[02:05] === trainguards: IMAGE 232 building (started: 20140908 02:05) === [03:05] === trainguards: RTM IMAGE 23 building (started: 20140908 03:05) === [03:45] === trainguards: IMAGE 232 DONE (finished: 20140908 03:45) === [03:45] === changelog: http://people.canonical.com/~ogra/touch-image-stats/232.changes === [04:15] === trainguards: RTM IMAGE 23 DONE (finished: 20140908 04:15) === [04:15] === changelog: http://people.canonical.com/~ogra/touch-image-stats/rtm/23.changes === === robru changed the topic of #ubuntu-ci-eng to: Train support: trainguards | Vanguard: cihelp | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: Ping robru if any citrain jenkins jobs have unexpected results. [07:01] trainguards, can I please have a reconf on silo 19 [07:02] added qtmir-gles there [07:02] Saviq: I'm not here, but ok ;-) [07:02] robru, yeah, you should *not* be here [07:02] hmm :) [07:02] Mirv: ah, you're here. ok I got this one but I'm really not here ;-) [07:03] robru: when you're not really here, you should really really not be here :) [07:03] Mirv: who are you talking to? I don't see anybody here. [07:03] exactly [07:26] Mirv: have you heard anything about how to (a) disable the edges intro and (b) how to unlock the screen remotely? [07:30] bzoltan: a) like, other than phablet-config edges-intro --disable? b) maybe see http://bazaar.launchpad.net/~ubuntu-test-case-dev/ubuntu-test-cases/touch/view/head:/utils/host/reboot-and-unlock.sh [07:31] Mirv: the `phablet-config edges-intro --disable` does not do the job, so yes a working solution would be cool [07:32] ok, I'll keep my ears open [07:33] Mirv: for the unlucking ... WOW, if that works then I am saved :) [07:33] bzoltan: for a), it's nowadays apparently as simple as... adb shell "sudo dbus-send --system --print-reply --dest=org.freedesktop.Accounts /org/freedesktop/Accounts/User32011 org.freedesktop.DBus.Properties.Set string:com.canonical.unity.AccountsService string:demo-edges variant:boolean:false" [07:33] ;) [07:33] (from http://bazaar.launchpad.net/~ubuntu-test-case-dev/ubuntu-test-cases/touch/revision/287) [07:36] bzoltan, hello - do you have results from you tests for silo 12 landing? [07:38] brendand: that is the 28.08 lanidng for RTM, right? [07:40] bzoltan, https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu-rtm/landing-012 [07:45] brendand: geez, that is an archaeological finding :) I have flipped the tested switch last week. I still have the logs. Do you need me to check something? [07:46] bzoltan, no i just want to have the logs for comparison === tvoss is now known as tvoss|test === tvoss|test is now known as tvoss [08:11] robru: "AttributeError: 'bool' object has no attribute 'isdigit'" https://ci-train.ubuntu.com/job/ubuntu-rtm-landing-007-1-build/25/console [08:12] Mirv: Do you know what does this mean: $ adb shell "gdbus call --session --dest com.canonical.UnityGreeter --object-path / --method com.canonical.UnityGreeter.HideGreeter && echo Greeter unlocked" [08:12] Error connecting: Cannot autolaunch D-Bus without X11 $DISPLAY [08:14] bzoltan, that it can't connect to the session bus I guess [08:14] bzoltan, try to adb shell; su - phablet; [08:15] bzoltan, running as root? [08:15] seb128: tvoss: with new adbd all my commands are run as phablet. but I can try to force [08:15] bzoltan, thats a bug with cgroups that mterry works on since a few days [08:15] bzoltan: I wonder if that one hasn't been updated to the new world similar to the provisioning [08:16] bzoltan, could be that the env is wrong/doesn't include the dbus session [08:16] the env is definitely wrong, you need the sudo wrapper, but even then it wont work [08:17] logind doesnbt hand out proper sessions due to the cgroups issue [08:17] ogra_, hasn't the cgroup bug fixed by desrt/slangasek in systemd-s on friday? [08:17] "debian/patches/0001-cgmanager-stop-doing-async-calls.patch: [08:17] cgmanager: stop doing async calls; fixes a race condition on login. [08:17] Thanks to Ryan Lortie . Closes LP: #1365095." [08:17] Launchpad bug 1365095 in systemd-shim (Ubuntu-rtm 14.09) "Greeter not asking for pin code in image 11 (krillin)" [Undecided,New] https://launchpad.net/bugs/1365095 [08:17] dunno, i have seen mterry and hallyn discuss into the (european) nige [08:17] k [08:17] I though that was ^ [08:18] *nite [08:18] but maybe there is another issue still [08:18] well, you definitely still need the sudo wrapper [08:18] dbarth: the syncing seems broken in yet new other way, I'll help your silo manually in a bit [08:18] adb shell doesnt have the full env [08:18] Mirv: seb128: ogra_: when I manually adb shell in and execute the command then the screen get unlocked [08:18] bzoltan, what ogra_ said [08:19] seb128: I just have read... so ogra_, what is the trick? [08:19] (i have been looking into that over the weekend, might be i can fix it in adbd) [08:19] bzoltan, wrap the gdbus command in "sudo -u phablet -i" as usual [08:19] ogra_: OK, I try that [08:21] (apologies for the spam, I edited the description) [08:21] ogra_: works now, thank you [08:27] Morning all [08:29] dbarth, ^ no no no for that rtm settings silo [08:29] dbarth, we have settings in sile 015 and are trying to land that to rtm for a week, don't hijack it with your version [08:30] Mirv, ^ [08:30] seb128: noted [08:30] https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu-rtm/landing-015/+packages [08:30] get that to land first please [08:30] that should be good, just need qa verification [08:31] will do that [08:31] thanks [08:31] should we clear 007 out meanwhile to avoid errorS? [08:31] it's going to need to be rebased on top of what is 015 anyway [08:31] yeah, I'll clear it too, the current build has no use [08:32] or well, is that a sync of current utopic u-s-s? [08:32] because then that would be fine, but it would required a new round of testing, so we should still land 015 which has been tested already [08:32] ogra_: meeting? [08:34] Mirv, i wish i could, firefox doesnt let me :P [08:36] seb128, hey - silo 15 seems to have issues [08:36] brendand, shrug [08:36] brendand, which ones? [08:36] seb128, wizard crashes, and sim cards can no longer be unlocked [08:36] at least. still more testing to do [08:36] brendand, settings have nothing to do with sim unlock [08:37] are you supposed you tested settings only? [08:37] (running with watch_only since it had wrong status) [08:37] seb128, i have only upgraded the system-settings packages [08:37] brendand, the way we work is a bit weird btw, that update fixes at least 15 bugs and some important rtm ones, and we block on those fix on one potential issue [08:38] block those fixes on* [08:38] brendand, that "can't unlock sim" looks like it could be the systemd issue fixed in friday [08:39] brendand, anyway, as said, settings don't do sim unlock, so if you found a bug it's not with settings [08:39] brendand, it's more likely a random one which is already there in the image [08:39] seb128, it works before installing the packages [08:39] brendand, that shouldn't count as a settings blocker [08:39] you got lucky [08:39] seb128, several times [08:39] it might work on another try who knows [08:40] you got lucky several times [08:40] what I can tell you [08:40] seb128, alternatively - you're wrong [08:40] brendand, settings don't have sim unlock code, they are not involved in that [08:40] shrug [08:40] k, I give up, do whatever you want with settings [09:03] actually we can't do any ubuntu-system-settings rtm landing because any new sync from utopic silo would include the landing-rtm-015 silo changes anyhow, so any QA signoff will depend on the rtm-015 changes being testably regression free. [09:04] dbarth: ^ so your location settings landing might take some time to reach rtm until the previous u-s-s rtm landing is agreed by QA [09:05] Mirv, brendand has issues with settings anyway, so it looks like it's not ready to land [09:05] brendand, btw having a bug report of those specifics issue would help [09:05] seb128, i'm just going to confirm (again) what's going on [09:10] Mirv: ok, noted [09:17] thanks for the allocation :) [09:19] pete-woods: you're welcome :) we just had 0 silos free so I didn't do that earlier. [09:24] Mirv: have you seen weird build failures like this before? https://launchpadlibrarian.net/184302130/buildlog_ubuntu-utopic-amd64.libusermetrics_1.1.1%2B14.10.20140908-0ubuntu1_FAILEDTOBUILD.txt.gz [09:25] my diff for this change is totally insignificant (just adds the X-Ubuntu-Use-Langpack: yes line) [09:27] pete-woods: ogra_: that problem pete is seeing is because of that not-done-by-me qtxmlpatterns upload. we probably need to revert it. [09:27] Mirv, ouch ... [09:29] pete-woods: thanks for letting us know, we just noticed in the meeting that a new qtxmlpatterns had been uploaded by mitya57. I assumed it was ok since we hadn't had error reports, but now we did [09:30] nothing wrong with the new upstream version otherwise, but Qt is just very strict about numbers... [09:30] no problem :) [09:30] I pinged him on #ubuntu-devel now, there's either a revert option or faking the module version option [09:31] well I'll just be happy to build my package :) will follow along to see when it's resolved [09:33] seb128, so the bug is reproducible by installing the silo and then resetting the device [09:34] seb128, on utopic it's not reproducible - at least not on mako [09:35] brendand, can you describe "the bug" [09:35] seb128, i'll file it [09:35] thanks [09:37] Morning - can someone help me get upstart 1.13.2-0ubuntu1 into the rtm branch? [09:37] seb128, until then, when you click on the unlock button, indicator-network just seems to reload itself [09:39] brendand, k, your are speaking about the SIM unlock on boot? [09:40] brendand, that doesn't make much sense to me, settings are not running at this point and none of their code is used by unity8, the code involved should be unity8 and indicator only [09:40] brendand, I hope somebody figure it out though [09:40] seb128, it doesn't seem to happen if the wizard doesn't run, so maybe that leaves something in a bad state [09:43] brendand, that would be and is more likely than having settings involved [09:53] Mirv: Hi - could you add a "sync: ubuntu,utopic,upstart" to the CI spreadsheet? I don't have write access and we need the latest utopic upstart in Touch for the "UAL with CGroups" landing (line 4 in the s'sheet) [10:01] jodh: sure, but add where exactly? line 4 is the utopic landing [10:02] jodh: or do you mean add a new upstart rtm landing