeah, not sure how else to describe it [09:53] Chipaca: people.canonical.com/~davmor2/upc [09:54] Saviq, swiping away the app seems to "think" for a split second and then quickly catch up making the animation look slighly odd [09:54] ogra_, right, screenshotting sometimes takes more time than usual [09:54] it is as if the swipe actually arrives delayed a bit at the shell [09:55] ogra_, if you can find a way to get it in that state reliably, would be good to know [09:55] Saviq: m&c unity-mir for you [09:55] thats a permanent thing on 301 [09:55] didrocks, thanks [09:55] yw [09:55] ogra_, across reboots? [09:55] Saviq, doesnt influence functionality, looks and feels a bit strange though [09:55] yes [09:56] ogra_, ok so obviously I can't reproduce that reliably... [09:56] ogra_, can you describe your device's state? how many apps open, things like that? [09:57] thress webapps open ... G+, heised.de golem.de ... all the latest versions (new framework) swiping away one of them the systme seems to not recognize the swipe for a fraction of a second, then catches up and plays the animation in a slight fast forward mode [09:57] *three [09:57] ogra_: do you know what revision we will have on the stable channel? [09:58] ogra_: like version "1"? [09:58] hmm, nope, let me check [09:58] thanks [09:58] 11 could be [09:58] yeah [09:58] great ;) [09:58] it is at 10 currently [09:59] didrocks, so I have bug #1308891 and the associated MP, should I add two landing requests with that? [09:59] sounds good! [09:59] bug 1308891 in unity8-desktop-session trunk "[SRU] indicators not started" [Undecided,In progress] https://launchpad.net/bugs/1308891 [09:59] Saviq: one is enough [09:59] Saviq: it's a bug fix and trusty and U will be synchronized [10:00] so until you add a new feature or diverge, I will really advise you just use one branch [10:01] kk [10:01] didrocks, ok, icanhassilo for row 32 then? [10:02] Saviq: landing-001 for you! [10:02] yay, a prime silo again /me likes [10:02] ;) [10:03] Saviq, hmm, funny, now it works as usual [10:03] (now i would think something hogged the system, but it was really just that one animation that behaved like this) [10:03] ogra_, that's where it's most apparent [10:04] davmor2: do you have a few minutes to try something on that device? [10:08] Chipaca: not right now no. Give 10-15 minutes to see if I can break it and I'm all yours :) [10:12] davmor2: k [10:12] wow, playing back music in the browser is quite stuttery when the screen is off [10:12] Stop finding bugs! [10:12] ogra_: run top [10:12] ;) [10:13] davmor2, it works toally flawless if the screen is on (using my laut.fm app on my flo atm) [10:13] psivaa: any chance you can help me with https://bugs.launchpad.net/reminders-app/+bug/1308949 [10:13] Launchpad bug 1308949 in Ubuntu Reminders app "Running AP tests fails" [Undecided,New] [10:13] ogra_: oh that's a web app, that apparently isn't meant to work at all if the screen is blank [10:13] davmor2, right :) [10:13] popey: 1 sec pls [10:14] davmor2, nontheless the suid root running sandbox keeps it up :) [10:14] psivaa: sure. [10:14] didrocks, so. what to test now ? [10:14] i guess thats something to talk about with the security team [10:14] om26er: start/stop application, test everything, the hell out of it [10:14] there is a new oxide, i wish my job today was to just go around the internet on my phone and play videos and music. [10:14] om26er: reports any new issues [10:14] ;) [10:15] that part as well :p [10:15] * ogra_ is always impressed by the speakers in the flo [10:15] ogra_: the manta ones are mind blowing [10:15] (not that they sound so great, but their stereo effect is really nice) [10:16] speaking of speakers, the earphone in mako have an crazy high level, gives bad call quality [10:16] oh, manta ... [10:16] need to upgrade it i guess :) [10:16] popey: do you want me to confirm the bug? [10:17] eeeeeeeeewree4eee [10:17] qwdzuwerqtwrewtzewr5z6rt3zeu7woiperuwouipeo [10:17] heh [10:17] say hello to my cat [10:17] ;) [10:18] ogra_, my nexus 4 so old now, are we ever going to nexus 5 ? [10:18] i doubt it [10:19] or maybe we'll skip the 5 and wait for 6 [10:19] we'll all go bq or meizu :) [10:19] psivaa: well, I need to find someone who can help me fix it. Guess i need to wait for balloons.. [10:19] or that. [10:19] the bq case actually looked very much like the N5 [10:20] popey: is that about failing AP tests of click apps? [10:20] popey: isnt that what we fixed now by evdev landing? [10:20] so in the image currently being built [10:20] asac: i don't know whats causing that issue. [10:20] the log looks not like its a runtime thing though [10:20] bzr: ERROR: Not a branch: "bzr+ssh://bazaar.launchpad.net/+branch/reminders/". [10:21] huh [10:21] * popey checks his working [10:21] looks like it cant pull via bzr [10:21] reading the log [10:21] right [10:21] its reminders-app [10:21] might be a misplaces lp:reminders alieas [10:21] oh, you said so above :P [10:21] so inconsistent naming [10:21] Fuck I might have to drop everything and run. We just had a call from the hospice that the Mother-in-law is at she is serious ill, and they are putting a morphine driver into her, which we know tends to signify the end :( [10:21] bzr branch lp:reminders [10:21] bzr: ERROR: Not a branch: "bzr+ssh://bazaar.launchpad.net/+branch/reminders/". [10:22] davmor2: sorry to hear that dude. [10:22] davmor2: go for it! [10:22] dont worry [10:22] asac: yeah, thanks for looking [10:22] davmor2, oh man [10:23] they're going to let us know how she responds later today so I'm okay till then [10:25] davmor2: sorry for you dude, yeah, if you need to run out, no need to loose time warning us, just go… === MacSlow is now known as MacSlow|lunch [10:34] Mirv: could you please upload this to the store when you get a moment - it's not in the default image ☻ http://s-jenkins:8080/job/reminders-app-click/lastSuccessfulBuild/artifact/out/com.ubuntu.reminders_0.5_armhf.click [10:35] popey: ok! [10:35] thank you [10:35] davmor2, how finishedis the mult party calling ? [10:35] davmor2, I am not able to merge two calls, is that expected ? [10:36] davmor2: i'm stepping away for a bit; when you return, could you run http://pastebin.ubuntu.com/7266946/ on the device? ta [10:36] om26er: Toykeeper and others can, I can't so it is believed to be a limit imposed by your provider I am going to contact mine and see if this is the case [10:37] Chipaca: indeed I can [10:37] davmor2: by "when you return" i meant "when you are able to get back to this" :) [10:38] asac, didrocks, Saviq: So without opening more than 5 windows I am able to open and close apps at will with no crash [10:38] * Chipaca steps away for real now [10:38] * om26er calls the helpline [10:38] davmor2, webapps ? or native ? [10:38] ogra_: mix [10:38] cool [10:38] is image built? [10:38] sweet [10:38] asac: not yet [10:38] asac, the bot will tell you [10:38] (with a 5 min delay) [10:38] asac: image is building [10:38] popey: reminders done, https://myapps.developer.ubuntu.com/dev/click-apps/403/ [10:39] thank you Mirv [10:39] imgbot, stunt [10:39] * imgbot rolls on its back and purrs [10:39] :) [10:39] imgbot, play dead [10:40] it only knows that one stunt :( [10:40] ogra_: that's amazing not a dickiebird from it [10:40] haha [10:41] didrocks: odd since it finished ten minutes ago. maybe there's an additional delay for mirror syncing [10:41] ogra_: ^- I mean [10:42] cjwatson: even the image diff is ready? [10:42] oh that I have no idea about [10:42] davmor2, will be activated in 15mins thats what i've been told [10:42] cjwatson, it looks for the system-image to appear [10:42] cjwatson, thats usually another 20-25 min [10:42] om26er: from your provider? [10:43] davmor2, yes [10:43] om26er: so then it should work after that :) [10:43] davmor2, lets see [10:43] ogra_: oh system-image, right [10:43] duh :) [10:43] :) [10:44] oh, a GLES game in the store ! [10:44] is it out yet? [10:45] hmm, sorry wrong channel [10:45] haha [10:45] dbarth, every time you ask the release will be delayed by 5min [10:45] ouch [10:45] davmor2, well didn't take 15 mins, and it works :) [10:50] hmm, installing fresh webapps i only get a black screen when starting them [10:50] they work after reboot [10:50] * ogra_ just tried hexgl [10:51] is it fast enough? [10:51] dog slow on my laptop [10:51] hmm [10:51] it doesnt start ... i get a grey screen instead of a black one though [10:52] the app name is wrong [10:52] LaunchProcess: failed to execvp: [10:52] /usr/lib/arm-linux-gnueabihf/oxide-qt/oxide-renderer [10:53] ah [10:53] lovely [10:54] what i find finny is that when apps dont start properly the X to close them appears on the other side of the window [10:54] yes! [10:54] i noticed that [10:54] (if you tap-hold in the app lens) [10:54] Chipaca: whoopsie id is: >>aebe5728705da7fbb9c278f252713be7fe0e9f22cba2e301676f330c838cff79c55130a377896cb3930f9ab33a3a893b7808acba42b29ab1e6fe7ab99001314a<< [10:54] * popey files bug [10:55] om26er: Mirv: sil2100: still everything's fine for you guys? [10:55] not OMG alert? [10:56] om26er: apparently my provider thinks they don't support it are they in for a shock [10:56] didrocks, important features are working great, will move to not-so-important stuff in a big [10:56] *bit [10:56] didrocks: no OMG:s, still [10:56] davmor2, hah that sounds crazy, its so important of a feature [10:57] om26er: excelent, in case davmor2 is going to run soon, you know how to dogfood the image with davmor2's test cases? [10:57] didrocks, yes, I have done that in the past. just need link to that doc from davmor2 [10:58] davmor2: you would have it more handy than I do ^ [10:58] * didrocks can't wait for #302 to be published [10:58] https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0Ai33BkOcORLLdE4xLTFtSE80ZkpITXZ3aV85cWtPX2c#gid=0 [11:00] ooh, interesting [11:00] on the manta the music stops when the screen goes off [11:00] switching between scopes is pretty good, too bad the scrolling vertically is still subpar [11:00] ogra_, stops as in fully stops or becomes choppy ? [11:00] ogra_: faster to power off and blocking the queue? [11:01] om26er, fully stops ... on the flo and mako it gets choppy [11:01] Mirv: agreed! the partial fix from Saviq is already really nice :) [11:01] didrocks, perhaps [11:01] didrocks, the actual bug is that it plays on on the other devices, the browser shouldnt be able to afaik [11:02] and the white background in the dash hurts my eyes with brightness set to a higher value [11:02] (not that i mind being able to listen to webapp music with the screen off) [11:02] oh, seems the network on my manta simply died when it shut off the screen :P [11:03] ogra_: that can explain [11:03] yeah [11:03] luckily we dont block on manta bugs :) [11:03] yep [11:03] well, luckily [11:03] I would like we would be into that position [11:03] juts reastically, we are not [11:03] the browser is also very flickery on scrolling [11:04] * asac restarts system after dist-upgrade [11:04] davmor2: that was on the device where those logs came from? [11:09] om26er: Mirv: davmor2: please do a flash on mako, the image is there [11:09] just not the diff [11:09] so that we can start real dogfooding now :) [11:09] alright [11:09] === trainguard: IMAGE 302 DONE (finished: 20140417 11:10) === [11:09] === changelog: http://people.canonical.com/~ogra/touch-image-stats/302.changes === [11:09] om26er: davmor2: if dave is still there, can you share the dogfooding? (one starting in the end, the other at the beginning?) that will enable to get more dogfooding done :) [11:09] there you go [11:10] didrocks: I'm here unless we get a phone call [11:10] popey: let's wait for me to install that image (download almost over) and then, I'll +1 on click apps publishing [11:10] sil2100: please help as well testing ^ [11:13] thanks didrocks [11:14] ogra_: if you want to play with it on your tablet… [11:14] ogra_: as you have one and we don't :) [11:15] didrocks, sure, all three devices around me are on 302 already ... and the fourth device just generates a bootchart for it :) [11:15] ahah, great! [11:15] * didrocks did a clean reinstall [11:16] (still in recovery) [11:18] wow, did the browser get more snappy or is that me ? [11:18] ogra_: seeing the fix, it's you [11:18] but let's see it's all desired! [11:18] heh [11:18] say* [11:18] indeeed :) [11:19] heh [11:19] hum, where is psivaa? [11:19] plars: around? [11:20] ev: we might want someone look at CI test results (and maybe winning some time by triggering the cron in advance to start testing?) [11:22] popey: ok, install + upgrades work [11:22] yay [11:22] popey: so you can publish click apps I would say [11:22] thank you! [11:22] yw ;) [11:22] Mirv: when you get a moment could you please push http://s-jenkins:8080/job/weather-app-click/lastSuccessfulBuild/artifact/out/com.ubuntu.weather_1.1.236_all.click to the store ☻ [11:22] popey: if you have time to participate in the dogfooding as well… we'll be loved! [11:23] #302? [11:23] sure [11:23] yeah ;) [11:23] whatelse [11:23] best than ever! [11:23] 101? [11:23] :p [11:23] "stable" ☻ [11:23] 101 is dead [11:23] for long actually ;) [11:23] long live 10 [11:23] It's not dead, it's resting. [11:23] and will be dead forever starting today! [11:23] yesterday actually [11:23] when it was renamed to 10 [11:23] :) [11:24] poor of him :) [11:24] ogra_: so stephane confirms that people on 101 will upgrade to 11? ;) [11:24] yes [11:25] ogra_: bug 1308973 [11:25] bug 1308973 in unity8 (Ubuntu) "Icon for closing apps appears on wrong side " [Undecided,New] https://launchpad.net/bugs/1308973 [11:26] didrocks: sure [11:26] popey, confirmed [11:26] thanks ev [11:26] didrocks: he confirmed it yesterday [11:26] (stgraber) [11:26] great ;) [11:26] guess he will be on soonish [11:27] unfortunately we will not be able to redo the test [11:27] right [11:27] plars, psivaa: can one of you help didrocks? [11:27] chromium has now an indicator entry on trusty? interesting :P [11:28] not here... [11:28] oh now its gone :) /me unchecked let chromium run in background [11:28] not sure how to get it back... just happened after final upgrade and reboot === greyback is now known as greyback|lunch [11:29] popey: if you go in advanced settings -> System you can check "continue running background apps if chromium is closed" [11:29] that makes this icon appear as an indicator [11:30] ev, didrocks: it's still 6:30 am here and I'm distracted getting my kids ready for school, but I can try. What's up? [11:30] plars: well, image started to get installed now, so I guess we're good [11:31] (we could have win 15 minutes, but no worry) [11:31] asac: yeah, latest chromium has the entry [11:31] asac: it's an upstream change btw [11:31] (chrome unstable directly from google has it for quite some time) [11:33] didrocks: ok let me know if I need to take a look, otherwise I'll be in front of my computer more permenently in just a bit [11:33] thanks [11:35] didrocks: do we have a bug for the 13.10 display issue on upgrade? [11:35] plz add to release notes [11:36] asac: no, we can create it but the design changed and it's not displayed anymore [11:36] would prefer to move the content and workaround etc. to the bug and just keep a short line in there [11:36] asac: so the status will be "fix released" [11:36] what? [11:36] when? [11:36] which will be puzzling [11:36] as told yesterday… [11:36] well, we need a bug to reference for the issue [11:36] even if its wo't fix [11:36] :) [11:36] the system settings update design changed [11:37] we don't show that and the protocol changed [11:37] sure, but we dont have that yet. also it seems one of our services exports bogus data still [11:37] so we need to fix that (either dont export or fix it) [11:37] asac: it doesn't export anymore [11:37] in 14.04 [11:37] (as told) [11:37] ok, well. i dont like keeping long text in the errata section [11:37] asac: did we get QA to open bugs for issues they found? [11:38] argh [11:38] and having something there withtout a bug isnt good either [11:38] stgraber, we dont have any target dirs for flo and generic/_x86 in stable yet [11:38] seems it would be faster to open one and set to won't fix anyway [11:38] than discussing [11:38] right [11:38] stgraber, i suspect we'll need that for copying ? [11:38] asac: QA should open bugs though [11:38] iahmad: can you open a bug? [11:39] for the 13.10 displayed at system upgrade? [11:40] let me open it… [11:41] err [11:41] * ogra_ just notices that the clock on his phone shows the wrong time === MacSlow|lunch is now known as MacSlow [11:42] hrm [11:42] suspending and resuming got it working again ... it was about 10 min behind and no action could get it forward [11:43] asac: done and comment ammended [11:43] cool [11:43] didrocks: added the upgrade issues [11:44] seen that, good [11:44] asac, sure can do, I thought didrocks suggeted one already exist [11:44] iahmad: no, I suggested that it was fixed in 14.04 [11:44] iahmad: yeah [11:44] as the design changed [11:44] he jumped on it :) [11:44] thanks [11:44] now done [11:46] didrocks, would you mind sharing the bug#? [11:46] iahmad: bug #1308980. Please open them in the future while reporting :) [11:46] bug 1308980 in ubuntu-system-settings (Ubuntu) "System upgrade dialog displays 13.10 while you are installing 14.04" [Undecided,Won't fix] https://launchpad.net/bugs/1308980 [11:47] didrocks, ack [11:48] davmor2: Mirv: sil2100: popey: all good on #302 for now as well? [11:49] didrocks: so far yes [11:49] popey: weather app at https://myapps.developer.ubuntu.com/dev/click-apps/147/ [11:49] thank you Mirv [11:50] didrocks: yeah, feels good so far [11:53] didrocks, yep, its going good [11:53] how are we from the list? [11:54] 30% of dogfooding tests done? [11:54] om26er: ^ [11:54] all done [11:54] atleast from the manual test plan [11:55] i'll just go exploratory now [11:56] didrocks: I'm not doing minor dogfooder I'm doing a full run like I would for the whitelisting, I figured it made sense as it has the potential to be the final image so far so good :) [11:57] davmor2: you are syncing with om26er rights? [11:57] on the manual test plan [11:58] didrocks: om26er is cruising through it, and popey's name is down. Hence me going for a full suite rather than a standard dogfood. [11:59] davmor2: ok, sounds good, please just ensure the communication flow is consistant :) [11:59] davmor2, how do you mean ... "potential" ?? [11:59] didrocks: but I'm quite happy with this one [11:59] sweet! [11:59] it *is* the final image [11:59] the best we ever produced :) [12:00] didrocks: yeah so from me you'll get a nice breakdown of everything that you can link to. I'll throw you the page when I'm back from lunch :) [12:01] davmor2: perfect! === psivaa changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: - [12:07] do we end TRAINCON-0 today? [12:08] om26er, yeah, go wild, upload away [12:08] :P [12:09] ogra_, speaking of upload, i am waiting for media-hub and greeter changes to land, both make life simpler for me. they were backed out mostly due to this state [12:10] didrocks: 302 seems good here [12:10] popey: +1 for you? [12:10] om26er, yeah, they will be in our first U image [12:10] om26er: not as long as we don't have the Qt fix + scope ones [12:10] ya [12:11] I guess [12:11] When will we switch to the new frameworks? [12:11] but let's discuss that next week :p [12:11] ☻ === alan_g is now known as alan_g|lunch [12:11] popey, hopefully earlier than last cycle [12:11] popey: at the last minute, after the incompatible change is done :p [12:11] popey: thanks for the testing! [12:11] well, the music app will soon drop support for grilo.. so we wont be able to update that in 14.04 [12:11] np [12:12] but we need to carefully manage that, or have two branches ☹ [12:15] we unfortunately ended up releasing with 14.04-dev1 [12:15] but not the end of the world :) [12:15] might try to do something about that post-release [12:17] that would be easy to drop [12:18] let's do that in the incoming week [12:18] is there a bug for it so that we can track? [12:18] do we ship 14.04 framework ? [12:18] or just 14.04-dev1? [12:18] IIRC, we ship both [12:18] but colin should confirm [12:18] that doesnt sound too bad if dev1 == 14.04 [12:18] not even sure if we really need to kill dev1 in that case? [12:19] I remember colin did an upload after the 5.2 transition to get the agreed