[02:03] Mhmh... I thought that sync could do that automatically for me (ehm, sorry if I tried too many times), but at this point... Should I sync changes to trunk manually? :o === chihchun_afk is now known as chihchun [04:57] Mirv: up yet? [05:25] robru: here [05:28] Mirv: oh just emailed you. No rush [05:32] robru: looks good! :) I'll take a good look and reply a bit later [05:39] Mirv: thanks, goodnight! [05:39] robru: goodnight! [07:25] Sigh, my silo build is toast: No space left on device [07:25] Not just mine, by the looks of things... [07:31] hey trainguards, problem with silo 44 - "failed to write (No space left on device)" [07:36] pstolowski: try again? I think it's being a bit temperamental [07:36] Mirv: ^^ if this happens again, try contacting IS. looks like /var/cache/pbuilder is a bit big at 5.6GBs: https://ci-train.ubuntu.com/job/cyphermox-test/1063/consoleFull [07:43] robru, that was a second try [07:43] crap [07:47] robru: ok! [07:47] trying [07:47] Mirv: I recommend asking #webops to delete /var/cache/pbuilder/, it looks like there's two wily images in there somehow. train will regenerate the images on next build. [07:48] ok [07:48] Mirv: uh but then after the pbuilder is deleted only let one build job run. if two build jobs run simultaneously it might make two pbuilder images or break in strange ways [07:48] hmm [07:49] I asked now generally over there for the vanguard to look at it [07:49] Mirv: I mean just for the first one. once the first build is complete then parallel builds can resume [07:49] ok [07:50] webops is a bit busy but will hear more in a few minutes [07:50] Mirv: or at least once the first build uploads it's packages [07:51] kenvandine: no [07:51] kenvandine: ^ [07:51] no? [07:51] kenvandine: no disk space left [07:51] damn! [07:52] ok... === _salem is now known as salem_ [07:52] robru, so it'll fail [07:52] Mirv: https://ci-train.ubuntu.com/manage I disabled jenkins jobs from running, once disk is freed you can re-enable that here (or sil can do it if you can't do it) [07:52] kenvandine: it'll most likely fail. I guess there's no harm in letting that keep running [07:52] on the off chance [07:52] kenvandine: we're waiting for #webops to free some disk for us [07:52] robru, yeah... any eta on freeing? [07:52] oh god why am I awake [07:52] ok [07:52] haha [07:53] that's what i was wondering! [07:53] kenvandine: webops says "gimme a few minutes" [07:56] robru: thanks [07:57] it seems it's a few minutes more still [07:58] Mirv: you're welcome. you'll want to click 'cancel shutdown' from that menu in case that wasn't obvious (don't worry, nothing will shut down, it's just in a mode that prevents jobs from running) [07:59] robru: it was pretty obvious, yes :) I've been kicking jenkins around a bit so I'm not completely unfamiliar. [08:00] robru, btw... thanks for the magic in the reconfigure :) [08:02] kenvandine: oh this is just the beginning! I'm working towards making reconfigures not even a thing. You just assign and then changes in bileto reflect in Jenkins live. [08:02] Lots of little fiddly things to get sorted on the way though [08:03] Mirv: great, hopefully you have permission ;-) [08:04] robru, that'd be nice [08:04] robru, will you give us a shout out when the disk space issue is resolved? [08:05] kenvandine: nah I'm going to bed, maybe Mirv can [08:05] ok [08:05] good night! [08:05] Goodnight! [08:05] morning here :) [08:05] in london this week [08:18] kenvandine: I will [08:18] Mirv, thx! [08:18] still no response, I guess the alert is taking vanguard's time [08:19] vanguard also changed, asking the new one [08:23] sil2100: so we have train down because of lack of free space, stay tuned, webops are on it [08:37] Mirv: I'm still here for some reason, I'll turn jenkins back on and start one build [08:38] Mirv: https://ci-train.ubuntu.com/view/1.%20Build/job/ubuntu-landing-010-1-build/211/console trying one build, we'll see if it passes now [08:41] robru: yes I can see you are still here :) you should move to China, then your hours would be fitting! [08:41] Mirv: cost of living is better too! [08:48] Mirv: so far so good with the wily build I think. just need to see it make the vivid pbuilder and then we'll know [08:58] Mirv: ok looks good, 5GBs free even with both vivid and wily pbuilder images [08:58] kenvandine: pstolowski: we're back, go ahead with your builds [08:58] robru, k, thanks! [08:58] pstolowski: you're welcome [08:59] ah, only 2AM, must be bedtime [08:59] uh [09:08] robru, thanks! [09:10] What is this? Company-wide insomnia? ;) [09:11] sil2100: Ken's on London for a sprint. I'm the only one with crippling insomnia [09:11] kenvandine: you're welcome [09:12] OK I'm going