for syncing from utopic? [10:02] yes, I think you mean that [10:02] Mirv: yes, a sync of the latest upstart in utopic which is a pre-requisite for line 4 actually working :) [10:04] bzoltan, with my latest android-tools upload the sudo wrapping shouldnt be necessary anymore (once it landed in all images) [10:04] (it wont do any harm either to keep it though) [10:05] jodh: ok it will shortly start building at https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu-rtm/landing-007/+packages . line 63 on the spreadsheet. [10:05] Mirv: thanks very much! [10:11] Saviq, are you landing https://code.launchpad.net/~unity-team/unity8/new-adbd/+merge/233684 somewhere ? [10:11] ogra_, yeah, silo 19 [10:11] ogra_, but need to fix a sh$tload of ap tests that assumed there's no password [10:11] Saviq, see above ... adbd should be fixed now ... while it wont do harm it also shouldnt be absolutely necessary anymore [10:11] Mirv: erk, looking [10:12] ogra_, yeah, k [10:12] Saviq, (might make sense to verify it though ... but i think it shoudl "just work") [10:12] ogra_, kk [10:19] Mirv: looks like you freed the offending silo. I'm happy to blame sil for this, amazingly this bug is not in a part of the code that I've been hacking on ;-) if you see that again, try freeing the silo and then reassigning it, sil made some changes to the silo config format, silos that were assigned prior to friday may be incompatible with code that sil [10:19] landed on friday. [10:24] ogra@styx:~/Devel/branches/phablet-tools$ adb shell "gdbus call --session --dest com.canonical.UnityGreeter --object-path / --method com.canonical.UnityGreeter.HideGreeter && echo Greeter unlocked" [10:24] () [10:24] Greeter unlocked [10:24] ogra@styx:~/Devel/branches/phablet-tools$ [10:24] yay !!! [10:26] GGRRR did anyone elses krillin touchscreen get super sensitive in the recent images [10:26] i cant even scroll in news weballs without having it open unwanted links all the time [10:26] *webapps [10:28] this is unusable ! [10:28] ogra_: ooh, i noticed that at teh weekend. have you got a bug for it? [10:28] funnily there is no issue in the G+ app ... but in all news apps i tried [10:28] yeah, same [10:29] popey, nope, only noticed it on sat and thought my display was to dirty or something [10:29] heh, ok, I'll confirm if you bug it. [10:29] guessing oxide? [10:29] but it persists (and feels like it got worse, though probably i'm just easier to annoy now) [10:29] popey, we didnt have oxide updates in rtm [10:29] i think ... [10:29] * ogra_ checks [10:30] i assume its the device tarball [10:30] http://people.canonical.com/~ogra/touch-image-stats/rtm/21.changes thats all that landed on the weekend [10:36] seb128, here's a twist [10:36] seb128, i can reproduce it in utopic without the silo installed... but only on krillin :/ [10:37] brendand, utopic already has the changes though, no need of a silo [10:40] robru: thanks for looking! a sync specifying a package worked, but then we postponed the whole landing. good to know, that workaround, makes sense that there was a format change. [10:43] seb128, well yes - i suppose that statement is a bit redundant [10:43] seb128, but i've tested the same thing on both krillin and mako with utopic and it only fails on krillin [10:43] brendand, k, fun [10:45] yay [10:45] gra@styx:~/Devel/branches/phablet-tools$ adb shell "gdbus call --session --dest com.canonical.UnityGreeter --object-path / --method com.canonical.UnityGreeter.HideGreeter && echo Greeter unlocked" [10:45] () [10:45] Greeter unlocked [10:45] ogra@styx:~/Devel/branches/phablet-tools$ [10:45] works on rtm fine as well :) === MacSlow is now known as MacSlow|lunch [11:00] cihelp, this jobs looks stuck, 'innit http://s-jenkins.ubuntu-ci:8080/job/generic-deb-autopilot-runner-mako/4375/console ? [11:00] so do the next two ones... /me worried that's what plars reported about devices getting stuck :| [11:01] what devices are these ? makos ? [11:03] (on krillin there was a udev rule from the device tarball that broke a shipped rule for adbd, which made the devices come up as offline, fixed on sat. devices that booted with the broken image might go stuck ... this is krillin only though) [11:07] Saviq: ogra_: these devices are not offline. could be some other issue [11:07] yeah, not that then [11:08] its only krillin utopic anyway, mako and rtm should be fine [11:12] Mirv: for some reason RTM silo 008 says it doesn't need QA sign off. Yes it does and looking at the description there is more than one fix so it is all lies. Also the telephony service isn't building [11:14] ogra_: have you guys hit this bug with the CI testing -> https://bugs.launchpad.net/gallery-app/+bug/1363190 It is a silent killer. You will not even notice that the flashing fails... [11:14] Ubuntu bug 1363190 in gallery-app "Gallery APP autopilot tests pollutes the file system" [Undecided,New] [11:21] Saviq: ogra_: so http://s-jenkins.ubuntu-ci:8080/job/generic-deb-autopilot-runner-mako/4375/console is not stuck, but it took a bit of time at one stage. they are progressing now [11:21] yay [11:21] good to know [11:22] psivaa, oh ok thanks [11:22] oof [11:27] davmor2: bfiller has written there ": these are isolated bug fixes don't need QA signoff". and I've written that telephony-service failed to build [11:27] probably because of some media related landing missing from rtm [11:32] Mirv: my concern is that we were told all rtm silos needed QA approval and now we see one that says it doesn't need it, so now we are wondering how many of these things are getting through [11:33] davmor2: the policy was that isolated bug fixes can go in without QA approval, but that 90% of landings would need the approval. better defining what's an isolated bug fix would help. [11:34] Mirv: okay I'll chase it up with jfunk latter [11:34] davmor2: see line 58/59 for another example. I think I haven't seen others lately. [11:39] trainguards, I can get silo for line 29 please [11:47] Saviq: with pleasure. how's the utopic silo testing looking? [11:48] Mirv, had to fix a lot of autopilot because of the adb change (we didn't support pinlocked tests until now) [11:48] Mirv, so that's building now [11:48] or built actualy [11:48] +l [11:48] so am testing now [11:51] Saviq: ok, cool === MacSlow|lunch is now known as MacSlow === iahmad_ is now known as iahmad === om26er is now known as om26er|afk [12:22] Mirv, do you have anything pending for a new image ? i would like to trigger a build (that should hopefully help with screen unlocking in the lab) === om26er|afk is now known as om26er [12:26] ogra_: feel free to fire, I'm not holding my breathe on anything specifically [12:26] heh, ok [12:26] triggered [12:29] Mirv: see -touch for your help [12:30] mterry: so i think we might wnt to move the systemd-shimd source packag efrom utopic in the system-setting silo? [12:30] mterry: do we also need li8ghtdm? [12:30] asac, no [12:31] mterry: no for putting it into the same silo? [12:31] think would help if it really fixes the problem that the system-settings silo bounced === iahmad is now known as iahmad|afk [12:31] woudl be good if someone who has the issue could confirm that first though [12:32] (that systemd-shim actually fixes the SIM unlocking) [12:32] right [12:32] davmor2 is at lunch [12:32] but given that mterry fixed the same issue [12:32] i am sure it does [12:32] before you pile up more stuff in that poor silo [12:32] https://bugs.launchpad.net/ubuntu/+source/systemd-shim/+bug/1365095 [12:32] Ubuntu bug 1365095 in systemd-shim (Ubuntu-rtm 14.09) "Greeter not asking for pin code in image 11 (krillin)" [Undecided,New] [12:32] ogra_: that silo currently onlyhas system-settings [12:32] asac, with a bunch of MPs though [12:32] asac, well... I am not sure that they are the same root problem yet [12:33] right, but if those depend on this one then they should go in together to avoid duping QA [12:33] mterry: ok. let us know [12:33] Mirv is chcecking whty the shim silo didnt ubild [12:33] Mirv: uscan warning: In watchfile /tmp/tmp6IdJ2a, reading webpage http://people.gnome.org/~desrt/ failed: 500 Can't connect to people.gnome.org:80 (timeout) [12:34] hmm. it cant find the orig it seem [12:34] yeah, that cant work [12:34] s [12:34] guess it should come from ubuntu archive [12:34] mabye that thing was stuck in proposed?: [12:34] no [12:34] i saw it enter the utopic image [12:34] ogra_: when? before the silo failed? [12:34] must be in the utopic archive proper [12:34] asac: if you say -touch, then I discuss at -touch :) I copied it now manually. [12:34] asac, yes [12:34] Mirv: thanks [12:34] === trainguards: IMAGE 233 building (started: 20140908 12:35) === [12:35] Mirv: yeah, i just thought as we continued we might wanna go back here. sorry [12:35] for me its one tab away though :P [12:35] hehe [12:35] right next door [12:35] asac, 7-3 was an upload, didnt go though silos in utopic [12:35] (systemd-shim 7-3 that is ) [12:35] ogra_: sure, but could have been stuck in proposed when wee tried the silo for rtm [12:35] https://ci-train.ubuntu.com/job/ubuntu-rtm-landing-016-1-build/6/console [12:35] ogra_: it was siloed for rtm after [12:36] but anyway, we got it in now :) [12:36] asac, silo buildds behave slightly different to distro builders ... [12:36] * ogra_ got bitten by that on the weekend with android-tools too [12:37] ogra_: its not in the buildds [12:37] its before... check the log [12:37] but doesnt matter now :) [12:37] * asac still believes that the silo build was attempted when it was still in proposed and the jenkins doesnt take source origs from proposed [12:39] is the theory here that the silo might have built against a package from -proposed? [12:40] asac's theory is, yeah [12:40] that would be anomalous as silos are not normally configured to build against -proposed [12:40] i think the copy-package would have copied the orig.tar.gz in any case [12:41] copy-package does not concern itself with such details :) but the publisher would republish that when it runs [12:41] if bzr-builddeb is relying on the .orig.tar.gz being in the archive already, then it may have to wait a publisher cycle [12:42] no ... i said that the silo upload through the jenkins job might have been tried when the package wasnt in proper archive yet [12:42] hence we get the weird "cannot download orig tarball" failyure [12:42] https://ci-train.ubuntu.com/job/ubuntu-rtm-landing-016-1-build/6/console [12:42] but *shrug* [12:42] IMO the problem there is that the version is wrong [12:43] asac, the "silo upload" is a manual script call ... [12:43] This isn't a package that has its upstream maintained by ci-train, so surely it should be 7-3~rtm or similar, not 7~rtm-3 [12:43] which essentially does something similar to a dput ... which in turn i would expect to have uploaded the orig tarball [12:43] * asac desnt know and leaves it to the smart folks [12:43] I seem to remember somebody mentioning that bug/misfeature/whatever here recently [12:44] yeah, i think lukasz was looking into that on friday [12:44] we could just do a source copy instead though, right? [12:44] yeah i see the version now [12:44] defiitely bogus [12:44] unless this is a cherry-pick [12:44] cjwatson: Mirv alreadu uploaded manually to the silo now [12:44] ok [12:45] i think the ci train source copy feature is what failed here initially [12:45] but guess the version mangling is really bogus still there [12:45] if it's doing version mangling it's not a copy ... :) [12:45] yeah [12:45] so i am not sure :) [12:45] have to admit I haven't used the sync feature yet; for ubuntu/landing-003 / ubuntu-rtm/landing-003 I just used copy-package since I could [12:45] I did copy-package now [12:45] hard to remember who did what there ... thought they used the sync: feature [12:46] Mirv: will the silo status figure that it got fixed? [12:47] i mean here: http://people.canonical.com/~platform/citrain_dashboard/#?distro=ubuntu-rtm&q= [12:47] if it was a dput it will ... with the next watch_only build [12:48] (which you can only do after the packages have built ... be patient :) ) [12:48] heh [12:48] you can do it a bit before and it'll wait [12:50] asac: yes, it just requires watch_only build that can't be done right away. it's now running. [12:50] cool. thx [12:51] * ogra_ thinks the last android-tools fix is the grossest hack he has ever done ... bt it makes things work [12:51] Sep 8 14:50:32 ubuntu-phablet sudo: phablet : TTY=pts/38 ; PWD=/home/phablet ; USER=phablet ; COMMAND=/bin/bash -c /bin/bash -cl env [12:51] (thats the log output of "adb shell env" [12:51] ) [12:52] (see the COMMAND) ... [13:17] rsalveti: hi, is the QMediaPlayer's audio role stuff already in RTM (or close to land at least)? telephony-service is failing to build on the RTM silo [13:17] rsalveti: https://launchpadlibrarian.net/184207757/buildlog_ubuntu-rtm-14.09-amd64.telephony-service_0.1%2B14.10.20140905~rtm-0ubuntu1_FAILEDTOBUILD.txt.gz [13:21] boiko, it's in silo 11. I'm testing it. [13:21] jibel: ah, nice! [13:22] hi. I have a new device tarball I'd be interested in publishing today, if possible [13:23] is someone able to give it a QA pass? [13:23] Mirv: would it work if after rsalveti's silo 11 is landed, I trigger a rebuild of telephony-service? I didn't know there was code in telephony-service not landed on RTM yet [13:23] ogra_, ^ (as person running the landing meetings...) [13:23] john-mcaleely, perhaps davmor2 [13:24] depends how swamped he is ... looks like he is alone [13:24] davmor2, something you could do today? (brendand did the honours on friday) [13:24] ogra_, understood [13:25] boiko: yeah, would work after a rebuild [13:25] boiko: didn't yet land because silo 11 is still to be tested [13:25] john-mcaleely: Yeap I can do that, currently looking at a silo so I could do with it not conflicting with that. Any idea on when you would like to land it [13:25] from my experience it takes a few days for the rtm stuff to be signed [13:26] davmor2, it is worth doing, if I can land it before cob (~6pm) UK today [13:26] davmor2, otherwise, it can wait [13:26] rsalveti: ok, I'll just leave my silo sitting there waiting, would you mind giving me a heads up once silo 11 lands? [13:27] boiko: sure [13:27] rsalveti: thanks [13:29] davmor2, http://people.canonical.com/~jhm/barajas/device_krillin-20140908-d8c11f3.tar.xz [13:29] davmor2, simply download, and use --device-tarball with your favourite ubuntu-device-flash command to test [13:29] john-mcaleely: no worries I'll give it a go asap [13:29] davmor2, I'll ping you in a bit and see if I get lucky on your capacity :-) [13:32] Mirv: hey [13:33] Mirv: I'm seeing mtp as landed in the spreadsheet, and no longer in any silo, but apparently not in ubuntu-rtm [13:38] ogra_: Mirv: how are the landings for the rtm proposed blockers going? [13:38] i think all but one had an MP according to sils mail on friday [13:39] let me check his list [13:40] ogra_: its the four bugs i cut and forwarded to ue-leads with you CCed [13:40] as you were appopinted the driver [13:40] nobody told me :P [13:41] ** Sometimes input breaks and only edges are responsive [13:41] https://bugs.launchpad.net/bugs/1295623 => Merge requests present -> It should land soon, it's next in the unity8 queue. [13:41] Ubuntu bug 1295623 in unity8 (Ubuntu) "Sometimes input breaks and only edges are responsive" [Critical,In progress] [13:41] ** Seekbar tests are skiped only on nexus 4 and 5 [13:41] ogra_: it was in the mail by sil [13:41] https://bugs.launchpad.net/bugs/1359040 => Merge requests present -> Is this really a blocker? Comment needed. [13:41] Ubuntu bug 1359040 in mediaplayer-app "Seekbar tests are skiped only on nexus 4 and 5" [High,In progress] [13:41] asac, yes, i see it [13:41] ogra_: so please ensure we dont pick up too risky landings while we try to get that [13:41] e.g. push back on crazy folks :) [13:41] the seekbar stuff is really ignorable [13:42] Saviq: when is landing of bug 1295623 scheduled? [13:42] bug 1295623 in unity8 (Ubuntu) "Sometimes input breaks and only edges are responsive" [Critical,In progress] https://launchpad.net/bugs/1295623 [13:42] but yeah, fix wouldnt hurt (four tests that should be skipped arent ... we tend to simply ignore their results ... purely cosmetic) [13:42] asac, am testing now [13:42] this one is definitely important [13:42] asac, unless you mean the bug itself, then it landed around the time we started doing the phone ;) [13:43] 15:40 < ogra_> the seekbar stuff is really ignorable [13:43] 15:41 < asac> Saviq: when is landing of bug 1295623 scheduled? [13:43] bug 1295623 in unity8 (Ubuntu) "Sometimes input breaks and only edges are responsive" [Critical,In progress] https://launchpad.net/bugs/1295623 [13:43] 15:41 < asac> ogra_: wll, we have an MP for that i read [13:43] 15:41 < asac> so why not get it in and get it off our worry list [13:43] asac, no idea why it is on that at all [13:44] rsalveti, I tested silo 11, ran the testplan of pulseaudio and tried music player/mediaplayer/webbrowser with alarms and alert (incoming calls, incoming messages). All works fine, is there anything else to verify? [13:45] ogra_: wll, we have an MP for that i read [13:45] so why not get it in and get it off our worry list [13:45] jibel: nops, that's it [13:45] 15:44 < asac> ogra_: well, sure that QA didnt like it and wants to get this resurrected [13:45] ogra_: well, sure that QA didnt like it and wants to get this resurrected [13:45] damn flaky connection [13:45] oops :) [13:46] * asac doesnt understand when irssi reposts stuff after disconnect and when not [13:46] asac, answered your mail [13:46] i didnt get tthat i was explicitly CCed ... evo just showed it in the thread [13:46] ogra_: anyway, i think that one is wihtelistable if we have the unity part [13:48] asac, phew had some laptop troubles -- in case you are still wondering, looks like the systemd-shim fix solves the SIM PIN problem [13:48] robru, So I deleted the status on line 50, but I'm not sure if that was the right thing to do :-) It's fixed. [13:49] mterry, right, it simply didnt land in rtm yet [13:50] ogra_, sure... Though note that the bug can still be reproduced on the latest utopic images until the next spin [13:50] mterry, already spinning ;) [13:50] ogra_, and there was some question on whether the bugs were the same root cause [13:50] mterry, with an adbd fois for your un lock issue too btw ;) [13:50] *fix [13:50] ogra_, why don't we just land the unity8 branch? [13:51] mterry: nice one [13:51] ogra_, that's been approved for like a week [13:51] davmor2: can you do a joint tesating of 015 and 016? [13:51] davmor2: seems they going in together will fix the SIM PIN problem [13:51] davmor2: and that will then clear the path for location landing [13:51] thanks! [13:51] boiko: yes, but please ask someone with ppa rights to trigger the rebuilds in there directly [13:51] jibel: ^ [13:51]  [13:52] asac: right okay so I'll actually need 016 to finish building then I can [13:52] davmor2: wait ... finish? [13:52] mterry, it hasnt been approved [13:52] davmor2: its just dashboard brokenness [13:52] davmor2: https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu-rtm/landing-016/+packages [13:52] mterry, i just commented on it a few mins ago [13:52] they are buitl!! [13:52] Mirv: ah because the sync: thing is broken already, right? [13:52] Mirv: ^ seems the dashboard didnt pick up that we directly uplaoded [13:52] Mirv: ignore... seems it did [13:52] davmor2: so all seems ready [13:53] cyphermox: hmm, maybe some publishing went wrong because of the cu2d code refactoring [13:53] mterry, seems it was merged with a giant other change that isnt 100% ready [13:53] asac: ah package built now yeap no worries [13:53] Mirv: should I just do the copy? [13:53] boiko: yes, the sync is no more there [13:54] Mirv: ok, got it, thanks [13:54] mterry: Saviq: maybe we cna cherry land this input fix? [13:54] asac: regarding blockers, the big qtmir/unity8 landing is still being prepared by Saviq. mediaplayer-app blocker fix is approved to land. [13:54] anyway, will not micro manage those :) ... lets hope we get the input landed [13:55] Mirv: cool. you think qtmir/unity8 is realisitc today/early-tomorrow? [13:55] ogra_, woah... someone took it and made lp:~unity-team/unity8/new-adbd which is enormous... [13:55] asac, I replied already [13:55] asac, am testing now [13:55] ah [13:55] thanks [13:55] yeah had reconnects here [13:55] sorry [13:55] rsalveti, music doesn't stop when music player is closed, but I think that's an existing bug. [13:56] jibel: yeah [13:56] mterry, yeah, what we forgot about is the adb change also made it that there's always a password.. [13:56] cyphermox: it'd probably need qa signoff so another silo. can you sort it out with robru? I need to flee [13:56] ogra_, you just went ahead and added an automatic "sudo -u phablet -i"?! I asked for that and you said no :-( [13:56] there was already a signoff [13:56] robru: ^^ [13:56] Saviq, I thought our autopilot ran with a lightdm mock