codename [12:19] I don't remember if it was something else to -dev1 [12:19] or -dev1 to the final name [12:19] * didrocks wasn't in the sdk name discussions [12:19] so what i assumed we would do is bump the dev names during dev cycle and that in the end keep the last dev1 version in (so apps dont need rebuild), but add the stable version on top [12:20] but then i didnt track the whole framework story to the end [12:20] asac, didrocks: preliminary testing of 302 is not super good here for webapps [12:20] ++ [12:20] same here [12:20] trying to track down the problem, though i don't think that's webapps per se [12:20] dbarth: how not super good? [12:20] Just got back from lunch [12:20] why didnt we see that yesterday? [12:20] like links taking 5s or so to be redirected to browser [12:20] oh, i missed the "not" [12:20] they are great here [12:21] or apps taking blank screens all of a sudden [12:21] havent had that here [12:21] i did a clean ubuntu-device re-install [12:21] davmor2: om26er: ^ can one of you you help extract/repro/confirming these observation? [12:21] will double check my setup [12:21] and re-confirm with alex [12:21] plz do [12:21] keep you posted [12:22] takes two seconds here for the browser to start if i tap an external link [12:22] yep, same [12:22] ogra_: and if the browser is already open? [12:22] asac: I think just dev1 right now by oversight, but as I say, easy to add post-release [12:22] ogra_: from which app? [12:22] dbarth, my own ... heise.de [12:22] cjwatson: ack. yeah no big deal i think [12:22] we can sneak in 14.04 if tits an alias [12:22] (its in the store) [12:22] ogra_: can you try from facebook or twitter? [12:23] asac: it would be, yes [12:23] and as I said earlier no reason to drop dev1 [12:23] dbarth, i have accounts for neither ... i can try G+ though [12:23] just reflashing to be double sure [12:23] ogra_: ok [12:23] if i find an exteral link there somewhere [12:23] cjwatson: right, so i think i understood the story correctly then :)... just wonder where we do that update right now, but lets check tuesday i guess [12:23] yeah, definitely not today :) [12:23] dbarth, with browser already open it switches instantly [12:24] let me kil the browser [12:24] sil2100: #302 rocks \o/ [12:24] Mirv: webapps too? :) [12:24] dbarth, again ... even with it closed [12:24] Mirv: it looks ok, but I just started testing it - aren't there problems with webapps [12:24] ? [12:24] sil2100, dbarth seems to see some [12:25] popey: the LAS apps is impressive [12:25] i cant repro at all here [12:25] ogra_: or that was the hexgl app i tried just before [12:25] didrocks: it is, isn't it? [12:25] but the phone became quite unresponsive [12:25] dbarth, yeah, thats broken [12:25] dbarth: how can I reproduce your problems? [12:25] popey: pure QML or webapps? [12:25] popey, ^^^ can we drop hexgl until it is fixed ? [12:25] qml [12:25] asac: well I don't see any particular problems myself [12:25] sil2100: just the webapp test plan, ie open facebook, follow links in wall posts and try to post a message [12:26] the link opening is where i found it quite slow [12:26] dbarth: what URL is the webapps testplan? [12:26] ogra_: not really our place to remove broken apps from the store, better to tell the dev so they fix it. I'll let them know [12:26] popey: I like clicking to "videos" and average users seeing they wait for Media HUB from SDK side though ;) [12:26] i only know webcontainer and webbrowser [12:26] heh yeah [12:26] sil2100: https://wiki.ubuntu.com/Process/Merges/TestPlan/webapp-container [12:27] ogra_: told them. [12:27] thx [12:27] dbarth: there is nothing specific in there... [12:27] Verify that links leading to external content are re-directed to the default browser application [12:27] is very generic [12:27] dbarth: ok, let me try that [12:28] please improve the testplan to be bullet proof and unquestionable [12:28] * ogra_ finds that pretty clear [12:28] ogra_: but how do you find those links? [12:28] where? [12:28] ogra_: agreed with asac, it needs to be "click there, there and there" [12:28] follow some lunatic on twitter [12:28] with stable test case [12:28] that could be described... e.g. open webapp gmail, click on "help link" [12:29] asac, well, you have to try some in a webpp [12:29] right, just giving concrete examples would help to avoid people interpeting this [12:29] wrong [12:29] how would you know where they point to [12:29] ogra_: we could give concrete examples, thats the whole point :) [12:29] we don have a status bar in the browser [12:29] nothing more [12:29] but we know our webaopps... like twitter/gmail etc. [12:29] and probably could say which links to click there to test [12:29] we could set up a mobile page somewhere and create a test app [12:30] its just odd to have such a thing in the store [12:30] you could surely automate this as well. but just saying that the test plan as of now might cause someone to test something wrong [12:30] hmmm [12:30] i would surely be lost for a bit [12:30] asac: patches welcome [12:30] asac, well, the testplan adapts to the nature of the apps ... [12:30] i wont change it [12:30] jus saying that if this slipped its probably because of that [12:30] nice [12:31] asac, if we can rip out the variables by providing an app that says "internal link" and "external link" it might be clearer [12:31] i dont think you get what i mean :); anyway [12:31] once webbrowser-app is opened, external links is snappy to me [12:31] but there is no way to distinguish internal vs external in any of our webapps [12:31] (at least in g+) [12:31] haven't heard anyone can confirm that its slow [12:31] you have to try [12:31] so i currently think its false alert :P [12:32] * ogra_ definitely cant confirm [12:32] dbarth: sil2100: I don't seem to have problems opening links from Twitter or GMail. but indeed links from Facebook posts don't work - but I don't use Facebook that much that I'd know if they worked before. [12:32] hope so [12:32] my phone flies [12:32] asac: and anyway slow > not working [12:32] yep [12:32] (even if it was) [12:32] still would be good to know [12:32] yeah, definitively don't reproduce here on a few webapps [12:32] Mirv, dbarth: so far it looks ordinary here [12:33] i think that was hexgl [12:33] dbarth, Hi! I am trying to reproduce the delay to open links. I might have seen that once today with camera -> gallery but thought system was busy/loaded [12:33] but that shouldn't have affected the other containers [12:34] clean reflash, start and test: so far so good with twitter, facebook, gmail, calendar, amazon, ebay [12:36] now loading hexgl to see if that was the cause (sorry to blame this app) [12:38] dbarth: ok, here webapps seem to work ok, no problems opening any links whatsoever [12:39] sil2100: yup, i can't reproduce anymore [12:39] something was loading the system for sure [12:40] cool (or not) [12:40] wonder what can slow down system so badly [12:41] eventqueue? crash dump? [12:41] 11 webapps loaded, including hexgl, and the phone still runs fine and switches between apps properly [12:41] dbarth: did you check if you had anything in /var/crash? [12:41] will do now [12:41] When you had that 'problem'? [12:41] asac, the renderer in the sandbox runs suid root ... i imagine for webgl it spans some extra stuff [12:41] Since maybe apport was busy collecting again [12:42] asac, definitely something to look at [12:42] ogra_: but its int he sandbox? [12:42] so is that a problem that its root? [12:42] asac, i have no clue ... but i tried hexgl too earlier today and it seems to put some load onto the system [12:43] nothing to inspect now, but something we need to look at [12:44] chrisccoulson: ^ "? [12:44] sil2100: must have been apport [12:44] _usr_lib_arm-linux-gnueabihf_qt5_libexec_QtWebProcess.32011.crash 15min. ago [12:44] theoretically everything should run in the sandbox and should be killed with the app [12:44] ogra_: ok, so hexgl is causing system business? [12:44] asac, its a broken app ... it seems to cause a crash (see dbarth above) [12:44] dbarth: use apport-cli to preprocess and then retrace i guess :) [12:44] dbarth: was that around that time? [12:45] yup [12:45] asac, cant blame us for broken apps ... but we should not crash indeed [12:46] ok [12:47] right we shouldnt crash for broken webapps [12:47] nor should unity ccrash for crashing native apps :) [12:47] yep [12:47] nor should unity crash for broken scopes :) [12:47] details :) [12:47] didrocks, asac: This will be the doc to watch ;) https://docs.google.com/a/canonical.com/spreadsheets/d/1mQ3Nu-3H9IQz8R7I3FfMV6rwW36B3EcXRliq5hwFhzI/edit#gid=0 [12:47] i still believe we have an architectural issue :P [12:47] we do [12:47] davmor2: thanks! [12:47] plenty [12:47] but we'll fix them the less we have to look at feature development [12:48] i dont know [12:48] its all about how you start [12:49] right, and we started focused on getting the basic features in [12:49] still :) [12:49] however, thats reality [12:49] once that drags less of our attention away we can start polishing and optimizing more [12:49] its not polishing... its rerchitecturing :) [12:49] thats a big different [12:49] polishing is fixing bugs and bad code [12:50] thats optimizing :) [12:50] and glitches [12:50] the fact that a container crashes if the client in there is bad, is something that normal polish won't solve in my experience :P... needs someone to go and redo the arch [12:50] we have a good working basic setup now ... we simply need to cnstantly review it [12:50] but i am always pessimistic [12:50] without knowing whats really going on, so ... :) [12:50] * asac shuts up [12:51] morning guys [12:52] Phew [12:52] asac, yeah, better rave a little in #ubuntu-release-party :) [12:53] sil2100: checking the unity8 failing test? [12:54] didrocks: I was just about to run AP:s for it after refreshing the tests page [12:54] cjwatson: should we rather sync the whole release with ubuntu Touch image promotion to stable channel? [12:55] asac: wdyt? as we put it in the release note? ^ [12:55] ogra_: good idea [12:55] didrocks: how long will it be until you're ready? [12:55] :) [12:55] didrocks: not sure what you suggest [12:55] cjwatson: I would say 2h