to bed for real [09:12] Goodnight everybody! [09:12] hehe [09:12] Yeah, but there's also infinity with his really scary insomnia [09:12] robru: goodnight! [09:12] get some sleep [09:12] True [09:13] sil2100: whoever is building 20 didn't reconfigure properly, they will be disappointed [09:15] sil2100: "preparing packages. Silo ready..." Means "i clicked build before the reconfigure finished and now I'm building with stale config" [09:16] Anyways. Really goodnight now! [09:19] sil2100: I've never understood the timezone infinity lives in anyway, he seems to be 24h [09:20] boiko: ^ [09:21] robru: sil2100: yeah, I clicked the reconfigure before the jenkins down thing, and reconfigured afterwards so I might have built before the second reconfigure finished [09:22] I didn't realize jenkins would save the reconfigure job to run after its reboot [09:24] boiko: it didn't reboot, it just disabled running jobs while disk was full [09:24] robru: ah ok, so that was it [09:25] boiko: I dunno what you've done now, please cancel the build, wait more than 1s, reconfigure, confirm reconfigure finishes, then rebuild [09:26] robru: ok, I will do that, thanks for the heads up [09:26] When a silo has multiple statuses like silo 20 currently has three statuses, it's because jobs are running in parallel [09:27] boiko: fortunately this nightmare will soon be over, I'm working towards making it so you never need to reconfigure. [09:28] robru: cool [09:29] OK I'm literally in bed now, goodnight again ;-) [09:31] robru: crazy man [09:31] we should have a remote switch for robru's internet connection, to cut it off at midnight [09:31] Pls no [09:32] or to be more pleasant, a remote switch to being able to screen -r or whatever he uses for IRC [09:32] sil2100, we are re-running sanity on latest mako, flo and emulator images for OTA6 and we are resuming the landings. We'll start with 48 which was problematic for OTA6 [09:36] jibel: excellent o/ [09:37] Those long night stays for work are bad for everyone, good those only happen for me during the release week [10:30] jibel: anyway, the arale images look good so far, right? [10:30] Since I could copy those to the rc channel now [11:00] sil2100, arale, krillin and mako are good. Waiting for results on flo and this emulator crash on wily [11:00] jibel: if the crash is reproducible, we might need to include it in the release notes as a known issue [11:02] sil2100, yes davmor2 is testing with an updated qemu [11:06] sil2100, jibel: sorry was on another pc trying it out on vivid when I found out there was an update there too, so that is up and running now, and I'll retry here on wily in one second [11:08] sil2100, jibel: so it is stable on vivid, retrying on wily === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem [11:40] hello out-of-silos situation, long time no see [12:07] Mirv: ;) [12:48] jibel: Were you able to reproduce https://bugs.launchpad.net/canonical-devices-system-image/+bug/1470308 recently? [12:48] Ubuntu bug 1470308 in Ubuntu Clock App "launch clock app display a white page,and it take a long time" [Undecided,New] [12:50] nik90, sometimes, and ouch, I open the clock app, allow the location and the MX4 rebooted :( [12:51] jibel: really strange. Not sure how looking for location would block the UI thread. I will try to reproduce it on my end. [12:51] nik90, I think it depends if there is already a location acquired or not [12:52] jibel: Are there any other apps that ask for location access on app startup? I know camera-app used to do that. [12:52] nik90, osmtouch [12:52] googlemaps and here are web based [12:53] jibel: ok. Will test with osmtouch as well === jhodapp is now known as jhodapp|sick [13:04] bzoltan_, Mirv: do you guys know if this silo is still needed? https://requests.ci-train.ubuntu.com/static/dashboard.html#?q=ubuntu%2Flanding-049 [13:05] sil2100: I pass it to Mirv [13:07] ping cihelp. Can I get utopic removed from ubuntu-filemanager-app-ci and ubuntu-filemanager-app-ci-autolanding? [13:10] brb in a moment [13:17] bzoltan_: it's your silo, don't pass it to me ;) [13:17] Mirv: I give it to you.. .do not ya say I never give you anything [13:18] sil2100: it's bzoltan's "late" gles landing split into wily + vivid because the gles twin package without the main package didn't really work out in bzoltan's first silo [13:18] bzoltan_: how generous! [13:18] sil2100: it's basically waiting for QA to review the vivid silo, if ok then publish both (047 + 049) [13:24] Mirv: how truze [13:27] Mirv: of course the funny twist is that the silo49 has a wily UITK-gles :) and so I need to test it with the wily emulator ... what might be disfunctional for a dozen of unknown reason... might