though? [13:57] are there any other trainguards about before robert in 2h? I can't stay longer than this 10h today alas :( [13:57] mterry, remember what our lightdm mock is these days? ;P [13:57] Saviq, no I mean the real mocks, the ones we use for qmluitests and such [13:57] mterry, nope [13:58] mterry, we only did the LD_PRELOAD thing, which doesn't seem to work any more [13:58] mterry, if you can fix it better, feel free [13:58] Saviq, well -- depends on whether we feel running autopilot via mocked lightdm is a "real" test or not [13:59] Saviq, we'd need LD_PRELOAD and QML_PLUGIN_PATH [13:59] Saviq, for just the lightdm plugin [13:59] mterry, yeah, let's go with my approach for now (I used the dbus unlock call) [13:59] mterry, and if we decide we want it different is when we'll have a look at this again [13:59] mterry, i said i wouldnt know if i can get it to work (we tried multipe times before) [14:00] mterry, i got it to work now though [14:00] ogra_, \o/ [14:00] mterry, with a gross gross hack though :) [14:00] but it works at least [14:00] Saviq, gotcha, that's fine. I just had remembered us using the mocks for AP so I didn't even consider having to fix them past the unlock-device script [14:00] asac, Saviq: so it looks like the no keyboard on initial boot is in place still but more important on reboot you get no unity8 dash just a black screen with the loading animation. [14:01] davmor2, bug #1362619 [14:01] The keyboard on initial boot bug isn't fixed? /me scratches head [14:01] bug 1362619 in unity-scopes-api (Ubuntu RTM) "unity8-dash hangs in scopes backend" [Undecided,Confirmed] https://launchpad.net/bugs/1362619 [14:01] davmor2, and I don't think the keyboard fix from mterry landed [14:01] mterry: this is before the silo [14:01] ah [14:02] mterry, "adb shell env" output in auth.log : Sep 8 14:50:32 ubuntu-phablet sudo: phablet : TTY=pts/38 ; PWD=/home/phablet ; USER=phablet ; COMMAND=/bin/bash -c /bin/bash -cl env [14:02] mterry, like inception ;) shell in shell in sudo ... [14:02] ogra_, heh [14:02] dbarth: mp approvals... ^ [14:02] mterry: need a fresh install system was getting messy and I wanted clean results [14:02] mterry, TBH we shouldn't be using any mocks in ap tests if possible, the lockscreen tests should probably be moved to QML ones anyway [14:03] Mirv: yup, spotted :/ [14:03] Saviq, well in theory I agree, but handling unlocking via system password is tricky. Not sure the dbus unlock is much cleaner than a mock, but at least it's using real code [14:03] davmor2: in silo or in general? [14:03] davmor2: i had this problem of no dash loading on krillin this morning [14:03] but also a couple days ago [14:03] rebooting fixed it [14:04] asac: rebooting indeed does fix it [14:04] Saviq, it'd be nice if there was a clean way for the unity8 AP test to be passed the system password [14:04] *phablet password [14:04] davmor2: right. but i saw it for a week or so from time to time on krilin... never on n4 [14:04] mterry, well, we should have integration tests that actually test setting/unsetting the password, locking/unlocking the shell, but I don't think unlocking with pin/pass should happen every time... [14:04] i mean.. getting an animation like starting an app forever [14:05] * asac thought a few times it would be great if he could kill the dash just like an app :) [14:05] through the right edge animnation [14:05] Saviq, well yeah -- ideally we could set / unset the password via the AP test :) [14:05] mterry, which would not be a unity8 ap test, but a high-level platform one [14:06] mterry, but yeah, with the adb change right now there's no way to test unsetting the pass, as that will lock you out of the phone ;) [14:06] Saviq, eh, only high level because of the password thing. Everything else would likely be local to unity8. But sure, however we get it done [14:06] asac, "restart unity8-dash" in the terminal app works fine [14:06] mterry, unless you waited for the adb connection again, as the test really happens on the phone still [14:06] asac, generally it shouldnt hang or crash indeed :) [14:06] ogra_: yeah, but its tempting to just shoot it in teh right edge aninmation if it doesnt accept input anymore :) [14:06] ogra_: hehe... true [14:06] but then most apps shouldnt need to be killed either [14:07] but we allow it anyways [14:07] like phone, mesasging etc. [14:07] they hog memory [14:07] even if they dont run [14:07] not with lifecycle [14:07] sure they do [14:07] lifecycle with swap them out if there is not enough mem [14:07] a sigstopped app wont free its ram [14:07] it first is sigstopped [14:07] only once it gets OOM killed it will [14:07] then it gets serialized and killed [14:07] right [14:08] so in general there shouldnt be a need to kill them [14:08] only to fix dysfunctional apps [14:08] thats hte only valid use case we have in theory forl killing imo [14:08] not really [14:08] why not really? [14:08] i wouldnt want to have to scroll through the list of apps i have started since i got the new phone [14:09] ogra_: ok thats a user experiennce challenge then [14:09] could be sorted by least recently used [14:09] hehe [14:09] if you make them unremovable in the edge swipe things will pile up and the edge switpe will become less useful [14:09] kind of true i guess [14:09] think maybe after not using them for like 2 days they could get hidden [14:10] or only last 10 apps shown [14:10] hehe [14:10] that might actually work [14:11] until that arm64 phone with 4GB ram comes out and you want to have 20 to stay open :) [14:11] but guess the "kill because app is buggy" isstill a valid use case especially since we have random 3rd party apps in store [14:11] yeah [14:11] but that use case also exist for dash :) [14:11] hehe [14:11] which it shouldnt though [14:11] system stuff needs to be rock solid [14:11] well it shouldnt for any of our key apps [14:11] like phone and messaing [14:12] and shell :) [14:12] terminal? [14:12] lol [14:12] ah the shell [14:12] well, that needs more love first [14:12] indicators and friends [14:12] right [14:12] not terminal-app [14:12] currently close to unusable on krillin ... [14:13] * ogra_ goes back to hack on pam before the standup [14:14] * asac goes coffee [14:15] === trainguards: IMAGE 233 DONE (finished: 20140908 14:15) === [14:15] === changelog: http://people.canonical.com/~ogra/touch-image-stats/233.changes === [14:15] whee, exciting ... [14:16] rhuddie: hey [14:16] rhuddie: do you agree we can close https://bugs.launchpad.net/ubuntu/+source/bluez/+bug/1273629 ? [14:16] Ubuntu bug 1273629 in bluez (Ubuntu) "HSP fails on Ubuntu Touch [Bluetooth headset does not work]" [Critical,Fix committed] [14:17] cyphermox, Hi, yes, I think that one can be closed [14:17] rhuddie: thanks :) [14:17] seb128, mterry - do we know when/how systemd-shim is going to land in RTM? [14:18] cyphermox, I also checked on the Vivaldi speaker and updated that bug report [14:18] rhuddie: for the Altavoz Vivaldi speaker, would you be able to gather some extra information for me, assuming you have it [14:18] brendand, I don't [14:18] brendand: looking at it now [14:18] ahah, great minds think alike! [14:18] brendand, ogra_ and asac were looking at it above. I think very soon [14:18] cyphermox, sure thing [14:18] let me refresh, I didn't see your update I think === om26er is now known as om26er|doc [14:19] cyphermox, no prob. Let me know if you need any more info on it. I have a big box of devices sat here :) [14:19] rhuddie: indeed I missed the update, that's all; but the hcitool output doesn't contain what I expect [14:20] rhuddie: I'll get you a package to test [14:20] cyphermox, thanks [14:22] brendand: davmor2 is testing that together with the systemd-shim [14:22] er system-settings silo 015 ans 016 [14:24] mterry: seb128: are landing the wizard crash in utopic already? i think we will soon need to pull that into rtm [14:24] guess the 015 silo will actually have the same crash [14:24] asac, not sure if kenvandine plans a landing today, we can probably do one [14:24] kenvandine: can you do that? [14:25] at best now :)... we need that to land a crash free system-settings in rtm [14:25] for HERE and all the good fixes that are in ss [14:26] asac, been trying... it failed qa verification again... [14:26] the biggest issue is the wizard crashing, along with the location service [14:26] which i suspect was introduced with the HERE stuff [14:27] the wizard crashing didn't seem obvious to me, it looks like the wizard finishes... but it leaves crash files [14:27] kenvandine: wait :) [14:27] for the location service and ubuntu-system-settings-wizard [14:27] kenvandine: talking about utopic [14:27] kenvandine: the rtm landing we have now understood... it was missing shim [14:27] oh... the HERE stuff is in utopic-proposed [14:27] and also we need this crash there [14:27] kenvandine, https://code.launchpad.net/~mterry/ubuntu-system-settings/wizard-crash/+merge/233589 [14:27] kenvandine, that should fix the crash in the wizard [14:27] great [14:27] right [14:27] so we need to land that :) [14:27] so once we have that in we can also add that to system-settings [14:27] and HERE silo [14:28] seb128, brendand also said he had trouble with the cellular panel [14:28] kenvandine: exactly :) [14:28] which didn't change... and it works fine for me :) [14:28] kenvandine, I've 2 trivial changes waiting review if you want to include that as well [14:28] kenvandine: isnt that the SIM PIN problem? [14:28] hats the systemd-shim that is in 016 [14:28] asac, yes... but it worked for me :) [14:28] kenvandine, there is a SIM unlock issues, but mterry thinks it's the systemd issue as asac just said [14:28] kenvandine: it worked on utopic because systemd-shim is in there [14:28] ok... i thought that too [14:28] but not in rtm [14:28] but i tried it on rtm 30 minutes after brendand failed it and it all worked [14:29] mterry confirmed that the systemd-shim fixes taht in rtm (which bounced you rlanding) [14:29] woot [14:29] ;-) [14:29] kenvandine: do you have a locked SIM? [14:29] you need that to reproduce [14:29] kenvandine, did you try on mako or krillin? [14:29] that bug is random one [14:29] asac, no... but i know the pin for my ATT sim [14:29] it's not consistent [14:29] and can unlock/lock/changepin [14:29] brendand, both... [14:29] if its not locked it isnt showing the problem according to bug [14:29] locked aka you need to enter sim pin to make it work [14:30] asac, but what if i lock it? [14:30] that worked [14:30] kenvandine: not