=== _salem is now known as salem_ === salem_ is now known as _salem [01:10] Who's testing OTA-7? I wanted to bring bug 1501330 to people's attention [01:10] bug 1501330 in webbrowser-app (Ubuntu) "[webapp-container] Crash with open context menu" [Medium,Confirmed] https://launchpad.net/bugs/1501330 [01:14] mterry: bit late for ota-7 [01:14] mterry: maybe poke davmor2 or jibel [01:15] popey, I know it's late, but it's a bad bug [01:16] mterry: sure, we have a few :) [01:16] popey, ok... it's a bad regression let's say [01:21] mterry: maybe ping davmor2 / jibel / sil2100 a mail? [01:24] yeah maybe. Maybe no one else thinks it's that bad and we ship anyway, which is fine. But maybe they agree === _salem is now known as salem_ === salem_ is now known as _salem === chihchun is now known as chihchun_afk [06:34] morning [06:34] trainguards, publish https://requests.ci-train.ubuntu.com/#/silo/035 please? [06:34] Saviq: let's see [06:39] Mirv, thanks! === chihchun_afk is now known as chihchun [07:59] cihelp, hey, seems D09add_ppa~ci-train-ppa-service~stable-phone-overlay does not add overlay to wily builds? [08:33] jibel, sorry about fix-releasing the bugs, I was fixing the status for unity8 and got trigger-happy for the image tasks, only realized later that might not be what we want [08:33] Saviq, np [08:40] trainguards, can you guys help with the version number of unity-notifications here https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-022/+sourcepub/5519545/+listing-archive-extra ? train insists on keeping the .is. part, do I need to bump the version to avoid that? [08:41] :D [08:41] Saviq: I think the answer is yes [08:41] Saviq: yeah, I already mentioned that a week ago... the train changed behavior and doesn't treat reverts properly [08:41] So you need to change it in the changelog [08:42] sil2100, yeah, I tried, but it was still lower in changelog because the .is. was for 15.10 [09:56] sil2100, is there a channel that does not use stable-snapshot for the overlay? /me just ground some teeth to understand why mir 0.17 wasn't available on my phone... [09:59] Saviq: my impression on channels is that there's a channel for EVERYTHING. precise + phablet PPA? sure! [09:59] (just joking about the example) [10:00] yeah, I think it wasn't clear to me that there actually is a snapshot of the overlay PPA, and that rc-proposed is switched to it during freezes [10:03] right, I can see that might be confusing. there's no overlay visibility on images until the OTA is out. [10:04] Saviq: sadly, no channel yet... [10:04] sil2100, does everyone know that to properly test vivid+overlay today they need to modify the PPA? [10:05] I'd think jibel's team does [10:05] Saviq: the reason for that is that we actually have only one infra for building images, so building every image would require manual switches to make sure we don't pull in the wrong rootfs into a wrong channel [10:06] Normally we'd use rc-proposed for that and just keep the snapshot images somewhere else, but the channel layout has many dependencies [10:06] So to have the snapshot images promotable for all devices, it's spread over multiple channels depending on rc-proposed/ubuntu [10:07] sil2100, I understand the reason, just I'm worried people might be surprised by it [10:07] it might not be clear that this modification is needed [10:07] at latest when you have a dependency in overlay [10:07] Saviq: will mention it in the landing e-mail if we won't unblock rc-proposed [10:11] sil2100, I think we need a better channel for those kind of announcements, landing email is quite high throughput and people miss it (like non-landing people were not aware of overlay getting enabled on wily) [10:12] sil2100, maybe just a wiki page with current status that people would subscribe to or something... [10:12] Saviq: sil2100 Mirv: what? That's not a train-generated revert, train reverts use ~is instead of "is" specifically so the new version generation still works. [10:13] robru, yeah, that was likely some manual thing around gcc transition [10:13] Saviq: weird, train has good support for generating reverts, very easy to do, people should use it ;-) [10:14] sil2100: i don't believe any revert-related logic changed recently [10:17] robru: well, in the old times that's how reverted version looked like [10:17] .is. [10:17] Saviq, do you have an example? (D09add_ppa~ci-train-ppa-service~stable-phone-overlay does not add overlay to wily builds) [10:17] Since ~is wouldn't work as it's smaller than the previous version [10:18] robru: didrocks's reverter was changing versions of 1.2-0ubuntu1 to 1.2.is.1.1-0ubuntu1 [10:18] sil2100, the bot will be down for a while, the instance on which it was running is lost [10:18] jibel: forever? [10:18] sil2100, yes [10:18] Wait, what? [10:18] Why? [10:19] sil2100: because the cloud! [10:19] sil2100, the base image the instance used got deleted [10:19] So I'll have to setup stuff again myself as well [10:20] sil2100, so only the diff remains which is pretty useless [10:20] sil2100, I don't know for you [10:20] sil2100, ask is [10:20] IS* [10:20] sil2100: yes i specifically changed it to use ~is so that newer releases could work without bumping versions [10:20] I had my scripts on lcy01 [10:21] robru: so how does it revert the version then? [10:21] robru: let's say we have a package version of 1.2-0ubuntu1 and want to revert it to 1.1-0ubuntu1 [10:21] robru: what will be the generated revert version? [10:21] sil2100: the reverted version is higher because of the period. Eg 1.2.~is is higher than 1.2 [10:22] sil2100: it would be 1.2.~is.1.1 [10:22] Ok, so the dot [10:23] sil2100: you can try it in a silo, it's live in production. Just assign a silo with a manual source, then run Jenkins "revert" job, will upload a revert to the silo automatically, very easy, no manual fuss [10:23] Anyway, the .is. was working pretty well in the past [10:23] And it's anyway strange that the train just continued to use that version [10:24] sil2100: .is. Didn't work because building a new version sorted lower, 1.2.1 is less than 1.2.is, but not 1.2.~is. [10:24] sil2100: it's not strange, the train tried to generate a new version, but it was less than .is. So it grabbed that version instead [10:25] robru: wait, what? Why? Since a rebuild is just enough to get a higher version and that's what we wanted to have [10:25] This exact problem is why i switched the reverter to use ~is [10:25] robru: for instance, if you have 0.2+15.04.20150930.is.0.2+15.04.20150922.2-0ubuntu1, then a 0.2+15.04.20151001-0ubuntu1 is good enough [10:26] sil2100: well i don't know the exact versions involved but if the silo/archive has a higher version than what the train generated, the train uses the higher version [10:26] fginther, jobs http://s-jenkins.ubuntu-ci:8080/job/unity8-ci/6474/ [10:26] Saviq: what were the exact versions in this case? [10:27] Just saying that .is. was working fine, and that a change to .~is. wasn't super necessary [10:27] Still, I would want to have certainity that the .is. case is properly handled as a revert even now as well [10:28] Since we have multiple packages with versioning like that [10:28] sil2100: change was made a long time ago, I can assure you I did it to solve a problem I faced at the time, I don't do things arbitrarily [10:31] sil2100: at any rate this seems like a bug in the code that generates version numbers, not a bug in the revert logic. [10:31] robru, unirt-notifications 0.1.2+15.10.20150623.is.0.1.2+15.10.20150804-0ubuntu1 [10:31] *unity [10:32] fginther, for example the wily qmluitests job under there [10:36] sil2100: this case is particularly confusing because the original "revert" was 0.1.2+15.10.20150623.is.0.1.2+15.10.20150706.1-0ubuntu1 (note the "reverted" version is higher than the new version). That makes no damn sense. It reverted to a newer version [10:37] sil2100: anyway I filed a bug, will fix tomorrow https://bugs.launchpad.net/cupstream2distro/+bug/1506417 [10:37] Ubuntu bug 1506417 in CI Train [cu2d] "Version number generator chokes on .is. revert packages." [Undecided,Triaged] [10:40] robru: I'm going to link this to you again. There's TL;DR too. https://www.reddit.com/comments/meaq9/the_real_quick_fix_to_get_your_fucked_up_sleep/ [10:40] Saviq, thanks. I think I see the problem. The qmluitest is a