be. I will see in 10 mins [13:29] balloons: I'm on it [13:30] bzoltan_: well as long as you test the vivid 047 emulator it should be indicative of the update's quality. === _salem is now known as salem_ [13:31] bzoltan_: with luck though wily emulator works too since the wily image works.. maybe some of the tricks just done not yet there though, I'm not sure if Mir is up-to it since wily has 0.15.0 which maybe didn't have this hotfix [13:31] Mirv: that is what I mean... for example [14:09] Mirv: as I was afraid the wily emulator does not show much.. black screen only [14:32] bzoltan_: well check the vivid then and mark that at least ready. if it works, we can probably assume the identical wily release is good enough too. [14:32] bzoltan_: the wily problem is probably the mir fix lagging behind there [14:33] Mirv: well at least the silo49 does not make it worst :) [15:08] jibel, pmcgowan: I will have to jump out in 30 minutes to drive my girl to a doctor's appointment - leave me a message about the OTA tests results and I'll release once I'm back [15:11] sil2100, arale upgrade passed, krillin in progress [15:11] \o/ [15:24] * sil2100 brb === chihchun is now known as chihchun_afk [16:09] jibel: sil2100: krillin upgrades too \o/ [16:10] davmor2, awesome, I'll send the testing report and that's be it for OTA6 [16:10] \o/ === salem_ is now known as _salem [18:08] Trevinho: the sources field is for sources that aren't defined by your MPs. The sources you specified in silo 56 are redundant to the MPs you've given. I'll fix it, but just for future reference don't do that. [18:09] robru: ok, thanks... It was my first experience which such issue (and landing), so I wasn't sure :) [18:09] (issue = not synced down and upstream) [18:13] Trevinho: right, and in this case what you've done is fine since it was just a no-change you're overwriting. but generally if somebody uploads to distro you need to sync it back to trunk, what you've done is effectively erased the manual upload that went to distro. [18:13] robru: I asked to sil2100, that's why I did that [18:14] Trevinho: yeah it's fine in this case, you should just understand that you have "synced" anything at all, you've just clobbered all over steve's work, that's all. [18:15] Trevinho: https://ci-train.ubuntu.com/job/ubuntu-landing-056-2-publish/5/artifact/ubuntu-push-qml_content.diff/*view*/ this diff shows what you removed. [18:29] Thanks for pushing the buttons sil2100, I'm glad this release is out! [18:35] jibel: big thanks to you guys for testing! [18:35] Took a bit longer since I had to modify the phasing scripts [18:35] Since now we actually phase more than one channel... [18:51] josepht, how's the core apps filemanager changes coming along? [19:10] sil2100: you still around? [19:10] robru: yeah, what's up? [19:11] sil2100: I made some changes at https://requests.ci-train.staging.ubuntu.com can you check that they're ok? ;-) [19:11] Oh oh! [19:13] robru: I'm thinking about, generally I like it - first thing I like the reordering of fields, it makes much more sense like this [19:13] sil2100: thanks [19:14] robru: the additional controls are also nice and I think they can be like that too [19:14] balloons: thanks for reminding me. I've approved the MP and will redeploy the jobs once it's merged. [19:14] First I thought: hmm, those might annoy after a while, but on second thought it's still good [19:14] sil2100: the trick is really to declutter the page, I hope the hover behavior isn't too mysterious [19:19] sil2100: next step: I'm going to add a view that shows all requests that require some kind of trainguard action [19:19] eg everything that is publishable [19:19] but first, lunch [19:19] robru: oh, that would be convinient [19:19] Anyway, good direction so far :) [19:20] sil2100: thanks! [19:36] balloons: the filemanager-app jobs have been redeployed and no longer call the utopic jobs [19:57] josepht, thank you. I re-ran and see success === blr_ is now known as blr [22:49] Trevinho: https://code.launchpad.net/~canonical-platform-qa/autopilot-qt/wily_fix-unittests/+merge/268990 need this top approved [22:50] robru: I know, but there were some concerns about failures, I wanted to wait nuclearbob full approvation [22:50] Trevinho: alright no worries, please ping me once it's approved and I can publish [22:50] I'm not sure whether those failures are related (they're just rebuilds), but still [22:50] robru: ok, thanks [22:50] you're welcome [23:11] It's best to propose landing branches that are already approved by someone usually [23:14] sil2100: yeah indeed... But those arrived later, as by landing xpathselect we saw that autopilot-{gtk,qt} had to be rebuilt as well after