sure what that means [14:30] you need to have it in locked state [14:30] then wipe user data [14:30] asac, in settings, you can lock the pin [14:30] and boot [14:30] on the sim [14:30] kenvandine, and you have to be on krillin I believe [14:30] maybe that too [14:30] * asac doesnt know [14:30] anyway, we think we have that under control. just need the crash fix in utopic [14:30] so we can clear it all into rtm [14:31] cool [14:31] kenvandine, i saw other problems in the cellular panel, but the silo is in davmor2's hands now so i trust that if what i saw was a real bug he'll find it [14:31] i desparately want to land rtm :) [14:31] kenvandine, so if davmor2 passes it you're ok [14:31] kenvandine: thanks. let us know when its in utopic later today so w can finish the rest [14:31] brendand, weird... all the cellular panel features worked for me... and the landing to rtm doesn't include any changes to cellular [14:31] asac, will do [14:33] fginther, any thoughts on what might be wrong with this job? s-jenkins.ubuntu-ci:8080/job/ubuntu-calculator-app-autolanding/35/ I thought it was something within jenkins, but s-jenkins.ubuntu-ci:8080/job/ubuntu-calculator-app-autolanding/34 seemed to run fine . . . [14:34] balloons, I'll be able to take a peek in about an hour [14:37] ty [14:38] whoops [14:38] asac, seb128: no available silos :) [14:42] thostr_, pmcgowan: can we free silo 15 (apneditor)? It's not ready to land and I need a silo to land the settings fix that we hope will clear the way for an rtm landing [14:42] kenvandine, i vote yes, Wellark is working with mpt on agreeing design [14:43] it's outdated anyway :) [14:55] Mirv: ^^ can you clear silo if wellark is fine? === plars changed the topic of #ubuntu-ci-eng to: Train support: trainguards | Vanguard: plars | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: Ping robru if any citrain jenkins jobs have unexpected results. [15:01] trainguards: hi guys. could someone re-trigger the builds in https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-017/+packages [15:01] the broken dependency seems to have made it into distro now :) [15:07] pete-woods: done [15:13] asac: I thought Mirv had left, I can clear silos if confirm the silo? [15:17] jibel: could ubuntu-rtm/landing-003 (spreadsheet line 46) make its way onto the QA trello board, please? the important change is re-enabling signature checking - so making sure that ordinary app installation from the store still works, and also that sideloading apps onto a device from the SDK still works (which broke the last time we did this, and we believe we've fixed). note that the latter requires the qtcreator-plugin-ubuntu ... [15:17] ... from ubuntu/landing-003, which is landing in utopic at the moment [15:18] jibel: this is moderately urgent as (a) it's one of the current blockers in landing team mails, (b) ubuntu/landing-003 was published a little prematurely and these two were meant to be in sync [15:18] pmcgowan, kenvandine, asac ? [15:22] Wellark, i freed silo 15 with the apneditor [15:23] cjwatson: thanks! [15:24] kenvandine: ack. [15:24] kenvandine: I will ping thostr_ when I've done the fixes from the design review. [15:24] cjwatson, done [15:24] Wellark, thx [15:24] kenvandine: need to revert the OptionSelectors :( [15:24] they look way better than the ListItem.ItemSelector [15:25] jibel: thanks [15:25] but it's better to suck consistently ;) [15:25] davmor2, how was my luck at a QA pass? will we be discussing results when the landing team meet in half an hour? [15:25] Wellark, consistency is best :) [15:26] john-mcaleely: having issue fighting sim pin locks on rtm currently [15:26] Wellark, i wish the OptionSelector had a better style [15:26] davmor2, ok, thanks :-) [15:26] davmor2, oh? in system-settings? [15:27] oh right.. we think that was because of the systemd-shim problem [15:27] kenvandine: Yeah I have the systemd-shim installed I think let me double check that and try again [15:27] kenvandine: the silo with the fix in I should say [15:29] davmor2, oh... rtm silo 1 is interesting [15:30] davmor2, looks like a fix for a regression in ofono that prevented entering pin [15:31] kenvandine: the other issue is that sim unlock constantly tells you there are 3 attapts left [15:31] davmor2, is that in system-settings? [15:31] or the shell? [15:31] kenvandine: everywhere [15:31] ok, that could be related to the fix in silo 1 [15:31] kenvandine: leading me to lock my phone [15:32] which i never even saw that regression [15:32] sim even [15:32] ugh [15:34] asac: are you around? [15:37] asac: I wonder if it would be possible to set up group highlights for this channel. For example we define and "sdk tools dependency" group with the "click, android tools, emulator, phablet-tools, QtCreator" and the CI train would ping the sdk-tools-deps when any of the dependencies put in a silo for landing. [15:37] pmcgowan: ^ [15:41] balloons, I'm a little confused by that one too [15:42] fginther, so the merge redoes the tests, but afaict the autolanding never gets past setup right? [15:42] bzoltan: hm, does your IRC client not have suitable highlighting facilities? [15:43] balloons, right, the setup failed during phablet-config writable-image [15:43] /hilight -channels #ubuntu-ci-eng -regexp Silos:.*click [15:43] or whatever [15:43] balloons, which it had successfully executed a few commands earlier [15:43] and is working in the next 2 test runs [15:43] cjwatson: I have :) But I would love to see something formal in place. So to make sure that certain information reaches the defined parties. [15:43] I dunno, I'm a big fan of decentralisation for this kind of thing :) [15:44] cjwatson: as the dude who's butt is kicked when something goes wrong with the SDK tools I do love centralization :D [15:44] fginther, so is it possible something in the merge is causing it? Seemingly the answer is no, but since they work on other mp's... [15:46] cjwatson: My dream would be actually to have an SDK signature flag in the CI sheet for all projects what can cause regression in the SDK. The same way I could imagine that all projects with dependencies would benefit from such policy. [15:46] balloons, no no no, this is during the generic setup of the device. Completely unrelated to the MP. It's possible there are some lingering setup and timing issues that need to be resolved after the adb user change [15:47] fginther, it's just odd that it worked for the others. That's actually the third try for that mp [15:47] autolanding 32 and 33 are also autoland attempts [15:47] bzoltan: I think what I'm saying is that you're in a better place to know what you need to pay attention to from your experience, and it seems like an odd choice to want to put that information somewhere where it's harder for you to update [15:48] regressions> this is what autopkgtests are meant to be for :) [15:48] cjwatson: what I am thinking about is an information pushing instead of pulling. [15:48] they're exactly meant as a way to assert things about the behaviour of your dependencies [15:49] cjwatson: can autopkgtests test emulator creation, deployment, chroot creating and usage, app launcher features from the QtC? [15:49] all this business with highlighting people to do manual testing isn't very scalable [15:49] it's all software, I don't see why not [15:50] (though nested VMs may become an issue) [15:50] cjwatson: yeps... exactly [15:50] all I'm saying is that figuring out how to ask more people to do manual testing on things isn't a very forward-looking approach [15:51] cjwatson: With that one I agree. I am not suggesting more manual tests. [15:52] cjwatson: I am thinking about how to make sure that projects get informed about the upcoming changes of their dependencies. [15:53] bzoltan: might be something to ask the CI Airline people about, rather than bodging more stuff into the spreadsheet [15:53] some kind of subscription approach [15:53] cjwatson: might sound childish, but in my past it proved to be working. Like making mandatory to walk to the owner of a project what depends on your project and have a handshake. Not more.. [15:54] fginther, so I guess I can just leave it in your hands then and you can get it passing? [15:54] cjwatson: I know the CI sheet is heavy already. [15:55] balloons, yes, there is one thing that I can tweak and then see how it goes after that. Am I ok to re-approve that MP? [15:55] fginther, yes [15:55] bzoltan: I suspect those were smaller projects than Ubuntu though :) [15:56] cjwatson: That is how Maemo/Meego used to work. That was not very small :) [15:56] bzoltan: considerably smaller than Ubuntu [15:57] certainly in terms of the platform's dependency graph [15:57] cjwatson: from that point, no doubt [15:57] meego cloud, meego server and meego desktop were surely rare occurences ;) [15:57] bzoltan: i would suggest to wait till sil is back and see if there are aliases to also include in the pings [15:58] cjwatson: the point is that making dependency lines to meet and shake hands at agreed milestones is something what i have seen working and it was improving stability [15:59] ogra_: by then Qt was an internal project :) [15:59] bzoltan: yeah, I'm just familiar with the dependency graphs above some of the things I maintain, and I fear having to talk to 100 people any time I change anything :P [16:00] so anything like that probably needs to be non-blocking most of the time [16:01] cjwatson: that is why I am suggesting an automatic step, if that is possible. If not, then just ignore me :) I am daydreaming too often [16:02] bzoltan: right, that's why I'm suggesting talking to the airline folks as they're better-placed to do things that aren't horrible ongoing  :)bodges [16:02] err rearrange the end of that to make sense [16:03] cjwatson: Yes, I will do so. Thanks for your input. [16:06] asac, i tested utopic silo 15 on both utopic-proposed and on my krillin with rtm + silo 1 and silo 16 (ofono and shim) [16:07] asac, no more crash in the wizard... but there is still a crash file left by _usr_bin_ubuntu-location-serviced [16:07] sim lock/unlock also worked in settings, and i verified i could unlock a locked sim from the shell after reboot === Ursinha is now known as Ursinha-afk [16:14] kenvandine: locationserviced is not a regression afaik [16:14] ok [16:14] good [16:14] so its fine and we can tackle it after imo [16:14] who maintains the 7digital scope? [16:16] asac, so syncing the landing will also pull in the location settings which is line 20 [16:16] asac, which