special snowflake and appears to be missing the hook to add the overlay in the correct location [10:41] it's starting to shift towards robru ending his Wednesday later than me ending my Thursday. [10:41] Mirv: but it's not even 4AM! [10:42] * Mirv mutters something about hopeless case [10:42] Mirv: yeah I guess the protein smoothie at 1AM was a bad move [10:42] Mirv: yeah you have grounds to talk :P [10:43] fginther, sry about that :| [10:43] Saviq, no worries, should be an easy fix [10:45] robru, sil2100, I've a feeling citrain tool should deal with stable-snapshot somehow as well in that case [10:45] Saviq: it treats stable-snapshot the same as overlay... is that not what you want? [10:45] robru, no, because train doesn't know about stable-shapshot [10:46] Saviq: uh, no? the apt pinning pins the silo above all else. overlay, snapshot, and archive are all "disabled" (in that deps will get pulled in if necessary but are otherwise ignored) [10:46] robru, my problem today: tried to citrain silo 22 and it failed in deps, after some digging I found out that overlay wasn't on the phone, stable-snapshot was, instead [10:47] robru, when overlay goes ahead after the snapshot, things get tricky (because qtmir, correctly, built against mir released yesterday, but the phone couldn't find it, because it's on stable-snapshot) [10:48] Saviq: then I guess it's up to you to add overlay ppa before using citrain tool? I'm not sure what you're expecting to happen in that case. [10:48] robru, well, yeah, that means you can't rely on the tool in general [10:48] Saviq: the bigger question is, why are you developing against a stable image? shouldn't you be using a later image with overlay ppa? [10:48] robru, there isn't oe [10:48] one [10:48] bah [10:49] robru, rc-proposed is on stable-snapshot until OTA7 is released [10:49] I have no idea why stable-snapshot even exists. Apparently our stable overlay wasn't stable enough. [10:49] indeed [10:50] well, I understand why it exist [10:50] s [10:50] otherwise we wouldn't be able to land between spinning the first rc-proposed image and its release [10:51] but the solution seems flawed, at that point I'm starting to feel like the rtm approach was clearer [10:51] Saviq: I'm off to bed soon, ask around if it makes sense to just unconditionally add overlay PPA when installing silos (so that stable phones have both stable snapshot and stable overlay). then I'll add that tomorrow with other fixes [10:51] anyway, //food first [10:52] robru, that might be a solution indeed [10:52] Saviq: I don't really understand the implications fully but it seems to me that if you're developing a silo, you want the overlay, not the snapshot [10:52] robru, totally [10:52] anyways, goodnight [10:52] o/ [11:13] kgunn, FYI silos 16 and 35 landed [11:13] sil2100: need moor power [11:13] :| [11:14] * pete-woods has kinda given up on landing that silo until you become a core-dev [11:14] jibel, how we doing this morning [11:15] pmcgowan, lol [11:15] pmcgowan, check your email :) [11:15] s [11:16] pmcgowan, OTA7-wise last image is okay, we re-ran a round of tests on it and it's fine. However we'll have a respin [11:16] jibel, so I see [11:18] Yeah, exciting morning [11:32] cihelp: are there plans to enable automated testing on desktop? IIRC otto was discontinued but I was told that something better, with real hw, would replace it === alan_g is now known as alan_g|lunch === _salem is now known as salem_ === dbarth_ is now known as dbarth === alan_g|lunch is now known as alan_g [13:16] fginther, so https://code.launchpad.net/~unity-team/unity8/use_sdk_13/+merge/271603/comments/693152 shows that vivid passed but wily didn't, because one had the overlay and the other did not [13:20] Saviq, I'm testing a fix. When it's working, I'll re-build that MP [13:20] fginther, ack, thanks [13:29] oSoMoN, that's a question better answered by ev or Ursinha [13:30] ev, Ursinha: are there plans to enable automated testing on desktop? IIRC otto was retired but I seem to remember that something better, with real hw, would replace it [14:26] trainguards: Hi there, I would need permissions to create silos. I need to land some