to get the automated testing done [12:55] didrocks: running the test suite, but it seems to be a flaky test - as the location indicator seems to be present all the time [12:55] imo we can copy it over once stgraber is avail [12:55] asac: AP testing is still happening, you don't care about their results? [12:55] we're still waiting for (at least) keystone to finish building and to migrate [12:56] cjwatson, we need stgraber ... there are subarch subdirs missing in the stable channel on system-image.u.c [12:56] didrocks: will we release the previous image if they fail now? :) [12:56] didrocks: did any AP succeed yet? [12:56] I expect infinity has some more things to do [12:56] asac: they do [12:56] asac: apart from unity8 which has one failure [12:56] and where I'm asking sil2100 to check [12:56] well, i think we would be ready whenever infinitiy is ready, given that stgraber is here [12:56] and can help us do the copy [12:56] didrocks, Mirv: just a quick note - when archiving results in the spreadsheet beware as sometimes some 'race' seems to happen and some assigned silos loose their data [12:57] I just re-inserted those that got missing now, not sure why this just happened [12:57] didrocks: right. so what i think is happening is that noone runs AP test on silos [12:57] sil2100: oof :( [12:57] didrocks: so :)... lets cross fingers its flaky [12:57] sil2100: weird, ok [12:57] asac: I do [12:57] let's see what sil2100 will come up with [12:57] right [12:58] * asac goes to supermarket getting some junk food [12:58] 10 min [13:01] ahhhhhhhhh, the "10 things to do after installing…" post [13:01] it's about time! :) [13:02] lol [13:03] hm, I just saw the Bluetooth indicator test happening on my phone and all seemed ok [13:03] But we'll know for sure after the whole suite is done [13:03] ok [13:04] Worrying is that it happened both on 301 and 302, 300 didn't have that failure - and in 301 a new unity8 was introduced [13:04] yeah [13:05] didrocks, FYI, silo 001 is ACK === alan_g|lunch is now known as alan_g [13:07] sil2100: Mirv: dealing with it? ^ [13:07] and see if the RT has time for a 0day SRU [13:07] (dekstop only) [13:07] desktop* [13:08] here's the sru bug #1308891 [13:08] bug 1308891 in unity8-desktop-session trunk "[SRU] indicators not started" [Undecided,In progress] https://launchpad.net/bugs/1308891 [13:08] ACK [13:08] Saviq: let me lookie [13:08] I'll let sil2100 handle it [13:10] didrocks: should I first ask the RT or I can publish anyway? [13:11] sil2100, can i get silo for 21? [13:11] mhr3: in a moment [13:12] sil2100: you can publish [13:12] it will be in unappproved [13:15] mhr3: sil2100 just found out it's already in landing-011 [13:15] sil2100: same for dialer-app? [13:15] ok, going for a run [13:16] I let you guys looking at the CI dashboard [13:16] and keeping a close eyes/retrying [13:16] Mirv, that silo has been there for over a week, can i get an override pls? [13:16] ACK [13:16] unity8 still running ;/ [13:16] It's taking ages [13:17] mhr3: can you coordinate with other landers if you are going to ditch their work and publish first? :) [13:18] mandel, ^^ mind? [13:18] mandel, building a bugfix for click scope [13:18] * didrocks away now [13:19] sil2100: I got the "frozen screen" with unity8 again, was there something to do when it happened? ie. stopping reacting to touch events when trying to run AP tests, ps ax shows it's trying to stop unity8 at the moment (/sbin/initctl stop unity) [13:19] ogra_: hello [13:19] stgraber, hey, happy release day ! [13:20] Mirv: it's hard to say, usually this resolves by its own after a longer while - I remember this: [13:20] sil2100: I killed the unity8 process so the stopper was then happy :) [13:20] When it was happening, I basically made sure unity8 was dead and then just switch off and on the screen [13:20] stgraber, we dont have subarch subfolders for generic/_x86 and flo in stable yet ... will copy-imagge create them or do i need you to pre-create ? [13:20] Yeah, so I guess that's the magic trick ;) [13:20] sil2100: ok then.. [13:21] I guess it's not a 'real' problem as it's probably AP having trouble killing unity8 after the test or something === jhodapp|afk is now known as jhodapp [13:23] yes, I haven't ever seen it real usage, only when running AP:s [13:23] ogra_: no, that was on my pastebin of yesterday. When we're ready to push trusty to stable we need to manually add the new devices and remove grouper and maguro, then do the copy [13:23] stgraber, right [13:23] stgraber: can we already do the cop and set permissions to 000 or something until we send out release announce later? [13:23] just wanted to make sure again [13:23] copy [13:23] or can we just do it before? [13:24] * asac thinks we could just do it after didrocks is happy with AP investigation [13:24] i dont think we have a mechanism to hold bacck [13:24] no phased roll out :)? [13:24] devel is our phasing channel here :) [13:24] we actually have phased rolled out support, we just never used it and I don't think that today is the day we want to try it :) [13:25] hehe [13:25] why not :P [13:25] yeah, what could possibly go wrong on release day [13:25] :) [13:26] all UK could still get a power outage i am sure === greyback|lunch is now known as greyback [13:36] didrocks, Mirv: all unity8 tests pass, whole test suite -> OK [13:37] didrocks, Mirv: to make sure I also ran the test_indicator_page_title_matches_widget test 'as is' and all is good [13:39] didrocks, Mirv: running dialer_app tests [13:39] popey: can you set an alarm I get No Active Alarms after setting one? [13:40] I'm a bit worried about dialer-app having 4 failures now on smoketesting [13:40] sil2100: works for me [13:40] davmor2: oh yes, i get that too, but the alarm shows at the bottom [13:40] popey: yeap [13:40] didrocks: hey, heads up - we just lost every single mako on ashes all at once [13:40] popey: and there is no activate button [13:40] didrocks: I don't have any more information for you yet, as I'm only about 4 seconds into looking at it [13:41] rfowler: any chance you, or someone, is in 1ss right now? [13:41] davmor2: sounds like a bug in clock. [13:41] nik90: ^ [13:41] didrocks, Mirv: dialer_app tests all pass here... [13:41] popey: also doesn't show the alarm in indicator [13:41] davmor2, yeah the indicator shows the upcoming alarm, just the clock text is wrong [13:41] I wonder what's going on with the devices in the CI lab [13:41] davmor2, I have it in indicator [13:42] Let me re-scan what it's doing when running those tests [13:42] davmor2, mine is set for 20 mins from now [13:42] sil2100: so you say that unity8 is flaky? [13:42] pmcgowan: I have Clock I have Calendar events but I don't see the alarm I set for 3:00 pm [13:43] asac: there seems to be a flaky test there, yes - at least from the looks of it here, as everything passes fine on my local device [13:43] plars: not now [13:43] asac: and also the test itself only checks if the Location indicator is displayed and working correctly [13:43] asac: which is true [13:43] plars: going to lex [13:43] rfowler: how soon could we get someone there to take a look? [13:44] sil2100: didrocks: yeah passes here too unity8 + dialer [13:44] sil2100: right [13:44] cjwatson, can you tell me who to talk to if we want to add an item to list of known issues? [13:44] rfowler: I can't see any mako in fastboot, nor in adb [13:44] rfowler: any chance all the makos are connected to the same usb hub or something? [13:44] The dialer-app failures worry me though - not from a 'regression' perspective, just that the test seem to be having so many problems but only on CI smoketesting [13:45] sil2100: Mirv: didrocks: do we believe unity8 is flaky or do we want to challenge whether its local vs. infra difference by retrying? [13:45] given that this happened for two images feels something isnt as sstable as before in infra [13:45] but guess we can keep observing [13:46] asac: I would try a revert, but once this run finishes completely and once we know we won't build anythingi n the nearest 30 min [13:46] plars: the makos are connected to one hub yes [13:46] rfowler: this doesn't look like an image problem [13:46] Since unity8 testing can take up to 30 minutes nowadays [13:46] rfowler: ok, that make sense then [13:46] asac: I mean, not revert! [13:46] sil2100: revert? you were not able to see the problem locally right? [13:46] rfowler: I think something happened to that hub, or its connection [13:46] asac: I meant rerun! [13:46] ! [13:46] yeah [13:46] :) [13:46] ok [13:46] TYPO [13:46] ;p [13:46] that i agree more :P [13:46] rfowler: most of the devices weren't even being tested on, but a bit ago they all went dark at once [13:47] lets have didrocks decide... he has the magic feel for this :P [13:48] * ogra_ sees sil2100 say revert above and uploads [13:48] plars: interesting === Ursinha is now known as Ursinha-afk [13:49] ogra_: nooooo~ o/ [13:49] ;) [13:49] :) [13:49] plars: power cycling the hub has no effect apparently [13:50] sil2100: asac: my feeling would be "infra" since none of us has seen that locally [13:50] mhr3: generally you can just add it to the release notes wiki page? [13:51] cjwatson, oh? and that's it? doesn't go anywhere else? [13:51] mhr3: afaik [13:52] cjwatson, ok, cheers [13:53] Mirv: yeah... [13:57] davmor2, popey: crap that is a regression... [13:57] davmor2, popey: The alarm switch was removed intentionally since it does not work due to a loop bug in the SDK. But the "No Active Alarms" issue is definitely a bug [13:58] nik90: got time to prepare a fix, or do you want me to revert the clock app in the store, or both? [13:59] (this wont revert the one shipped on the image unfortunately of course) [13:59] popey: let me try to investigate the issue and propose a fix asap [13:59] popey: if the alarm isn't meant to work I would leave it it doesn't hurt anyone it's just confusing [13:59] thanks nik90 [13:59] it is confusing, yes. [14:00] not an image blocker though.. i hope [14:00] davmor2: didrocks:I think we should release note it if the alert is broken [14:00] not at all === Ursinha-afk is now known as Ursinha [14:00] asac: alrams is broken due to the Qt bug [14:00] mhr3: so, did you have a talk