[08:11] I see CI Train spreadsheet error reports in my inbox [08:11] We might have a problem... [08:17] :( [08:22] Anyway, time to open the gates officially === sil2100 changed the topic of #ubuntu-ci-eng to: Need a silo or CI Train support? ping trainguards | Need help with something else? ping cihelp | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: low on vivid silos. Spreadsheet again having issues [08:23] Yeah, just got a fatal error on the spreadsheet, damn it [08:23] The funny part the problems started in the middle of the weekend! [08:24] So actually when no one was really using the spreadsheet, so all the theories about too many users accessing at once has been made invalid [08:24] yeah, I already thought it's probably not that [08:45] === IMAGE RTM 271 building (started: 20150504-08:45) === === chihchun_afk is now known as chihchun [10:20] === IMAGE RTM 271 DONE (finished: 20150504-10:20) === [10:20] === changelog: http://people.canonical.com/~ogra/touch-image-stats/rtm/271.changes === [10:46] ogra_: ^ can we get a proper changes file? [10:59] ugh [10:59] Indeed, something is b0rken [10:59] ogra_: ^ [11:04] imgbot, status 271 [11:04] Status: succeeded, Started: 2015-05-04 08:41:26 UTC, Finished: 2015-05-04 09:32:37 UTC [11:04] Build URL: https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu-rtm/14.09/ubuntu-touch/+build/26101 [11:04] Changelog: http://people.canonical.com/~ogra/touch-image-stats/rtm/271.changes [11:07] Mirv, that will take a bit ... i need to fiddle with the scripts to use the LP build url of the manifest (cdimage flushed all old images/manifests because they got to old to keep) [11:07] ogra_: your scripts are bash-based, right? [11:07] well, shell based ... not bash :) [11:08] Yeah, sorry, old habits of using those terms interexchanably ;) [11:09] the scripts are in the respective dir with the .changes files ... [11:09] (compare-manifests and compare-manifests-rtm) [11:50] Damn, some workers are drilling next to my flat, hard to concentrate [11:50] brb, lunch [12:20] Mirv, http://paste.ubuntu.com/10984182/ ... should be it [12:21] ogra_: thanks! ooh, updated translations for my Bq upcoming! [12:21] yeah :) [12:21] the store categories have been a bit annoying but I read they were translated recently [12:24] talking about Bq, I just turned the display on and u8 lockscreen is there but touch is not working [12:24] oh, it was probably running apport and it crashed [12:25] CI Train dashboard broken again? [12:25] "A server error occurred. The page will reload." [12:25] rvr: the spreadsheet, yes, causing some troubles [12:27] It doesn't let me sign off a silo [12:32] Ah, good, finally [13:24] kenvandine: ping [13:25] rvr, pong [13:25] kenvandine: Good morning [13:25] good morning [13:25] kenvandine: I'm with the lists updates hotfix [13:26] kenvandine: I was wondering whether the bug still applies [13:26] which bug? [13:26] kenvandine: I mean, the test case (umetronome) ... http://launchpad.net/bugs/1449582 [13:26] Ubuntu bug 1449582 in ubuntu-system-settings (Ubuntu RTM) "lists updates depending on frameworks newer than the device ones" [Critical,In progress] [13:26] yes it does [13:26] well it did friday :) [13:27] kenvandine: Ok, checking [13:27] rvr, it's an important fix to land, it's going to affect lots of rtm users [13:27] as packages get updated in the store [13:28] so far we only found one that caused the problem [13:28] but the bottom line is the update plugin was using the wrong url for the store and not filtering arch or framework [13:28] it was using the old dev api, before any of that existed [13:31] tvoss: could you add landing description to the request? [13:31] sil2100, sure [13:31] tvoss: thanks :) [13:31] tvoss: sadly, the landing will have to wait a bit since we currently have no vivid silos [13:32] We'll increase the number of silos shortly though [13:32] sil2100, fixed [13:54] sil2100, heya [13:55] cwayne: hi! [13:56] sil2100, hey so, since we agreed that updating any scope that's in the store should go through QA, I was wondering how to best accomplish that? do we just add it to the spreadsheet with a link to the click? [13:57] cwayne: which scope would you want to update? Generally that's what we would love to start happening, yes [13:57] Usually an e-mail or an entry in the spreadsheet as with clicks would be sufficient [13:58] sil2100, twitter-trending, believe