changes for the sound indicator [14:28] xavigarcia: hey! What's your LP ID ? [14:28] sil2100: xavi-garcia-mena [14:29] jhodapp: Silo 55 approved, as you can see [14:29] :) [14:29] rvr, yay [14:30] sil2100, can you please publish silo 55 then? [14:30] xavigarcia: added you to the team, you should be good now :) [14:30] jhodapp: looking [14:30] Fingers crossed that I have enough permissions to publish it [14:54] Saviq, is this an acceptable solution to the wily qmluitests? http://s-jenkins.ubuntu-ci:8080/job/unity8-wily-amd64-qmluitest-ci-test/2/ [14:55] fginther, it's green, so I'd say so :) [14:55] Saviq, it doesn't use the VMs as the current test does and just runs the whole thing as a regular pbuild with the extra hooks to install xvfb [14:55] fginther, but let me verify it actually gets packages from overlay [14:56] fginther, hmm I can't see in the log that it actually downloads any package from the overlay ppa [14:56] Saviq, I tested it with trunk, I can rerun it with the other MP [14:56] fginther, yeah, that'd be useful as that's where we have the problem (requiring UITK 1.3 fix that's only in overlay) [14:56] rvr, jhodapp xavigarcia thanks for your work on silo 55 [14:56] ! [14:56] yeah, glad to see it landing! [14:57] thanks jibel and rvr for your patience [14:57] Saviq, ack. The log indicates that the PPA was added, but no packages were pulled from it. I'll update you with the results from the MP having problems [14:57] fginther, it might be because we're not dpkg-depending on anything, so if the packages are there, it's fine [14:58] tsdgeos, can you add a >= 1.3.1679 for UITK dep in your 1.3 branch? [14:58] 1676 rather [15:00] jibel, jhodapp: brilliant it was finally approved! \o/ [15:00] Saviq: ok [15:01] sil2100, seems that qtmultimedia got hung up landing though [15:01] jhodapp: *sigh* yeah, we'll need a core dev [15:01] kenvandine, mterry: could you guys look at https://requests.ci-train.ubuntu.com/#/publishable and publish those 2 silos? [15:02] Saviq: pushed [15:03] sil2100, who can we ask? [15:04] jhodapp: poked kenvandine and mterry [15:04] tsdgeos, thanks [15:04] sil2100, perfect thanks [15:05] fginther, we now have a dependency on a package from overlay in that MP https://code.launchpad.net/~unity-team/unity8/use_sdk_13/+merge/271603 [15:05] Saviq, thanks, giving it a test now [15:07] sil2100, looking [15:07] kenvandine: thank you! [15:18] sil2100, jhodapp: done [15:18] thanks kenvandine [15:18] np [15:21] oSoMoN: ping [15:25] rvr, pong [15:25] oSoMoN: Checking silo 14 [15:25] oSoMoN: In the previews, I see none for Google Maps [15:26] oSoMoN: It's weird, because I can see previews for Google Maps on the tab list [15:27] rvr, yes, I was seeing that too in my testing, not sure exactly what’s happening there [15:27] rvr, as it was only for google maps I decided to let it through and will file a bug to investigate properly the issue [15:32] sil2100: would you mind verifying I setup https://requests.ci-train.ubuntu.com/#/ticket/518 correctly? this is not a dual landing (upload already pushed to wily) and I plan to upload directly to the ppa [15:32] jdstrand: looking [15:33] jdstrand: looks good so far! [15:33] Let's assign it and upload [15:34] jdstrand: assigned, silo 016 [15:48] oSoMoN: In the top sites, when a site is long tapped, only the first one is selected (grayed) [15:50] rvr, yeah, the highlight should be disabled when there is no keyboard (it indicates keyboard focus), that’s a minor issue that sneaked through [15:55] popey, jibel, rvr, davmor2, robru, ogra_: I think we might skip the LT meeting today since we have the malicious meeting right now [15:55] yeah [15:55] Not sure if we'll finish in time [15:56] OK [15:56] our snappy standup runs over as well here [15:56] sil2100: ok [16:04] Saviq, that MP with the updated dependency is passing here: http://s-jenkins.ubuntu-ci:8080/job/unity8-wily-amd64-qmluitest-ci-test/4/console [16:10] trainguards: Error in tempdir() using /tmp/XXXXXXXXXX: Could not create directory /tmp/19bxoPhTef: No space left on device at /usr/bin/uscan line 883. :o ^^^ [16:11] Trevinho: oh no, again [16:11] Give me a moment [16:11] sil2100: thanks [16:12] Trevinho: could you retry now? [16:12] k [16:21] sil2100: how come the rc-proposed image still pointing at stable-snapshot ppa? shouldn't it be using overlay-ppa again? [16:22] bfiller: no, since daily-builds are disabled [16:22] The rc-proposed channel is still used for OTA-7 as per announcement I made :) [16:22] (this will be fixed for next OTA btw., as per my earlier e-mail) [16:22] sil2100: ok, sorry I missed that [16:22] sil2100: it will switch to overlay once ota7 is released? [16:23] bfiller: yes, which should happen this week even I suppose [16:24] bfiller: the new idea is to use the rc channel for OTA-x re-spins and leave rc-proposed as overlay-only [16:24] But that's for the next release [16:24] sil2100: ok [16:32] rvr, how is the rest of silo 14 looking? [16:35] fginther, ack, it doesn't say anywhere where it downloaded it from, but as long as it's there, fine by me ;) [16:36] Saviq, thanks for examining it. I'll get it updated. [16:36] oSoMoN: The rest looks good. I'm not very happy with the two found issues, though. They are a bit annoying from the user perspective. [16:37] oSoMoN: OpenStreetMap doesn't show either [16:37] sil2100, you can free silo 47 btw, I don't need it anymore [16:37] rvr, I agree, but they qualify as minor issues, I’ll file high priority bugs for them as soon as it lands, if that’s ok with you [16:38] oSoMoN: Ok, please address them as soon as possible. [16:38] rvr, will do [16:40] rvr, thanks for the thorough testing! [16:40] oSoMoN: You're welcome :) [16:42] jdstrand: remember to upload to https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-016 when you have the packag ready [16:43] sil2100: yes, I tried to upload but it isn't there [16:44] I used lp:~ci-train-ppa-service/ubuntu/landing-016/vivid [16:44] jdstrand: hm, did you get a reject? [16:45] dput ppa:ci-train-ppa-service/landing-016 should just work [16:45] my dput configuration handles the lp:~ bit for me (or at least, it always has) [16:46] I didn't see a reject. let me see if it went somewhere [16:47] ah yes, I did get it [16:47] I forgot to include the orig.tar.gz [16:51] it is weird that I need it-- the ppa should have it... [16:52] ok, they are there now, and building [16:58] jdstrand: \o/ === alan_g is now known as alan_g|afk [17:30] sil2100, still around? [17:34] robru, ping [17:35] jhodapp: her [17:35] Hey [17:35] robru, hey man, you can free up silo 47 at your leisure...I don't need it anymore [17:36] jhodapp: thanks [17:36] np [17:36] jhodapp: actually you have the power to abandon if you want to try it [17:36] robru, oh, am I able to do that myself with the abandon button? [17:36] Yeah ;-) [17:36] sure let me try it, thanks [17:37] awesome, done! [17:37] sil2100, so is the silo loaded? testing? [17:39] jibel, silo16? [17:41] jdstrand, silo says ready to build [17:42] yes. it just finished [17:42] ah ok [17:43] pmcgowan, it is not ready for qa, is it? [17:44] guess not [17:44] jibel: I'll ping you. I want to do a quick test first [17:44] jdstrand: let me run a watch-only build to change the train status [17:45] Oh, you're doing that now :) [17:45] Good! [17:45] sil2100: I did already [17:45] is devel-proposed the proper channel for wily overlay? [17:45] jdstrand: yes, we build devel-proposed images from wily + overlay [17:45] rc-proposed for vivid + overlay of course ;) [17:46] jdstrand, can you ping alesage and ToyKeeper too [17:46] yeah, I knew about that one, wasn't sure what wily + overlay really meant [17:46] jibel: ack [17:46] I go exercise now for a bit, but I'll be around for package copies and image builds in a bit [17:49] oh sigh [17:49] udf gave me an error: [17:50] $ sudo ubuntu-emulator create --channel=ubuntu-touch/devel-proposed/ubuntu --arch=i386 devel-proposed.x86 [17:50] ... [17:50] Setting up... [17:50] open /home/jamie/.local/share/ubuntu-emulator/devel-proposed.x86/sdcard.img: permission denied [17:50] sure enough, sdcard.img is owned by root [17:50] * jdstrand attempts to flash a mako [17:58] rvr, FYI I just filed https://launchpad.net/bugs/1506579 as discussed earlier, and I will test again the lack of previews for google maps and openstreetmap (I