requires a manual step for trust-store for testing [16:16] asac, should i change rtm silo 15 from a utopic sync to sync from the current utopic landing? [16:17] the end result will be the same [16:17] davmor2: ^^ fine with thsi? [16:17] i'll copy dbarth's comment over to the other line with instructions to manually start that [16:18] or rather a link to the bug it works around :) [16:18] davmor2: he is talking about getting system-settings that has the ziward crash into the silo 15 for one sign off [16:18] davmor2: alternatively we could first clear that 015 silo with the crash whitelisted given that we have it in the pipe and then just land that fix after [16:18] davmor2, since the location access stuff landed already in utopic, it'll pull that in regardless [16:18] kenvandine: wait for davmor2 ... he is testing those silos right now [16:19] asac: let me have a quick think about it while I test shouldn't take too long. [16:19] davmor2: thanks [16:19] kenvandine: just continue on your utopic silo for now :) [16:19] e.g. get that fixed, landed etc. [16:19] i like option 2 :) [16:19] davmor will ome back [16:19] asac, landed already [16:19] just not merged back [16:19] popey: someone in thostr_'s team, pete-woods would be a wild guess [16:20] sergiusens: ta [16:20] davmor2, note i tested the utopic silo on krillin rtm and it was all good for me [16:21] back to check, is there anything urgent? robert should be here as soon as his morning meeting ends. [16:21] popey: you need to speak to facundobatista I think [16:21] pete-woods: ok [16:21] robru: note the cyphermox's missing mtp package, it needs to be hunted. possibly a case of "publish returns success but does not actually publish anything"? I've gotten into habit of double checking rtm publishings on what they I actually did [16:24] Mirv: yep, there absolutely was (and may still be) a bug where rtm publishings report success without actually doing anything. [16:24] cyphermox: ^ [16:24] robru: and that ^ a watch_only build seems to fail after sync build, the configuration seems empty. I've resolved such cases by adding the package names manually instead of using sync:N, after the build, reconfiguring, and building with watch_only after that. [16:25] Mirv: yep, the sync:N logic isn't implemented very well. unfortunately I'm drowning in fixing up basic code quality of citrain, it's hard for me to even look at large bugs like that. [16:25] so basically nothing trainguards can't workaround, but the workarounds need to be applied every time [16:26] robru: yes, I don't think it should be tackled now, but sync works for syncing the sources, then manually reconfiguring + building with watch_only should work for now. [16:26] for users of the train it should all be transparent, we just have some more clicking to do than optimal :) [16:30] Mirv: "requiring more clicking than is optimal" is the core ethos of citrain. [16:32] ogra_, if you don't mind: https://code.launchpad.net/~nskaggs/phablet-tools/remove-python2-support/+merge/233754. I fixed the bug nik90_ mentioned on g+ the other day by just removing python2 [16:38] robru, In silo 11 I think I fixed the approval, but the dashboard is still showing an issue. [16:38] robru, Is that a manual thing? [16:38] tedg: the dashboard is showing the result of the last publish attempt [16:39] Ah, okay. [16:39] tedg: yeah, the MP looks good, just need to run the publish job again. I'm on it [16:39] robru, Can you please publish silo 11? [16:39] Thanks! [16:39] balloons, i assume we dont have anyone running these tests on older images at home ? [16:40] asac, kenvandine: so I don't think that the wizard crash is the last of the blockers. So let me try and get settings, sim unlock and location landed, if it is just the wizard crash we can whitelist, if it isn't there will still be 3 free silos in about 20minutes that you can add the new fix too, hope that makes sense [16:40] cjwatson: ah, that silo has a NEW source package, can you review it? or should I just publish it to the real NEW queue? ;-) [16:40] robru: NEW source? easier for us to review that in the real queue [16:40] it's only new binaries that need pre-review [16:40] davmor2, ok, do you know what the other blockers are? [16:40] ogra_, I think that is a very fair assessment. Some of the testsuites even require python3 now [16:41] balloons, ACK then ... not sure when i can land it though, i'm a bit swamped in developer mode fixes [16:41] davmor2: cool. yeah, lets get the big pile pushed and then get the crash fix in right after [16:41] davmor2: the crash fix we can put into the HERE silo then [16:41] so we can land those together [16:41] cjwatson: ok i'll publish it for real then, thanks [16:41] we could already built it there [16:41] sergiusens, second pair of eyes for: https://code.launchpad.net/~nskaggs/phablet-tools/remove-python2-support/+merge/233754 ? [16:43] ogra_: why not [16:43] cjwatson: do you remember if there's some kind of whitelist before citrain is able to sync a NEW source into utopic? I remember it used to be that way on the old daily_release system (pre-citrain), but I can't remember if it's still like that. anyway let me know if you don't see indicator-datetime show up in NEW shortly. [16:44] asac, kenvandine: oh sure if you can build it in the here silo go for it I thought it needed targeting against 15 which seems to be fixing stuff already so didn't want that having to be re-tested [16:44] sergiusens, dunno, you tell me :) [16:44] robru: not afaik, I think the only reason this whole preNEW idea exists is because otherwise it used to block a whole stack [16:44] ogra_: already added needs fixing ;-) [16:44] davmor2, i think asac meant if we can, land silo 15 [16:44] oh [16:44] then we'll push the crash fix in the HERE silo [16:45] kenvandine: right. think you can already push the fix into the HERE silo [16:45] Mirv: any reason you didn't get packaging ack for silo 2? I see you published it but it's just sitting there requiring ACK [16:45] seems a simple diff [16:45] so its ready for davmor to get on right after signing off the current silo thatjust has teh wizard [16:45] if he doesnt trust us he could also take a sneak preview then :) [16:45] robru: IMO preNEWing of source packages is an obsolete concept; just having code review of the packaging should be good enough [16:45] cjwatson: ahh makes sense, yeah I'm glad we don't have stacks anymore ;-) [16:46] robru: indicator-display in https://launchpad.net/ubuntu/utopic/+queue now [16:46] cjwatson: anyway I looked over the packaging, seems sane to me [16:46] asac, how do i add it to the HERE silo? that's a sync:10 [16:46] cjwatson: thanks! [16:46] asac, can i add an additional sync? [16:46] err [16:46] robru: ^^ can you help kenvandine ? [16:46] kenvandine: no. no you can't do that. [16:46] robru: he wants to sync from archive now i think [16:46] into the HERE silo [16:46] I THINK [16:46] :) [16:46] well... i was thinking from utopic silo 15 [16:47] it's in utopic-proposed [16:47] not release [16:47] kenvandine: isnt that in the archive now? [16:47] right [16:47] but will soon be [16:47] robru: let us know what options we have :) [16:47] settings takes longer than most to publish [16:47] or rather let kenvandine know [16:47] kenvandine: you can't have multiple sync:s, but you can list additional source packages and copy-package them directly. and then reconfigure the silo and WATCH_ONLY build... [16:47] i think the autopkgtests [16:48] queuebot: yay! ;-) [16:48] robru, oh... i could do a copy package then :) [16:48] kenvandine: copy-package ALL THE THINGS [16:49] robru: I am still testing the UITK in silo10 for Ubuntu. The QtCreator MRs have no Silo. Do you think it would be OK to use the silo10 for QtCreator plugins too? [16:49] robru: because I see that it is hard to get silo today :) [16:50] bzoltan1: yes there are currently none available. feel free to merge your requests and then I can reconfig the silo for you. [16:53] robru, i don't have upload rights to the ppa [16:53] copy-package --from=~ci-train-ppa-service/ubuntu/landing-015 --suite=utopic --to=~ci-train-ppa-service/ubuntu-rtm/landing-004 --to-suite=14.09 -b ubuntu-system-settings [16:53] robru, ^^ [16:54] kenvandine: are u core-dev? [16:54] yes [16:54] but i guess that isn't good enough for the ppa [16:54] i can add you to the uploaders for that ppa then [16:54] asac, cool [16:54] without any alignement [16:54] one sec [16:54] done [16:55] kenvandine: just dont misuse :) [16:55] you can now [16:55] ok, copied :) === Ursinha-afk is now known as Ursinha [16:57] I think it would be possible to give ubuntu-core-dev an upload ACL on the ci-train-ppa-service PPAs without making them a member of the team and causing everyone to get loads of build failure mails [16:57] though it would have to be done individually for each PPA [16:59] cjwatson: why is everybody afraid of the build failure mailvalanche? don't people have mail filters? ;-) [16:59] it's rude to impose that on a big team === boiko_ is now known as boiko === plars changed the topic of #ubuntu-ci-eng to: Train support: trainguards | Vanguard: cihelp | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: Ping robru if any citrain jenkins jobs have unexpected results. [17:05] robru, So desrt said that he should have the systemd-shim patch we need for cgroups by tomorrow morning, which makes me want to try the UAL cgroups landing tomorrow afternoon. [17:05] robru, How does that work from the "expected amount of chaos" perspective? [17:06] robru, Are there are big landings you're expecting? === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [17:20] 19:16 < davmor2> asac, ogra_, robru: also I'm happy with 001, 015 and 016 combination, but they will all need to land in the same image or they will break stuff between them. [17:20] robru: ^^ [17:20] thanks! [17:20] all push, then image, then we go for HERE silo :) [17:24] asac, we're talking about rtm silos, right ? [17:25] yeah [17:25] robru: ^^ [17:25] see above [17:25] :) [17:25] ogra_: dont you know how to publish? [17:25] asac, i sure do [17:26] ok ... well i will not suggest anything that would not be normal as i am sure then something bad will happedn :) [17:26] asac, 001 cant land [17:26] why? [17:26] according to the dashboard it hasnt been signed off or even tested [17:27] http://people.canonical.com/~platform/citrain_dashboard/#?distro=ubuntu-rtm&q= [17:27] davmor2: davmor posted above that he signed it off [17:27] err [17:27] ogra_: ^^ [17:27] asac, i