it fixes a crasher in today-scope. very very low-risk as well :D [13:58] Famous last words ;) ! [13:58] lol [13:58] Yeah, let's give it to QA I suppose, they might decide it doesn't need much QA anyway [13:59] right,t hat's fine, I just want to start doing updates more properly :) [13:59] well, I mean they did go through qa during the custom update, but still, more proper I suppose :) [14:25] is there something weird going on with the spreadsheet? [14:25] i keep getting some popup saying there's a fatal error [14:28] cwayne: yeah, as per topic, spreadsheet is brokenish again [14:29] derp, sorry [14:35] jibel, robru: I need to jump out now to the vet with my two lizards, I'm pretty sure I'll be back before the meeting but it all depends on traffic/queue [14:44] kenvandine: Approving ubuntu-rtm silo 5 [14:45] rvr, thx! [14:59] rvr, don't forget to mark silo 5 as qa verified on the spreadsheet [15:03] kenvandine: I did, but the spreadsheet is broken :P [15:03] sigh [15:03] rvr, ok, thanks! [15:03] * kenvandine grumbles about the spreadsheet [15:03] It presented "Granted" to me... just re-granted [15:04] cool, thx [15:11] sil2100: yeah, spreadsheet is kind of broken [15:11] kind of ... [15:29] rsalveti, Hi! powerd package install is failing with: http://paste.ubuntu.com/10985054/ [15:29] re: silo 16 [15:32] om26er_, sudo umount /usr/share/powerd/device_configs/config-default.xml ... then try installing it again [15:33] (make sure to reboot afterards to have it properly mounted again) [15:33] yeah [15:33] we have the instructions in the wiki page [15:33] om26er_, https://wiki.ubuntu.com/Process/Merges/TestPlans/Powerd [15:34] yeah, that one [15:34] thanks [15:34] om26er_, you must install the packge from recovery [15:34] jibel, ogra_ aah, didn't see the TestPlan, was expecting installing instruction in the spreadsheet :) [15:34] ogra_, that seemed to have worked. [15:34] :) === om26er_ is now known as om26er|otp [15:35] spreadsheet giving tons of erros [15:35] errors [15:35] yeap ... [15:36] see topic [15:42] Yeah... [15:43] As always, not much we can do [15:43] But the replacement is closer every day! [15:43] send google a bill for the work on the replacement is what we should do :P [15:43] Google doesn't care ;) [15:43] indeed [15:44] We filled in formal complaints and bug-reports and they just send those back without doing anything [15:44] I'll know more about the ETA after todays integration meeting possibly [15:48] rvr, you passed rtm/silo 5 right? [15:48] its status reverted to needs sign off [15:50] jibel: Yes, 4 and 5 [15:51] rvr, ack, I delete the card [15:51] Argh, ok, publishing then === om26er|otp is now known as om26er [16:41] rsalveti, I am not able to find the unlock script do you know where it exists ? [16:41] om26er, plars should be able to tell you [16:41] I need that to verify "See if the lab-specific screen unlock logic still works:" [16:41] * ogra_ isnt sure ... i think it used to be shipped in the unit8 tree [16:42] om26er: it lives in unity8-autopilot [16:43] plars, hmm, how do I call it ? [16:45] om26er: iirc, it's meant to be called from the host side, so you'll want to start with adb pull /usr/share/unity8/unlock-device [16:45] chmod +x on it [16:45] then run it, I think it can take args for the serial and a wait command [16:59] plars, if I have passcode set, does it still unlock ? [17:02] om26er: if you look at it, it's just a dbus call that's happening. Should work fine. [17:02] adb shell "gdbus call --session --dest com.canonical.UnityGreeter --object-path / --method com.canonical.UnityGreeter.HideGreeter && echo Greeter unlocked" 2>&1 [17:03] om26er: the reason we use the script from unity8, is because that is their "official" way of doing it. If at any point in the future it changes, it just requires them to update that script and keep the calling mechanism the same, rather than coordinating to fix everyone using it [17:16] plars, after a few attempts it fails. [17:17] "I: Unlock attempt 5 failed, script output: 'error: closed' [17:22] om26er: that seems odd, do you see anything happening on the screen? How about if you try to run the gdbus command manually? [17:23] plars, error: closed [17:26] plars, isn't that expected ? If the screen is locked with passcode adb is not supposed to work. [17:26] om26er: ah, you aren't running with developer mode