haven’t been able to reproduce on desktop so far) [17:58] Ubuntu bug 1506579 in webbrowser-app (Ubuntu) "Highlight doesn’t follow right-clicked preview in new tab view" [High,Triaged] [17:59] oSoMoN: Cool [18:00] meh, I may not be able to test wily. I can flash an emulator and I can't flash a mako ('error pushing:' [18:01] * jdstrand tries different cable [18:34] trainguards, cihelp: anyone know how one might debug why occasionally tests are hanging for 10 minutes on launchpad buildsers, but seem to always work fine locally? [18:38] jibel: are you aware of any issues with the keyboard not coming up on mako? [18:38] jibel: rc-proposed [18:38] jdstrand, nope, I saw it once on krillin on first boot [18:38] * jdstrand feels like he is being thwarted at every turn [18:38] but no one could confirm [18:39] I can't flash an emulator, I can't flash wily overly and now my dogfooding phone doesn't have a keyboard [18:40] jibel: I was able to test the example clicks. feel free to test silo 016 whle I figure this out [18:40] jdstrand, a vivid emulator or wily? [18:40] jibel: I only tested vivid overlay due to the above [18:40] jdstrand, where is the test click? [18:40] I test vivid overlay mako with the updated packages [18:41] http://lillypilly.canonical.com/~jamie/click/1506467/ [18:42] thanks [18:59] dobey: dunno, sorry [19:01] jdstrand, alesage will do the verification of silo 16 [19:01] jdstrand, hiya [19:02] trainguards: hi, any suggestions on how to debug this [1] failure? It's been failing consistently, but I can't replicate it (even on am64). [1] https://ci-train.ubuntu.com/job/ubuntu-landing-021-1-build/197/consoleFull [19:02] dobey, maybe try an s-build locally? [19:02] alesage hey :) [19:02] jgdx: looking [19:02] robru, thanks [19:02] jibel, sil2100 I need to be gone for a while but you got it under control [19:03] pmcgowan, yes, it should be all right. [19:04] jgdx: did you see the full error log? https://launchpadlibrarian.net/221339749/buildlog_ubuntu-wily-amd64.ubuntu-push_0.68%2B15.10.20151015-0ubuntu1_BUILDING.txt.gz there's failures in the testsuite, that's not something I have any kowledge of, sorry [19:04] jgdx: ie, not a train problem [19:04] i did [19:04] why is it building for wily at all? [19:04] jgdx: because it's a dual silo. [19:05] okay, I thought wily was closed up [19:05] jgdx: "closed up"? wily builds go into overlay PPA until wily+1 opens to accept those packages [19:06] i thought vivid+overlay builds went into the overlay [19:06] but that makes sense, I guess [19:06] jgdx: currently dual silos are configured so that everything goes into the overlay. [19:07] okay [19:07] jgdx: wily is in feature freeze so phone stuff can't land new feature there, but that doesn't mean that we just ignore it. [19:07] jgdx: the policy is always that the newest devel release has to be developed for then features backported to vivid. you can't just develop for vivid, otherwise the devel release rots. [19:08] jgdx: so wily overlay ppa is just a holding spot until wily+1 opens so we can send packages there. [19:16] pmcgowan: yep [19:16] robru, ah, got it. I'll reproduce on wily---thanks! [19:16] jgdx: heh, you're welcome [20:04] robru: hey, I'm looking at https://requests.ci-train.ubuntu.com/#/ticket/518 (this is the click hotfix update). alesage is doing the QA testing. How do I say that this is 'Ready for QA'? [20:05] jdstrand: click edit and then change 'QA Required' to 'Ready for QA' in the dropdown [20:05] oh meh [20:05] I just looked at that [20:05] thanks [20:05] you're welcome [20:06] alesage: ok, I just now toggled this to 'Ready for QA'. delay was due to ignorance [20:06] alesage: do you have an questions on it? [20:07] jdstrand, not at the moment no, jibel directed me to the faux clicks for testing [20:07] Saviq: if you're still around: http://bazaar.launchpad.net/~robru/phablet-tools/further-fixes/view/head:/citrain [20:10] alesage: ok, so, in addition to the install of apps from the store, for those, install, see if it did what is expected, uninstall, then move to the next one [20:10] jdstrand, ack--testing updates atm [20:10] great [20:10] alesage: let me know if you need anything [20:10] jdstrand, shall do [20:11] robru, will try it out tomorrow, "loudly" feels weird when you can set -x, but ok [20:13] https://code.launchpad.net/~phablet-team/phablet-tools/trunk/+activereviews looks sad :P [20:13] fginther, looks like it didn't work... https://jenkins.qa.ubuntu.com/job/unity-phablet-qmluitests-wily/508/console [20:15] Saviq: set -x was spewing all kinds of garbage and printing things out of order. [20:15] Saviq: http://paste.ubuntu.com/12793387/ output is much cleaner now [20:16] Saviq: yikes @ activereviews. [20:17] alesage, jdstrand: testing click in progress? :) [20:17] robru, I'll be trying it out for sure [20:17] I did my part. it is 'Ready for QA'. alesage is testing it now [20:17] sil2100, yessir [20:18] Saviq: great, thanks [20:18] sil2100, on it [21:06] sil2100, jdstrand passing 16 in a moment [21:06] \o/ [21:06] \o/ [21:06] sil2100, copy to overlay and snapŝhot + respin ? [21:25] Yep! [21:25] jibel: I copy it to the OTA-6 snapshot for a OTA-6+ respin [21:26] sil2100, tell me when the builds are in progress and I'll send an email to the testing team for results tomorrow morning [21:27] jibel: ok, the copies are in progress [21:27] sil2100, what is the channel for OTA6 hotfix? [21:27] I think we'll have to use rc-proposed... ;/ Since otherwise we won't have the fix on all channels [21:28] So I would recommend first building the OTA-6+ image to rc-proposed, then next one the OTA-7 re-spin [21:28] That way rc-proposed users will not go too much into the past for too long [21:28] sil2100, so we cannot test OTA6 hotfix and OTA7 at the same time? [21:28] No, it's anyway not possible as we only have one building infra for rootfses [21:28] sil2100, is anyone using the rc channel? [21:29] So when we are building the OTA-6 version, we can't build anything until it finishes [21:29] not really efficient [21:29] Yeah, I know, but not much we can do here, that's how cdimage works [21:29] sil2100, you could build an OTA7 image push it to RC then build an OTA6 hotfix? [21:29] jibel: sure [21:30] We can do it like that [21:30] sil2100, this way we can tests both tonight [21:30] test* [21:30] I'll kick the OTA-7 one in a minute, since I want the binaries to publish fully [21:31] sil2100, make sure devs are aware that rc-proposed will contain OTA6 not the latest proposed [21:32] sil2100, otherwise there'll be some surprises tomorrow [21:32] jibel: yeah, I'll try to build an overlay-based rc-proposed image as soon as the OTA-6+ one is done... [21:32] But not sure if I'll be awake for so long === chihchun is now known as chihchun_afk [21:32] So probably tomorrow [21:33] sil2100: where did you push the ota6+? I don't see it in https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/stable-phone-overlay/+packages [21:34] sil2100, okay in summary which channels should I ask to review? rc (krillin build 28) for OTA7 and rc-proposed (krillin build 151) for OTA6+ correct? [21:35] jdstrand: you need to publish the silo, I don't have the permissions to do so - I copy-packaged it to our two snapshot PPAs [21:35] One we use for OTA-7 and one for OTA-6+ [21:36] jibel: let me check the OTA-6+ number, but yes, that's more or less the idea [21:36] jibel: yes, 151 as you said [21:37] The second image we will get [21:37] oh [21:37] good luck to remember this mess in a week ;) [21:37] I thought I saw someone else would do that [21:37] * jdstrand does it now [21:37] jibel: yeah, I proposed a better release model through e-mail for the next OTA [21:37] Actually using the rc channels [21:38] * sil2100 still waits for the package to pop up in the snapshots [21:38] sil2100, I agree with your proposal but didn't really think about the details yet [21:39] It wouldn't be so troublesome if we didn't have so many goddamned channels [21:39] sil2100, but it makes sense to use the rc channel for a Release Candidate :) [21:39] sil2100, it also means we'll never freeze proposed [21:40] which is what we discussed at the same time than the snapshot ppa [21:46] jibel: OTA-7 re-spin building [21:48] seb128, cool, thanks [22:00] trainguards: we need a new manual upload of qtpurchasing to silo 22, thanks! [22:01]