think awe_ just said it misses top approbval in one MP [17:27] robru: the silo10 is ready for reconf [17:27] asac, which means it will block [17:27] ogra_, asac, it's *not* ready [17:27] rsalveti: can you check out the MP for 001 [17:27] which is why it's not top-approved [17:27] awe_, thanks ... thought so [17:28] regressions found and bugs filed on Fri, I will update the MP shortly [17:28] asac: check #phablet :-) [17:28] k [17:28] shorly meaning? [17:28] :) [17:28] hehe [17:28] * asac will go for dinner and check back later [17:28] this is a key part to get system settings in and also HERE [17:28] etc. [17:29] many thing are depending on this [17:29] so no pressure [17:29] hehe [17:29] rsalveti: ^ [17:29] cu [17:29] asac, sure...but we can't land things that are obviously broken [17:29] * asac checks out for 2h [17:29] just make it so that its not more broken than the current situation :) [17:29] so no regressions [17:29] hehe [17:30] anyway 2h gone [17:30] will check then [17:30] we could go and just do the fix enter pin part [17:31] we could just fix everything and land all silos a day before release, sure [17:31] right, but i am suggestint the opposite ... just cherry pick the sim enter one if the other part is the problem [17:31] split up landings [17:31] hehe [17:32] anyway, think its under control for the next couple hours [17:32] asac, sure [17:32] robru: no, just ran out of time [17:32] plars, looks like we're missing a device for the 233 utopic-proposed tests ? [17:33] wow, filemanager is as bad in utopic as it is in rtm [17:33] and we seem to have the same 9 failures in unity8 [17:33] in both images [17:35] while here, I'll try fixing boiko's landing before going towards sleep, now that the media landing happened. [17:37] Mirv: please let me know if you need anything from my side [17:39] boiko: looking good, it should be ready for testing once the build job reports so (it's running watch_only build on the triggered rebuilds) [17:39] and amd64 already succeeded [17:40] nice! [17:40] * boiko flashes the phone with the rtm image again for testing [17:42] ogra_: good catch! I restarted one earlier but there's a 3rd one that needed it. It's running now [17:42] ogra_: and the security fix seems to work locally, merging it now [17:42] jdstrand: ^ [17:43] plars, yay [17:43] overall that looks really good ! [17:43] nice, thanks! [17:45] these security issues once again show how moot the percentages are :P [17:45] tedg: i dunno, it's all chaos, as long as we build images in between the big ones I'm happy [17:45] sergiusens, replied back. I would rather keep the changes separate as phablet-click-test-setup has some other issues as well and may take more time to land [17:46] bzoltan: sorry for the delay, was having breakfast. done: https://ci-train.ubuntu.com/job/prepare-silo/1866/console [17:47] balloons: as this change requires a ci run of every test I don't think it will be fast enough anyways; I'd rather do the full run once only. [17:47] robru, K, do you know when other big ones are? [17:47] balloons: unless you take care of the actual landing; then it's fine [17:47] sergiusens, fair enough, fair enough.. I'll drop the second MP and roll the changes into the one that is open [17:49] tedg: sorry no, just published rtm1 and rtm15, will build an image once they land [17:49] robru, 001 was definitely not ready [17:50] and according to davmor2 there wwere three silos that need to go together [17:50] (001 being one of them= [17:50] ) [17:50] robru: Mirv: so what's your suggested course of action for mtp; should I copy it in rtm, should something else be done? [17:50] ogra_: yeah, on davmor2's approval I published them... [17:50] robru, if you scroll 50 lines up you see the conversation about this [17:51] ogra_: ah sorry I just read the ping itself not the whole scrollback. [17:51] robru, that was blocked by the owner of silo 001 [17:51] and the MP sholdnt be landeable at all [17:51] it was not top approved [17:51] how could you publish that ? [17:52] ogra_: yeah, how could I publish that?? maybe because rtm syncs don't deal with MPs, they do source package copies from other silos and/or archives. [17:52] robru: can I get a silo for row 67? I see you're running low, so I have no problem with "not today" :-) [17:52] ogra_: seriously rtm1 doesn't have an MP. [17:52] ouch, right [17:52] robru, well, it has a corresponding utopic silo thats blocked [17:53] ogra_: ok, well, fix it in utopic and we'll re-sync it to RTM I guess. [17:53] awesome. [17:53] robru: thank you [17:53] bzoltan: you're welcome! [17:53] ralsina_: eh, can you be fast? [17:53] robru: well, it has to go to rtm but I can get the utopic stuff done in 45' or so [17:53] robru, except that everyghing will be broken now in rtm til that is fixed (by tomorrow i was just toold) [17:53] ralsina_: ok cool [17:53] ogra_: well, here we are. [17:54] yeah, bad ... [17:55] ogra_: perhaps we'll be saved by that bug that makes rtm silos report publication without actually having been published properly. [17:56] robru, Oh, I think I just merged and cleaned 11 before the rtm silo was made. [17:56] robru, Did I screw that up? [17:56] tedg: yes but no worries, we can just sync from utopic rather than the silo, no worries. === Ursinha is now known as Ursinha-afk [17:57] grumble [17:57] robru, Okay, so I just need to put the package name there? [17:57] tedg: on the rtm row, additional sources column should read 'sync:ubuntu,utopic your package names here' [17:57] tedg, commit message added [17:57] ralsina_: you got 12 ^ [17:57] robru: thx! [17:57] ralsina_: you're welcome! [17:58] robru, K, updated line 51 [17:58] robru, Can I get silo 11 for line 53 ? [17:58] It should make everyone else build faster :-) [17:59] tedg: sure, once it finishes freeing (sloooow) [17:59] it's a race, let's see if silo 6 or 11 frees faster ;-) [18:00] Heh, go! [18:01] robru, can you help me clear up some confusion re: the ofono silos? Apparently I've been told that silo-001 has landed, although the "Testing pass? Image #" hadn't been recorded? [18:01] charles, Thanks, think we're good. [18:01] awe_: right, so what happened was that davmor2 said it was tested, so I published it [18:02] robru, right... we need to figure out some way for there to be coordination between QA and the lander [18:02] I will keep this in mind for the next landing [18:03] awe_: well somehow davmor2 got it in his head that it was his job to be testing silo 1. he probably got that idea by reading from the dashboard that silo 1 was tested and ready for QA signoff (or maybe somebody told him that, I dunno). anyway there are systems in place for communicating this stuff, we just need to use them. [18:04] robru, well he got into a situation where he had sim locking problems, because of the shim landing... [18:04] and added silo 1 to try to get out of it, because of the description on the silo [18:05] robru, he wasn't intending to test it when he did... but he hit a problem that seemed related [18:05] greyback: Saviq: please approve your MPs https://ci-train.ubuntu.com/job/ubuntu-landing-019-2-publish/11/console [18:05] robru, oop [18:06] kgunn, can you https://code.launchpad.net/~mir-team/qtmir/gles-sync/+merge/233650 [18:06] Saviq: done [18:06] greyback, and I can't reproduce your issue with the sudo pwd :/ [18:06] Saviq: ok I'll approve it as it unblocks stuff [18:07] greyback, yeah, it's definitely better than before... [18:07] slow internet is slow [18:07] robru, done, sorries [18:07] kgunn, unping [18:07] Saviq: robru: ok all should be approved [18:10] robru, can we just publish silo-018, and then merge? We'll handle the reported regressions in a subsequent landing. [18:11] awe_: sounds good to me! [18:11] awe_, robru, it would probably also be nice to have a new blocker bug for this and have it mentioned in the landing mail [18:12] since that image we're trying to build is planned to go to the customer for QA [18:13] tedg: ok you got silo 6 [18:13] ogra_, have we promoted an rtm image yet? [18:13] robru, Great, thanks! [18:14] tedg: you're welcome! [18:14] awe_, no, but the landing robru just did closes one of the blocker bugs [18:14] ogra_, rtm14 tag added to the new bug [18:14] ogra_: oh, I fixed something? yyaaaaaaayyyyy! [18:14] awe_, so it would be promoted without having a new one [18:15] ogra_, you mean a block-promotion bug? Note sure how this is indicated [18:15] ogra_, I added rtm14 to the new bug and it's Critical [18:16] awe_, we just need a bug for the issue and robru needs to add it to the blocker list [18:16] https://bugs.launchpad.net/ubuntu/+source/ofono/+bug/1366231 [18:16] Ubuntu bug 1366231 in ofono (Ubuntu) "[krillin] When FlightMode disabled ConnectionManager interface isn't restored" [Critical,In progress] [18:16] awe_, so give the bug number to robru and he should put it in the blocker list of the landing mail [18:16] (not sure where else) [18:17] ogra_, how does the merge happen now? Do I just top-approve the MP? [18:17] awe_, yeah, and someone needs to publish the silo after tthat [18:17] that should then auto-merge [18:18] awe_: I was testing the shim that helped fix the sim card lock issue but the rtm branch meant you could type in any number and it would say it was wrong including the right number, So without your fix from silo 001 it was impossible to test and land the other fixes that were needed so we could unblock the location fixes [18:19] robru, you landed 016 too, right ? [18:19] awe_: In general a pig of a day that needed 3 silos landing in order to test 1 [18:20] davmor2, right, but ofiono wasnt even remotely ready ... [18:20] ogra_: no, somebody else did before I even saw it. it was already merged & cleaned before I even got the ping to publish it. [18:20] robru, ok [18:22] ogra_: Why are we landing stuff in silos that isn't ready? Surely that is what trunk is for, Then you request the silo when it is ready to land? [18:23] davmor2, both silos get assigned at the same time ... not sure why 001 had content before ofono landed in utopic though ... it shouldnt [18:23] ogra_: I think it might of been filled directly [18:23] davmor2, I care about and do most of my testing on rtm [18:23] davmor2, right, still, the rtm silo should only be filled once the utopic one landed [18:23] davmor2, and I don't like landing things in either place with regressions [18:24] davmor2, by the rule nothing should be in the rtm silo before that [18:24] anyways, let's try and keep more in sync with landings... this kinda