enabled? [17:26] plars, no [17:26] om26er: certainly you will want that, if you want this to work in any sensible way [17:26] plars, ok, lemme reflash :) [17:29] om26er: hm, yeah, that might have changed, plars would know the details [17:50] rsalveti, while running the last test here https://wiki.ubuntu.com/Process/Merges/TestPlans/Powerd [17:50] rsalveti, "active" stays in the last till the screen is on. It goes away after screen turns off. Is that expected ? [17:52] * rsalveti checking [17:52] om26er: guess that active one you have with screen on/off is from unity [17:53] rsalveti, right, so it needs a bit more clarity on how the fix can be verified [17:53] let me check here [17:55] om26er: yeah, it's unity, you need to test this without any app opened [17:56] let me change the wiki page [18:03] om26er: did it work now? [18:03] rsalveti, no, I am not fully sure about the steps. Did you update the wiki ? [18:03] aah, ok. I see your point now. [18:03] om26er: yeah, just run the same test but make sure you close all the opened apps before testing it [18:04] rsalveti, thanks. I was able to verify the fix now. [18:04] om26er: great [18:07] rsalveti, you need to update the test plan to point at how to unlock the screen. The current TestCase is wrong :) [18:08] trainguards: can you please land silo 16 ? [18:08] given the spreadsheet is broken. [18:08] I can land it [18:10] done [18:26] om26er: ACK [18:26] rsalveti: oh, thanks ;) [18:26] We're in a meeting now o/ [18:41] robru: ubuntu{,-rtm}/landing{31-40} adjusted to match the older ones. [18:41] infinity: thanks! [18:52] bfiller, Hi! === mhall119 is now known as mhall119|afk [19:13] excellent, everybody who touched silo rtm 3 is gone. [19:24] alright! 10 new ubuntu silos online, plus one free from before. who wants one? [19:26] bzoltan: zbenjamin: silo 16 [19:28] bfiller: brand new silo 33 is yours [19:30] ricmm: rsalveti: silo 34 for you [19:31] robru: that might be the one we just landed [19:31] rsalveti: awesome [19:31] can't see it in the spreadsheet [19:35] rsalveti: lol, awesome. well I assigned it from somewhere, now it vanished [19:36] wut [19:37] rsalveti: indeed that merge is merged [19:37] :-) [20:07] rsalveti, silo 4 rtm seems to have the wrong status [20:09] pmcgowan: hm, also another one that probably landed already [20:10] part of this ci spreadsheet mess that was caused when it was reverted to a previous revision [20:10] robru: mind cleaning up rtm 4? [20:11] rsalveti: oh, that one too? ok [20:12] rsalveti, can we fix silo 3 as well? sees stuck [20:14] let me see what is going on there [20:14] pmcgowan: rtm 3? yeah it's in an inconsistent state, I need info on that before I can publish it [20:15] rsalveti: pmcgowan: it seems the PPA has unity-scopes-api but the silo is not configured to have that or not, not sure if that package should be published or not. [20:15] robru: another issue with the spreadsheet? [20:15] since both were built via the ci train [20:15] rsalveti: not as far as I can tell. if the spreadsheet loses data that's one thing, but in this case the silo config itself is inconsistent. could only happen if somebody reconfigured the silo [20:16] they were not copied, so built via ci train, which probably means it was configured like that before [20:16] right [20:16] rsalveti: yeah, so at one point somebody configured the silo to have both, and then later reconfigured to remove unity-scopes-api. but it's not clear to me if that was on purpose or by mistake. so there's this unity-scopes-api package that I'm not sure if it should be published or not. [20:16] rsalveti: I sent an email to pstolowski about that since he isn't on IRC [20:17] got it [20:17] pmcgowan: ^, so we'd need his input/feedback here [20:17] rsalveti, ok lets wait then thanks [20:17] thanks robru [20:18] pmcgowan: you're welcome === mhall119|afk is now known as mhall119 [23:45] Is someone on top of mangling CI bits to know what wily is? [23:54] infinity: the train is mostly fine. the necessary cowbuilder chroot will spring into existence the first time a wily build is run. just some spreadsheet bits need to be updated to offer wily as an option. is it ready to go? [23:55] robru: Not quite, but getting there. [23:55] infinity: no worries, I put a bunch of effort into eliminating cases of distro-name hard-coding in ci train. so when it's ready the train should just spring to life.