with mandel and sergio regarding the click scope? [14:01] asac: they would only go off if your phone was awake [14:01] davmor2: do you have a summary of the most high impact symptoms of that event bug? [14:01] we might want to expand in bug/release notes [14:01] "nothing works while screen is off" [14:01] i think that summarizes it nicely :P [14:01] davmor2: yeah, this makes them useless :P ... i never keep my phone on while i am sleeping [14:01] you just have to tap the screen every 30sec [14:02] asac: You mean you don't sleep test like the rest of us shame on you [14:02] i do, but in cases where i dont i need the alarm, otherwise i also track the time :P [14:03] sil2100, mandel didn't respond [14:03] haha [14:04] i am now in the party [14:04] didrocks: are we good to start copying styuff to stable channel? [14:06] asac: let's wait for things to finish testing on CI smoketesting [14:06] messaging_app seems to still be running [14:06] sil2100: thats the last? [14:06] Ah, maybe I'll simply run it now [14:07] whats the name of the rotate sensor it has gone out of my head [14:07] asac: no... [14:07] asac: stilla lot to go ;/ [14:07] yeah === plars changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: all makos down in CI right now [14:07] wait [14:07] popey: all makos are down? [14:07] err [14:07] plars: ^^ [14:07] HUH?! [14:08] asac: yes, I mentioned that just a bit ago [14:08] yes [14:08] was discussed above [14:08] asac: I should clarify - all makos for smoke testing [14:08] Ok, this would explain why messaging_app is taking so long [14:08] asac: looks like a failed usb hub, we have someone in IS on the way to take a look [14:08] plars: well, thats all makos for whatever we care about [14:08] didrocks: so what do we do? [14:08] didrocks: test locally? [14:08] asac, didrocks: let me run all the remaining AP tests locally then [14:08] asac, trust our dogfooding [14:08] asac: we are on it, should hopefully have a resolution soon [14:09] plars: ETA < 10 minutes? [14:09] * sil2100 runs the tests anyay [14:09] yeah do that sil2100 [14:09] i dont believe in this getting fixed before i want to take a nap [14:09] ogra_: Pffffff [14:09] :) [14:10] asac: not sure, rfowler do you know about how far out you are from 1ss? (he may already be driving) [14:10] if he is driving then surely it isnt soon [14:10] asac, i think we are confident enough about the image atm ... we wont build a new one today so plars has enough time to run the tests later [14:10] sure [14:10] but [14:11] i want didrocks to come back [14:11] meantime sil2100 is runnign the rest [14:11] (i mean ... would you block it now if one more test fails) [14:11] sure [14:11] I guess these AP tests are just a 'triple-check' [14:11] right [14:11] ogra_: depends on what test fails [14:11] asac, ogra_: if needs be, I can always go back and rerun them too, even if we get a new one [14:11] if something fails that might indicate somethign super broken taht we missed, then yes :) [14:11] Since if something would be bad davmor2, me, Mirv and all the others would notice I guess [14:11] but very very unlikely at this point [14:11] * rsalveti waves [14:11] o/ [14:12] plars, right [14:12] rsalveti: just in time to the party:P [14:12] rsalveti, happy release day !!!! [14:12] popey: it is not a clock app bug. [14:13] popey: I just opened an older version of the clock app and it has the issue [14:13] popey: sometime else must have changed for sure..did a new release of the SDK land? [14:13] nik90: eds issue then maybe? [14:13] davmor2: either EDS or SDK Alarms [14:14] ok, lets get a bug filed with some data to back that up and we can see where it broke [14:14] ogra_: whats the name of the rotate sensor please [14:14] err, dunno ... frank ? [14:14] ;p [14:15] davmor2, are you looking for a package for a bug ? [14:15] platform-api i think [14:15] ogra_: no to add to the chart of working things [14:16] gyroscope or accelerometer ... i can never remember which is the right one [14:16] second one ta [14:17] didrocks: asac: if it's going to be a bit before rfowler gets there, I can repurpose a device connected to another adb host to see about kickstarting things too, but I suspect he's already on the way [14:17] didrocks: asac: I'm hopeful that it's just a bad connection and can be restored very easily, but if not, we'll get it running one way or another [14:18] plars: no hurry I guess, I'm running the test locally now [14:18] *tests [14:18] plars: if its an option to repurpose do it [14:18] we are waiting [14:18] sil2100: I have a complete set that I kicked off locally as well for other reasons, so if you see anything odd let me know and I'll check my results too [14:18] you should have done that right away without hesitation [14:19] asac: well, it will take a bit of time, but yes I'm already looking at that [14:19] asac: ogra_: are we good enough to celebrate already? :-) it seems we got some issues still [14:19] ok well, then dont bother, lets finish tests locally [14:19] not sure if we have time to fix them though [14:19] rsalveti: we dont have issues that i kinow [14:19] just infra burning :) [14:19] so we wont get final results in time, but tahts ok i guess [14:19] haha, cool then :-) [14:20] rsalveti, we're "good enough" [14:20] ogra_: fair enough [14:20] get a beer [14:20] :) [14:20] awesome [14:20] (and some breakfast too) [14:20] well, beer is kind of a liquid bread anyway [14:21] yeah, drink a saint bier :P [14:21] i will not do it just now [14:21] yeah, got a few here :-) [14:21] haha [14:22] didrocks, ogra_, popey, asac, sil2100: can you guys see anything obvious missing from the list https://docs.google.com/a/canonical.com/spreadsheets/d/1mQ3Nu-3H9IQz8R7I3FfMV6rwW36B3EcXRliq5hwFhzI/edit#gid=0 [14:23] davmor2: mention welcome screen stalls on unlock due to eventloop issue? [14:23] davmor2: let me look [14:23] davmor2, looks ok to me [14:24] davmor2: also, location has no agps so takes a long time to get a fix (outdoors) [14:25] davmor2: mention 'not being able to change the dash background' ;) j/k [14:25] sil2100: I already did [14:25] davmor2: oh! :) [14:25] sil2100: setting backdrops [14:26] davmor2: see it, awesome o/ [14:26] davmor2: it seems to look ok [14:27] sil2100: and note it's green cause it is a feature :D [14:27] :| [14:27] To me it's a regression, as in the previous images I could change that and in the new scopes I can't! [14:27] Give me my backgrounds [14:27] ! [14:27] ;) [14:28] sil2100: until it is decided otherwise it's a feature damn it stop trying to create bugs ;) [14:29] dammit unity crash [14:32] popey: ?! [14:32] popey: did I not just finiksh telling sil2100 to not create bugs? [14:32] -k [14:32] popey: awww come on, what happened? [14:32] i had a bunch of webapps open [14:32] left it alone for maybe an hour [14:32] tapped power, swiped in from right, *bang* [14:33] no crash file [14:33] might not have been unity, but everything went black then i saw the welcome screen [14:33] lemme try and reproduce [14:34] plars: makos are back... it looks like the hub got hosed [14:34] rfowler: that's what we thought [14:35] huh [14:35] asac: well, I'll bring the other devices back online shortly too, but I did kick off a job a bit ago on a device connected to a different adb host. It's still installing but when it finishes we should see results start to roll in on the missing ones via that device [14:35] didrocks, sil2100 ^ [14:35] cool [14:35] plars: awesome, is that for #302 ? [14:35] sil2100: yes [14:35] popey: the alarm bug is it in the stable 14.04 image or just the latest daily image? [14:36] sil2100: I just rescheduled the missing ones, plus messaging and dialer since there seemed to be a bit more errors than we normally see with dialer [14:36] nik90: latest daily is the final image [14:36] plars, in any case that will teach asac to not make jokes about power outages on release day (which he just did before you joined) :) [14:36] ogra_: oh, so it's his fault! [14:36] yeah [14:36] plars: yes, we know about these ones, I even did normal local runs [14:36] davmor2: :/ [14:37] nik90: its in the latest stable [14:38] i dont understand how this passed autopilot testing nik90, is there a test for setting alarms? [14:38] didrocks: Testing complete as far as I can tell. 1 blocker as expected, Some minor that have been reported, one new issue with the alarm not working. Other than that it's good to go and you can quote popey on that /me wonders off nonchalantly whistling as he goes [14:39] not happy about the alarms being broken. [14:39] sil2100, can i just get the silo? i promise to not land it before i talk to mandel [14:39] at least it can be tested [14:39] popey: there is but the branch that removed the alarm status switch also modified the AP tests to not check the alarm status. So it only checks the alarm name, time, days of the week. [14:39] mhr3: I guess, let me assign and note the conflict [14:39] nik90: i see [14:39] popey: no indeed [14:40] sil2100, ty [14:40] nik90: have we determined where the bug is? [14:41] popey: I am currently discussing with charles and renato to figure out if the bug is in EDS or indicator-datetime [14:41] popey: zsombi confirmed that it is not a SDK bug since the alarms API hasnt changed since feb 2014 [14:42] Its frustrating, but given alarms can't trigger anyway, while it's annoying to have it broken in this image, I doubt it will be considered a release blocker. [14:43] thanks davmor2! [14:43] * didrocks refreshes the dashboard === alan_g is now known as alan_g|tea [14:43] not over yet [14:44] popey: + [14:44] didrocks: see plars's messages ^ [14:44] asac: sil2100: what's up with the tests? [14:44] sil2100: local testing? [14:44] didrocks: mako's are down, there are some countermeasures taken [14:44] I'm running locally anyway [14:44] Yes [14:45] sil2100, didrocks: actually countermeasures are in place, but the makos are back online again too [14:45] sil2100: where are you with the remaining tests? [14:45] tests are running on the remaining tests in the lab now too [14:45] didrocks: from what plars said, the mako's are back but the tests are still not finished [14:45] didrocks: UITK tests running now [14:45] It's taking a while