slipped thru our own processes [18:24] not sure why it did here [18:24] ogra_, no not true [18:24] robru, Can I get an rtm silo for line 52 please? [18:24] silos can be done in parallele [18:24] awe_: Indeed and that is why I thought it was safe [18:25] davmor2, well... the utopic MP hadn't been approved, and neither utopic nor rtm silos had a testing pass set to "yes" w/image # [18:25] awe_: I figured it just needed to wait for you to come on and test it against rtm but had already landed in utopic and so was safe. But to be fair is did fix the sim unlock issue I had landed in [18:25] anyways, what's done is done [18:26] awe_, they shouldnt [18:26] tedg: ok you got rtm16 [18:26] robru, Great, thank you! [18:26] awe_, the plan is that you get an rtm silo reserved when you get your utopic silo [18:26] awe_, but only fill it once utopic landed [18:29] ogra, marked the MP as top-approved; looks like someone already clicked publish for utopic [18:29] does the merge happen automagically? [18:29] hmm, dunno [18:30] i know that sil added a check that blocks promotion if an MP isnt top approved [18:30] i dont get how that could get published at all [18:32] * davmor2 blames the shell script, in the python script, in the perl script, in the javascript, in the ether [18:39] robru, hey - is there any code around in citrain that can update the spreadsheet? preferring python [18:45] robru: silo 12 ready to publish, if you can start a copy to rtm, I'm done with it [18:57] brendand: no, it doesn't work that way. you have to write JS code in the spreadsheet that polls your other thing and updates itself. you can't push data into the spreadsheet without triggering horrible revert explosions (it doesn't scale and then google solves this by throwing away hours worth of work at a time) [18:58] brendand: but don't write any code in the spreadsheet because that particular code base has no version control, no unit tests, and no review process. we just live-edit production. it's a total clusterfuck that i'd like to see shrink, not grow. [18:58] robru, ah i was worried about swearing but well, you've set the precedent now [18:58] so i say, poo === Ursinha-afk is now known as Ursinha [19:01] ralsina_: ok you're building in silo 17: https://ci-train.ubuntu.com/job/ubuntu-rtm-landing-017-1-build/4/console [19:01] robru: ack [19:01] robru, so where should i put my code then? [19:02] brendand: you just... don't. I dunno. you're trying to make the trello board update the spreadsheet status right? [19:05] robru, yup [19:08] brendand: yeah I'm not sure what to tell you. citrain wasn't designed for this level of web-service integration hooks. we're not github. this isn't web2.0. This is a pile of shitty python scripts and a spreadsheet. it's a miracle that any of it works, and it's a delicate balance. [19:08] brendand: the dashboard and queuebot only work because they are read-only things that poll the spreadsheet, they don't put anything into the spreadsheet [19:10] robru, no worries - i can tell anyone who asks it's just not feasible right now [19:10] robru, i'll focus on the things i can do with read-only [19:15] robru: hahahaha nice I'm glad to see you rank your fellow team-mate coding skills so highly ;) [19:16] robru: silo rtm-17 job says SUCCESS but the packages are not in the silo [19:16] robru: I'm with you though sometime I'm amazed the bandaids are still stuck on :) [19:17] ralsina_: https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu-rtm/landing-017 looks fine to me? [19:23] robru: ha, it's there now. I'm impatient I guess :-) [19:26] ralsina_: no worries [19:36] tedg: your'e building in rtm1: https://ci-train.ubuntu.com/job/ubuntu-rtm-landing-001-1-build/45/console [19:41] robru, How does that work with rtm 1 and 2 ? [19:41] robru, 1 should pick up 2's changes since we built them in utopic in order. [19:42] robru, Should we just kill rtm2? Or let QA test them in order? [19:55] tedg: oh I didn't notice the duplication there. yeah I guess just skip right to the newest one, if they were released to utopic in order then rtm1 should contain everything in rtm2. no point wasting QA's time with two very similar landings. [19:55] tedg: make sure you test rtm1 in in rtm ;-) [19:55] before sending to QA [19:58] robru, Cool, on it. [20:04] robru: can you publish silo 12 please? [20:05] ralsina_: it was already, but merging it for you now [20:05] robru: oh, thx [20:09] ralsina_: you're welcome! [20:09] * robru -> lunch [20:22] hi trainguards! Dobey and I are working on some bugfixes for pay-ui, because gatox is on vacation. Now we got a new click for pay-ui that we'd like to submit to the store. I know that gatox usually asks fginther to upload those packages... but is there any other thing that we should keep in mind? [20:24] alecu, there is a review utility that that app store reviewers use, make sure that passes [20:25] alecu, at the moment I can't find what this called [20:26] alecu, I think this is it: https://code.launchpad.net/~click-reviewers/click-reviewers-tools/trunk [20:26] great, I'll check with that too [20:29] seems to be already packaged as click-reviewers-tools [20:31] but doesn't seem to be up-to-date. [20:33] alecu: yeah just get the trunk of that [20:35] robru: fginther: trunk gives two errors, but I think that should be expected from pay-ui, since it runs unconfined and it has a special hook: http://pastebin.ubuntu.com/8293383/ [20:35] popey: do you know about that? ^ [20:36] alecu: seems reasonable, i'd say go for the upload and let popey sort it out ;-) [20:36] great. fginther, robru, where should I upload the click package? [20:39] alecu: not sure. only a tight cabal has the password to upload core apps. ping popey [20:48] ok, I've sent mail about that to popey and francis. [20:54] robru, Can I please get a silo for line 60? [20:55] tedg: silo 9! [20:55] robru, Great, thank you! [20:55] tedg: you're welcome! [20:59] alecu: if fginther uploads it, I'll approve pay-ui [21:18] robru: all going fwell? [21:20] asac: yep, fixing up citrain piece by piece [21:20] asac: publishings coming alnog as well [21:37] robru: citrain busted? what did you do :)? [21:40] robru, once you are done playing with the citrain :-), can you give a quick peek at the autopilot packaging changes I have in my MP and approve/deny? It's just adding a couple depends https://code.launchpad.net/~nskaggs/autopilot/fix-1328600/+merge/227399 [21:42] robru: anyway. i will be out today. wonder how close we got in elimiating our big blockers today on rtm [21:42] but will check the mail tomorrow [21:42] so we can continue on that mission [21:42] balloons: lgtm, but I'd prefer if you could sort the dependency list. Please run 'wrap-and-sort -a -t' in the source tree root and then re-add the comment in debian/control that that'll discard [21:42] promotable == rtm branch [21:43] asac: yeah, from what I gathered, I published a regression but I fixed another, we'll have to wait for QA to get the next image to know for sure [21:43] good luck. if you need any support or so, slangasek is probably around for a while longer; otherwise text me :) [21:43] robru: published a regression? [21:43] hmm. with QA sign off? [21:43] no wayt o back out? [21:43] well, i will let you figure and hope we can resurrect that tomorrow to promotable state at least [21:43] asac: yeah, no worries, upstream is on it [21:43] for backouts etc. steve can probably help if you want to get that out [21:44] thanks [21:44] robru: right, but remember that that will take days usually [21:44] :) [21:44] upstream fixing someting quick is rare [21:44] hehe [21:44] backout is usually better [21:44] cu [21:45] robru, command run. what do you mean by 're-add the comment in debian/control that that'll discard'? [21:46] balloons: running wrap-and-sort discards comments in debian/control, which you should have one if you follow the citrain packaging guidelines. check your diff. if you see a comment block deleted from debian/control, copy&paste it back in there [21:47] robru, ahh.. k, no comments in the debian/control :-) [22:14] robru, if you can add your 'OK' to the MP I'd appreciate it [22:19] balloons: oh sorry, sure [22:21] balloons: ah that wrap-and-sort was noisier than I'd expected ;-) [22:21] robru, yes, veebers said the same thing ;-) But it's done [22:22] I thought it would just inline my changes [22:22] I'm blaming you for that robru ;-) [22:32] robru: hi. any idea why something would be on the "Archive" page of the spreadsheet, but not actually in ubuntu-rtm? [22:33] dobey: probably something exploded. it's happening. if you notice it, just make a new request and I can take care of it [22:34] robru: ok [22:34] alecu: ^^ [22:34] veebers: blame away! I love sorted debian/controls ;-) [22:35] dobey: ouch :P We better take a close look at the latest landings. [22:35] robru: is that fixed if we do a new landing + srccopy? [22:36] i guess it would be [22:36] if just asking for a srccopy again would fix it, anyway [22:36] alecu: yep, any new landings that go through utopic will contain previous releases from trunk by definition, so everything that gets released to rtm will by definition contain all previous releases, even if the previous release never made it to rtm [22:37] and we need to do a new landing for ubuntuone-credentials at least (once my branch is approved), so maybe we should just do that [22:38] alecu: do we need to do a unity-scope-click landing? [22:38] bfiller: please approve your MPs. https://ci-train.ubuntu.com/job/ubuntu-rtm-landing-020-2-publish/2/console also I have no idea why you have MPs in an rtm landing, that makes no sense to me [22:41] brb folks, gotta get dinner started [22:52] alecu, are you still around? [23:29] robru: do you need help with a revert for whatever this regression is you mentioned above? [23:32] slangasek: I don't think so. upstream said they'd have a fix by morning. if not, europeans can look at reverting it. I'm not in a great position to know what the image quality is currently like, currently I'm knee-deep in citrain internals [23:33] slangasek: the thing is, the regression I published fixes a different blocker. so if we revert it, from what I understand, we don't actually get ahead there. so I think we should just wait for upstream to fix that regression and enjoy the other blocker that got fixed. [23:36] robru: ok; so the regression and the fix weren't in unrelated uploads, and a clean revert doesn't get us anywhere - thanks for confirming [23:37] slangasek: you're welcome!