though... [14:45] didrocks: yeah, murphy just said hello one more time :) [14:45] Since almost half of the tests didn't run yet [14:45] asac: unity8 passed for Mirv, sil2100 and I AFAIK on the previous image [14:46] didrocks: i think devices are back up now, but i asked sil to just finish locally to void further delay [14:46] argh too many pings to backlog [14:46] asac, dont balme dholbachs dog ! [14:46] *blame [14:46] what's up? will be easier :p [14:46] asac: yeah, sounds reasonable :p [14:46] didrocks: yeah, so, unity8 is cool, dogfooding is +1 (as per davmor2), local AP tests are running, makos are back but tests are just rescheduled [14:46] sil2100: so, compared to the dashboard, what did you complete? [14:46] didrocks: messaging_app, address_book_app and ubuntuuitoolkit done [14:47] SO still a lot to go [14:47] sil2100: dialer_app was +1 too? [14:47] Yes [14:47] cyphermox: around? [14:47] sil2100: ok, which one are you running? [14:47] so that we don't do the sames [14:48] Now it's ubuntu_weather_app [14:48] sil2100: ok, I'm starting from the top then [14:49] doing music_app, then calendar_app [14:49] ACK [14:49] Mirv: eow for you I guess? [14:49] or are you around to help? [14:50] Knowing Timo he still might be around ;p [14:50] let's focus all efforts on this [14:50] didrocks, asac, no meeting today i guess ? [14:50] or did you plan one [14:50] Would be nice to have one I guess? [14:51] yeah [14:51] let's have a quick one [14:51] i might crash before [14:51] but who knows [14:51] yeah, would be my first one on release day in 10 years ... but ok [14:51] let me take a shower while the first tests are running :p [14:51] didrocks: what's up? [14:51] is this the standard landing meeting or something else? [14:51] cyphermox: CI devices are screwed [14:51] cyphermox: we need to all run AP tests manually, and we try to parallelize [14:52] yeah i saw the email === alan_g|tea is now known as alan_g [14:52] cyphermox: can you take some of the share as well? [14:52] didrocks: I *guess* they're back now, but I guess it'll take longer for them to finish than we will finish [14:52] sure, what do you want me to pick, and which image? [14:52] cyphermox: so #302 [14:52] sil2100: yeah [14:53] cyphermox: terminal & short app? [14:53] ok [14:53] let's sync here [14:53] didrocks: I'll do filemanager then [14:53] so that we can get to next ones [14:53] sil2100: great! [14:59] So far so good === doanac changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: doanac | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: all makos down in CI right now [15:01] didrocks: not very around, but I can run some stuff in case of emergencies [15:01] Hm. Seeing a bit of a memory leak in clock.. [15:02] popey, clock seems like it really regressed [15:02] for alarms anyway [15:02] Mirv: I guess we're almost done [15:02] thats not clocks fault [15:02] pmcgowan: yeah, being discussed [15:02] I'm not sure if this is a memory leak caused by clock itself or maybe the eventloop thing [15:03] where do those events queue up? [15:03] sil2100: are you sure we are almost done? we have quite some tests to run still [15:03] * didrocks still running music-app AP tests [15:03] (horrible digital voice btw) [15:03] +1 [15:03] plug headphones in or stick a screwdriver in your ear, both preferable [15:03] Mirv: do you have a device ready for running one or two AP tests?? [15:04] popey: ahah :) [15:04] didrocks: still music? Oh god... [15:04] Private + Shared = RAM used Program [15:04] Ok, and I thought only filemanager and weather have many tests [15:04] bah [15:05] paste.ubuntu.com/7268648/ [15:05] that ram usage is climbing alarmingly (pun intended) fast [15:05] 5 mins between each line [15:05] popey: I want to murder someone now :p [15:05] * popey cuddles didrocks [15:05] if I heard that music once more… [15:05] ahhhhhhhh [15:06] ;) [15:06] 4MB/min [15:06] * sil2100 is glad he doesn't run the music-app tests then [15:06] ;) [15:06] popey, logs ? [15:06] ogra_: of? [15:06] popey, clock ? [15:07] dunno [15:07] nothing worth seeing [15:07] k [15:07] where do events queue up? [15:07] inside qmlscene? [15:07] wonder if it goes down if I unlock the phone and let the events play out [15:07] somewhere in Qt [15:07] try it [15:07] i bet it does [15:08] unless it fragments the memory :) [15:08] didrocks, cyphermox, Mirv: running ubuntu_clock_app o/ [15:09] sil2100: still suffering music… [15:09] 134.6 MiB + 6.4 MiB = 140.9 MiB qmlscene [15:09] still climbing [15:09] didrocks: you'll see we already have results coming in on the dashboard for many of the restarted tests: http://ci.ubuntu.com/smokeng/trusty/touch/mako/302:20140417:20140411.3/7782/ [15:10] plars: can you start by the one we didn't run? [15:10] yet [15:10] didrocks: yeah, device is ready and I've been running some AP:s already (without fails) [15:10] didrocks: it's running those now [15:10] so, I'm running dropping letter [15:10] plars: hum [15:10] plars: we did run some manually [15:11] plars: to not delay the image, do you mind running some? [15:11] the one we didn't run manually yet [15:11] calendar_app now [15:11] didrocks: I'm running some locally too, but I can't change the order part way though without stopping what I have in progress [15:11] didrocks: I can see what I have so far, what are you looking for? [15:11] plars: argh, ok, so let's not count on the dashboard [15:12] didrocks: why? [15:12] Mirv: mind running online-account-ui [15:12] didrocks: I mean that I have what's running on the dashboard, but I also have local tests in progress [15:12] plars: because we don't have the time to wait for all of them to sync [15:12] plars: so we started manually [15:12] continued* [15:12] didrocks: online_accounts_ui already has results on the dashboard! [15:12] and it would have been nice that you sync up with us to only run the ones we didn't finish yet [15:12] ah great on that one :) [15:12] Mirv: sudoku_app then? [15:13] didrocks: and ubuntu_system_settings just finished [15:13] yeah, we already got results on that one [15:13] didrocks: running, and queued some others too [15:14] Mirv: great, let's continue to sync up :) [15:14] Mirv: so only sudoku_app for now for you [15:14] I guess, one cyphermox and sil2100 finishes, we'll have then only ubuntu_calculator_app remaining [15:15] 70% flashing... [15:15] cyphermox: ok, unsure you will finish before we are done then :) [15:15] let us know when you start [15:15] aye [15:16] well, u-d-f says 8 minutes to the download finishing [15:16] ah, then you flash [15:16] so, ok, I think we'll be done before you [15:17] didrocks: the toolkit tests are running now in ci [15:18] plars: yeah, we already validated that one :/ [15:18] didrocks: I can stop the ones I have at home and run some others too if you like [15:18] bug 1309065 [15:18] bug 1309065 in Ubuntu Clock App "Memory leak in clock 395 on #302 mako" [Undecided,New] https://launchpad.net/bugs/1309065 [15:18] plars: I mean it's good for the long term, but not for the current issues to just validate the image :) [15:18] didrocks: right, so I don't know which ones you still need. just let me know [15:19] plars: terminal_app, shorts_app [15:19] to start [15:19] hmmmm [15:19] hum, calendar_app fails [15:19] clock_app failures here === gatox is now known as gatox_lunch [15:19] * didrocks runs calculator [15:19] FAILED (failures=5) [15:19] will redo calendar after that [15:20] sil2100: did you retry? [15:20] Retrying now [15:20] Mirv: sudoku done, [15:20] ? [15:20] didrocks: starting those at home now [15:21] clock app AP test issues might explain our clock alarm rergression? [15:21] didrocks: does your device have google accounts / calendar stuff? [15:21] asac, ram usage piling up when screen is off ? i doubt that [15:22] popey: when do you know the memory leak? While in the alarms tab? Or just about anywhere? [15:23] s/know/notice [15:23] popey: only g+ [15:24] I just got a bug report from thomas strehl that clock app crashing while trying to edit alarms [15:24] didrocks: ok [15:24] https://bugs.launchpad.net/ubuntu-clock-app/+bug/1309057 [15:24] Launchpad bug 1309057 in Ubuntu Clock App "Clock app crashes when trying to access a saved alarm" [Undecided,New] [15:24] nik90: i left it on clock screen [15:24] calculator pass [15:24] relaunching calendar [15:24] oh, I'm in French [15:24] probably that [15:24] nik90: lemme clarify on bug [15:25] Holy shit [15:26] sil2100, so we finished testing of 004, ready to land it, but still no response from mandel, what to do? [15:26] didrocks: sudoku pass [15:26] didrocks: terminal_app is good [15:26] mhr3: we're busy with the release now anyway, not sure if we intend to build a new image today - let's wait a bit [15:26] didrocks: I got again failures... this time 10! [15:26] o_O [15:26] sil2100, ok, thx [15:28] popey: can you revert clock-app to rev 394. The memory leak and the clock app crasher wasn't present until rev 393 for sure. Otherwise we would have noticed it immediately. [15:28] almost booting... [15:29] Mirv: can you check clock app? ^ [15:29] plars: ok, weather app now? :) [15:29] nik90: uh, i can revert to the previously publised version [15:30] popey: when was the last publish date? [15:30] didrocks: I'll do that one as soon as shorts_app finishes [15:30] plars: oh sorry, shorts is good [15:30] weather is done [15:30] 373 was the previously published version nik90 [15:30] nik90: might be easier to revert 395 in bzr, and push a new 396 and upload that [15:30] camera_app wasn't done, right? [15:30] Crap, now 13 failures [15:30] s/easier/cleaner [15:31] didrocks: I have no idea what's up with the clock app ;/ [15:31] sil2100: camera_app? ^ [15:31] popey: ok I will do that. [15:31] didrocks: I can proceed if necessary... [15:31] popey: can you try clock app? [15:31] anyone done shorts yet? [15:31] sil2100: please do camera_app meanwhile [15:31] nik90: ping me when you need me to build a click, and ping fginther to force the app through jenkins. [15:31] didrocks: should I go to camera_app? Ok [15:31] popey: ok [15:31] cyphermox: plars is doing shorts [15:31] cyphermox: can you try clock? [15:31] ok [15:31] didrocks: wait, but we had results for camera_app already [15:31] didrocks: sure, lemme test clock [15:31] didrocks: from smoketesting? [15:31] ah ok [15:32] so, once we have shorts [15:32] didrocks: shorts just finished, all pass, what's next? [15:32] we have 2 issues [15:32] clock [15:32] and calendar_app [15:32] plars: can you try clock as well? [15:32] calendar_app, I'm in French [15:32] cyphermox: mind trying? ^ [15:32] on it [15:32] (in english) [15:32] meanwhile, I'm reverting to english [15:33] trying what exactly? [15:33] sil2100: maybe try as well? ^ [15:33] clock in french? [15:33] cyphermox: calendar_app in english [15:33] ping doanac, I think the jenkins in 91.189.93.70 is not using the latest autopilot version. [15:33] can you check please? [15:33] hmmm [15:33] didrocks: right [15:33] Let me try that [15:33] But wait [15:34] I'm on english anyway I think, so clock anyway fails I think [15:34] didrocks: should I try calendar in english? [15:34] sil2100: yeah, possible it's something else [15:34] sil2100: but yeah, calendar in english [15:34] I hope it's failing because I was in French [15:34] those apps passed yesterday when i uploaded them to the store [15:34] (in english) [15:34] elopio: ack. [15:34] fwiw [15:34] popey: clock is weird… I don't see anything that can impact it that entered [15:35] just rebooting to run it here [15:35] popey: I trust your environment :) [15:35] * popey looks around at his environment [15:35] popey: btw the clock app tests will now fail since rev 393 will check for alarm status which was broken today [15:35] popey: I'm sure it's the UK humidity! [15:36] sil2100: just by any luck, did you set your timezone to something else than UTC? [15:36] nik90: ok, we need passing tests, balloons can help? [15:36] didrocks: mine is BST [15:37] didrocks: what's up [15:37] davmor2: BS Timezone? :p [15:37] didrocks: http://imgur.com/E2Gzh0O my environment - including inspirational photo of jono bacon [15:37] davmor2: we have clock and calendar AP tests failing [15:37] popey: ahah! I like the inspirational photo in particular :p [15:37] but I wouldn't have use that word [15:37] ☻ [15:37] haha [15:37] no offense jono :p [15:39] didrocks: no, it's all default :| [15:39] hmmm [15:39] didrocks: but from the failures I see that hm, there seem to be some DBus issues there [15:39] org.freedesktop.DBus.Error.NoReply [15:40] So it can't find some views simply [15:40] Looked like flakyness to me, as I had an increasing number of test failures [15:40] Maybe I needed a reboot ;/ [15:40] yeah, can be [15:41] didrocks: I had my queue running. calculator, terminal, shorts all pass. shall I move it to clock now? (it's currently running calendar) [15:42] Mirv: yeah, calendar and clocks are the unknown remaining ones [15:42] oh, maybe I'll let that calendar run to completion then [15:42] Mirv: thanks! [15:42] clock still running here. [15:42] calendar still rerererunning [15:43] do I need to do something special to run the tests? I have 18 failures right now :/ [15:44] cyphermox: for click apps, see the wiki page for the setup [15:44] Ran 18 tests in 495.503s [15:44] FAILED (failures=7) [15:44] again [15:44] I did [15:44] cyphermox: on which one? [15:44] calendar [15:44] clock or calendar? [15:44] didrocks: I had 4 errors and 20 passes on a local run of clock [15:45] Ran 20 tests in 523.686s [15:45] FAILED (failures=5) [15:45] * didrocks don't get it [15:45] so we all have failures [15:45] try to roll back autopilot ? [15:45] didrocks: and there was a qmlscene crash with it too [15:45] ogra_: no, it's a dependency which is added [15:46] didrocks, i know, i worked the whole night on it [15:46] didrocks, but its also a rebuild of the package ... if deps changed etc ... [15:46] ogra_: python? [15:46] i was voting for a seed change instead ... [15:46] where is sergio? [15:46] off i think [15:46] I wonder if the bzr branch revno changed [15:46] with new commits [15:48] elopio, doanac, it's indeed using an older version. I'll update [15:48] http://paste.ubuntu.com/7268887/ [15:48] thats the log of my clock run [15:48] thanks fginther. [15:48] popey: we don't have the rev of the bzr branch [15:48] https://launchpad.net/ubuntu/+source/autopilot/1.4+14.04.20140416-0ubuntu1 [15:49] didrocks, see the second diff :/ [15:49] oh. [15:49] popey: there are some errors [15:49] didrocks: calendar pass for me [15:49] ogra_: urgh? [15:49] Mirv: phew! [15:49] Mirv: can you try clock? [15:50] Mirv: seems your machine is magic :) [15:50] didrocks, not sure what it means if LP shows two [15:50] ogra_: not [15:50] ogra_: no [15:50] it's the first one [15:50] the real one [15:50] the older revision seems quite old actually [15:50] phew [15:50] why is it generating two ? [15:50] yeah [15:50] didrocks: all tests for calendar pass! [15:50] ogra_: no idea TBH, maybe an old version staying the in the silo? [15:50] sil2100: \o/ [15:50] ok [15:50] didrocks: yeah, nothing weird in there http://pastebin.ubuntu.com/7268898/ . running clock now. [15:50] so my phone is screwed [15:50] * didrocks opens the window [15:51] testing GPS in motion ? [15:51] will transform it to an ogra's phone :p [15:51] popey: I got the same number of failures on first run of clock app... [15:51] ogra_: ahah [15:51] :) [15:51] popey: I see some failures on your pastebin yesterday [15:51] still don't understand why 301 passed [15:53] ogra_, popey: Guys the test is failing due to https://bugs.launchpad.net/ubuntu-clock-app/+bug/1309057 [15:53] Launchpad bug 1309057 in Ubuntu Clock App "Clock app crashes when trying to access a saved alarm" [Critical,Confirmed] [15:53] didrocks: ^^ [15:53] aha [15:53] nik90: hum, but why was it passing on 301? [15:53] ogra_: I have no idea what is causing the crash [15:54] nik90: the image diff is oxide + unity-mir + webbrowser-app [15:54] nothing else [15:54] didrocks: erm [15:55] nik90: but you're right, I'm seeing this QML Scene crash [15:55] still don't know what the magic which triggers it now is [15:55] didrocks: I can confirm it locally on my system. It seems to crash when accessing the alarm status [15:55] nik90: oh that's weird if you click on the name of the alarm instead of the time of the alarm the time is blank [15:55] it's really random [15:56] nik90: for instance: [15:56] Ran 20 tests in 554.815s [15:56] * balloons tries to keep up [15:56] FAILED (failures=2) [15:56] only 2 issues [15:56] crashes [15:57] davmor2: when I click on an alarm (on the label or time), It crashes...Segmentation fault (core dumped) [15:58] davmor2: and I have tried this with older versions of the clock as well..same result [15:58] ok [15:58] seems something to release note [15:58] but not block on it [15:59] (as alarm are more or less already useless due to the Qt issue) [15:59] sounds sane? [15:59] +1 [16:00] and apps can be updated independently [16:00] and bizarrely my phone is now alarming me that we have a meeting [16:00] oh, this meeting notification always gives me a heart attack [16:00] * ogra_ wants adjustable sounds for that ! [16:01] heh [16:01] ogra_: but you are still late! === bschaefer_ is now known as bschaefer [16:01] :p [16:01] ARGH [16:01] didrocks: + [16:01] Meeting! [16:01] didrocks, yeah, not used to having to have meetings on a release day [16:01] nik90: let's do that then! [16:01] cyphermox: coming? [16:02] didrocks: i think you are correct [16:02] release note, and go [16:02] oki! [16:03] indeed [16:03] what else would you do ? [16:03] not release ? [16:03] :P [16:03] we can relase the image 2 days ago :) === gatox_lunch is now known as gatox [16:03] didrocks: will you work with stgraber ? [16:03] that half breeded crap ? nah [16:03] :) [16:04] asac, we'll manage, dont worry [16:04] i dont get though why we got a regression on clock in last two/three landings [16:04] ok goodie [16:04] asac, not only clock [16:04] calendar seems to be flaky too === plars changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: doanac | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: none [16:04] ogra_: and those are new? [16:04] seemingly [16:04] what changed? [16:05] note: dont think it has an impact on wehtehr to release or not [16:05] we didnt have that many errors on the dashboard for any of the recent images [16:05] asac, well, i would blame AP :) [16:05] its the only thing that could explain it [16:05] i am sure its evdev [16:06] evdev also breaks runtime behavioru :) [16:06] i am sure [16:06] well, nothing broken here [16:07] during simple usage [16:07] i thought that the clock and calendar app is even more odd now [16:07] not only event [16:07] but completely weird behaviour [16:07] yeah [16:07] so thats what regressed in last 2 days [16:07] but both are apps [16:07] or last day even [16:07] they can be updated in the store [16:07] but the apps didnt change [16:07] so we broke them [16:07] with evdev [16:07] or mir [16:07] or unity [16:07] as long as we dont discover serious systemic issues we're fine [16:08] we had evdev for ages [16:08] didrocks: clock-app one failure http://pastebin.ubuntu.com/7268990/ [16:08] asac, even 300 tested with evdev [16:08] if that was the last one, I'll let it rerun [16:16] Happy release daaay \o/ [16:16] Mirv: we are good it seems :) [16:16] shortest metting EVA! [16:16] Mirv: we are going to promote to stable [16:16] * sil2100 is dancin' [16:17] happy release! [16:17] didrocks: we need to have a more public discussion / announce of how the next set of images will work [16:17] IMO [16:17] didrocks: \o/ !! [16:17] cjwatson: all is good on our side, ogra_ is going to promote to the devel channel and work with stgraber to put it in the stable one [16:17] because people will want to know "why am i getting no updates" [16:17] popey: more than agreed [16:17] happy release, everyone [16:17] popey: and I'm with you on that one ;) [16:17] popey: but let's see next week [16:23] didrocks: can we send a mail like last time that landing team will resume operation on tuesday? [16:23] but adjourns until then as the archive needs bootstrapping etc.? [16:23] asac: already in my daily email draft :p [16:23] ah, yeah, good idea [16:23] ok, i think it would justify keeping it a separate mail [16:24] for adjourning [16:24] asac: people will want to merge to their trunk [16:24] just short [16:24] asac: we'll accept SRU [16:24] asac, didnt you say you are pessimistic ? [16:24] asac: I'll just tell we'll start working with upstream on SRUs on tuedsay [16:24] right [16:24] asac, "the archive open on tuesday" definitely is very optimistic [16:25] tell them that landing team recedes foor eastern vacation and will be back operational helping on SRUs and starting to ramp up on normal landing business on tuesday [16:25] and thank everyone for the amazing release [16:25] asac: the thanks was already in the other email, but I can always thank twice! [16:25] didrocks: yeah, you can state that depending on how archive bootstrapping goes we might start doing that earliest on tuesday [16:25] sure thanks [16:25] twice :) [16:29] second email -> loaded :p [16:29] g+ -> prepared [16:30] * didrocks is going to be a spam machine [16:30] yay, spam us ! [16:30] ogra_: is it done yet? dad dad dad ;) [16:30] * ogra_ hands didrocks a lollipop [16:31] soon kid soon [16:31] \o/ [16:34] ;p [16:37] sil2100, didrocks so what's the plan for silo 4? marked testing pass, what should I do, just publish that and let it sit in UNAPPROVED until U opens? [16:37] thank you landing and release teams for taking care of our projects !! :) [16:37] dbarth, you're welcome! [16:37] robru: is it a feature? [16:37] yw dbarth :) [16:38] didrocks, fix a crasher bug (line 21 in spreadsheet) [16:38] * ogra_ isnt sure yuo can let it "sit in unapproved" [16:38] the changelog will target the wrong distro [16:38] robru: that's ok for today to publish, it will go to -updates [16:38] as a 0-day SRU [16:38] (to have to go to U) [16:38] yeah, will work as 0 day [16:38] all others will need to follow the SRU process [16:39] didrocks, can somebody explain to me the concept of the 0-day SRU? do I just have to mark the bug to look like an SRU or does it really need full SRU team acking? [16:39] nik90: https://bugs.launchpad.net/ubuntu-clock-app/+bug/1309103 initial bug report I've added that from your initial digging it looks like it is a lower level than the clock app [16:39] Launchpad bug 1309103 in Ubuntu Clock App "In the latest image alarms now say No Active Alarm" [Undecided,New] [16:39] robru: no, you don't need to, it's like "it was too late, delivered before the release and it's in -updates" [16:40] robru: so that people dist-upgrading to trusty just get it [16:40] davmor2: thnx..but I reported it already at https://bugs.launchpad.net/ubuntu-clock-app/+bug/1309041 [16:40] Launchpad bug 1309041 in Ubuntu Clock App "Alarms are shown as disabled despite it being enabled and ringing at the intended time" [High,Confirmed] [16:41] didrocks, yeah, I get that people will update to it after the release, just wondering how much of the full SRU process is necessary to get this published from silo into trusty-updates [16:41] nik90: oh nice [16:41] robru: until the release is done (for one line fix like that), I would say nothing [16:41] robru: just after, I would say… whole process [16:41] ok [16:42] they should just get it into U and it will show up in the next devel image [16:42] :P [16:42] who runs stable anyway [16:42] ;) [16:42] (except press people for making reviews) [16:43] mhr3, ok I published silo 4, I guess that will find it's way into trusty-updates eventually [16:43] didrocks, so how much do i have to pester the release team to get that through? will they just know and accept it? [16:44] robru: I think don't bother them with it [16:44] ok... [16:44] doesn't seem that important and don't affect desktop [16:44] and on touch, people will only get it with a new image build [16:44] didrocks, but if I don't pester them, how will it get into -updates? [16:45] robru: well, they are looking at the queue regularly (especially when they will open -updates) [16:45] ok [16:45] robru: if you don't see it accepted in 6 hours, maybe ping the available ones [16:45] ok [16:45] you can always fallback to Laney anyway, if he doesn't have a beer in one hand :p [16:45] didrocks, thanks. this is sure a strange transition we are in [16:45] robru: yeah, it's always special ;) [16:46] robru: and I don't testify everything I'm saying is 100% accurate ;) [16:46] just my interpretation ;) [16:48] popey, dbarth: is there a bug for youtube videos not opening, I know it's known but I'm just after a bug number for the chart [16:48] pass [16:48] i dont have one [16:59] davmor2, heh, i was assuming teher is one [16:59] but apparently there isnt [17:00] there is one but mostly on desktop,for flashplayer [17:00] davmor2: point to: https://bugs.launchpad.net/oxide/+bug/1303997 [17:00] Launchpad bug 1303997 in Oxide "youtube does not work" [Undecided,Confirmed] [17:01] robru, despite the conflict? [17:01] dbarth: thanks [17:01] mhr3, which conflict? [17:01] robru, > OTHER SILO ALSO HAS unity-scope-click, proceed with caution! [17:02] mhr3, it just means that silo 11 will need to be rebuilt after this silo lands, but before it can be published. [17:02] robru, indeed, can you add a comment about that to 011? [17:02] mhr3, yep, just did [17:02] robru, ok, thx === plars changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: plars | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: none === alan_g is now known as alan_g|EOD [17:13] nik90: http://davmor2.co.uk/~davmor2/screenshots-phone/device-2014-04-17-181235.png [17:19] robru: landing-006 ready to go [17:20] boiko, published! [17:20] poor thing ... [17:21] published into jail [17:25] jail? [17:25] is that a new mechanism? [17:25] we have silo and jail :) [17:25] maybe we could use the same for both :P [17:25] robru: nice, thanks! [17:26] asac, jail was always there, it just had a nicer name before the release freeze... [17:26] boiko, you're welcome [17:27] asac, we have a jail ... there is a sign saying "Unapproved" at the door ;) [17:28] i like the idea [17:28] we might want to take that and work that into our train concept :) [17:28] ogra_: what is jail? upload queue waiting area? [17:29] asac, my word for unapproved after release :) [17:29] robru: this unapproved queue, does it require any action on my side? [17:29] so similar to new queu [17:29] i was referring to the unapproved queue [17:29] boiko, hmmmm.... no. basically we just have to wait for the extremely overworked release team to notice and approve the upload, just takes some time. [17:29] +1 on the jail idea [17:29] lol [17:30] * ogra_ likes how his jokes turn into reality [17:30] didrocks: maybe we should rebrand CI Airline to CI Alcatraz :) [17:30] j.k. [17:30] asac: where are the sharks then? ;) [17:30] or rather "who"? [17:30] we have plenty of those, dont we? [17:30] easy answer :p [17:30] didrocks, just go to #ubuntu-release-party ... we store them there today [17:30] ahah [17:31] haha [17:31] robru: ok, good :) [17:31] ogra_: how many are in there? [17:31] 513 [17:31] hmm [17:31] ok [17:31] it bumped to 560 or so inbetween [17:31] thats pretty ok [17:31] what was our record? [17:31] 1400? [17:31] in 08 09? [17:31] nah, below 1000 [17:31] sure? [17:32] 980 or something [17:32] i am quite sure we had a thousand at least [17:32] popey, might remember [17:32] wow [17:32] we have nearly 2000 in #ubuntu today [17:33] are there some alltime stats for freenode channels somewhere? [17:33] i am sure that must exist [17:34] if someone has them its AlanBell i guess [17:34] he has bots everywhere [17:34] pass [17:34] ha [17:36] bah ! [17:36] https://twitter.com/MarkABaker/status/456839297878863872 [17:36] i want such a cupcake ! [17:37] very good [17:37] ogra_: make a photo reply, post how your desk is prepared-by release [17:37] :) [17:37] ugh, no ! [17:38] nobody wants to see that [17:38] hehe [17:38] well, you can see a place where hard work was done [17:38] so might be cool :) [17:59] asac: ogra's isn't the only desk nobody need to see [18:00] haha [18:06] ogra_: I have a smattering of desktops, laptops, tablets, phones notes, books and DVD-RW scattered liberally :D [18:12] sounds we're all good! have a nice week-end for those being on easter holidays! [18:13] nice work again, see you ;) [18:13] didrocks: bye! get rest [18:13] o/ [18:14] * sil2100 still has some code to write [18:18] so say we all [18:19] have fun with the release everyone and see you next week! [18:22] see you! :) === jhodapp is now known as jhodapp|lunch [19:16] robru: can I have a silo for line 22 please? [19:16] bfiller, ok, you got silo 1 [19:17] robru: thanks [19:17] bfiller, you're welcome! happy release-ter! [19:17] releaster ;-) [19:17] robru: same to you :) === jhodapp|lunch is now known as jhodapp [19:52] releaster very good [19:55] lol [19:55] releaster [19:55] wish I thought of it [20:02] robru: silo1 ready to be published [20:09] pmcgowan, ChickenCutlass ... yeah, we've been releasing in april for 10 years, how did nobody ever think of releaster before? [20:09] robru: its fantastic [20:09] bfiller, published [20:09] ChickenCutlass, thanks [20:11] I guess easter is at a slightly different time each year, so it wouldn't always be so close. [20:11] I bet somebody thought of releaster in 2003 though... [20:12] (because easter was april 20th then) [20:13] robru: cheers [20:31] ... 2003? Isn't that a year too soon? [20:37] ToyKeeper, maybe ;-) [20:37] ToyKeeper, realized it shortly after saying it [20:37] ToyKeeper, 2006 then. easter was just a few days off release then [20:39] But I think this is the first year that we get to have our post-release coma during the time from Good Friday to Easter Sunday. [20:40] I don't know about anyone else, but I plan on sleeping for a few days. I'll wake up when those tasty little cream eggs go on sale; time to stock up. :) [20:41] 6.04 was delayed to 6.06 [20:41] :P [20:43] TBH, the first time I actually installed Ubuntu was in 2008. Before then I had all Debian boxes running testing or unstable on a "update when I feel like it" schedule. One install lasted from 1997 to 2008. [20:44] (er, installed for my own use anyway... I happily set it up for other people though) [21:34] robru: can you rebuild mediaplayer-app in silo 17, from a new MP push? === plars changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: none === jhodapp is now known as jhodapp|afk === bschaefer_ is now known as bschaefer