[05:29] hello zsombi === bzoltan1 is now known as bzoltan [05:29] bzoltan: hai [05:30] zsombi: have you heard anything from elopio or kalikiana about the AP failures I linked yesterday? [05:30] bzoltan: nopez [05:31] zsombi: it would be cool to get back the 1-2 days release cycle ... but it seems to be impossible [05:34] bzoltan: well... and this time I haven't seen any UITK related failures... but aing tot anything confirmed... [05:39] bzoltan: I'm sorry, today I was stuck with apps [05:39] and tomorrow I'll be on holidays. [05:39] I can help on wednesday. [05:49] elopio: for me that silo16 with this failures http://pastebin.ubuntu.com/7690500/ seem to be tottaly safe [07:19] bzoltan: I struggled to repro as my phone discharges faster than it can charge [07:19] as soon as I do so much as enable wireless :-( [07:59] bregma: hi! Once you're around: I have some minor nitpicks about the packaging changes of libgrip in your silo [08:03] ogra_: hey! I would need a core-dev packaging ACK for some of the main package in the blank powerd policy changes silo [08:04] sil2100, why, isnt that landing from rsalveti ? [08:04] ogra_: yeah, but he's not the author of the commits - you think he reviewed all the branches? [08:05] He's just one of the landers [08:05] if he took care of the silo i would expect so ... but let me take a look [08:05] So I don't have certainty that he was the one reviewing [08:05] ogra_: anyway, diffs: https://ci-train.ubuntu.com/job/landing-020-2-publish/lastSuccessfulBuild/artifact/packaging_changes_powerd_0.16+14.10.20140620-0ubuntu1.diff , https://ci-train.ubuntu.com/job/landing-020-2-publish/lastSuccessfulBuild/artifact/packaging_changes_indicator-power_12.10.6+14.10.20140620.1-0ubuntu1.diff [08:06] The universe components had small changes, recommends additions and etc. which I +1'ed [08:06] These look safish as well I guess [08:08] sil2100, looks fine to me, ACK [08:08] \o/ [08:08] ogra_: thanks [08:09] sil2100, hmm, btw ... powerd = universe [08:09] Oh, powerd as well? [08:09] you could have acked yourself ;) [08:09] I actually only checked indicator-power and assumed powerd is the same ;) [08:09] Packages with power should be in main, as main has power [08:10] haha [08:12] sil2100, ogra_: -1 for indicator-power [08:13] seb128, is it in desktop ? [08:13] unity-system-compositor and powerd are in universe and not on all architectures [08:13] indicator-power is used on desktop and in main [08:13] dang, we need to get rid of that desktop stuff :P [08:14] sil2100, right, cant do that [08:15] :| [08:15] No worries [08:15] seb128: thanks! [08:16] yeah, i would have forgotten about that [08:16] i guess it wants to become a suggests ... on the phone both packages are seeded anyway [08:16] ogra_, sil2100: how can you ack stuff like that? you for sure should know for sure that main can't build-depends on universe [08:17] seb128, its not a build dep [08:17] depends sorry [08:17] binaries in main can't depends on binaries in universe [08:17] but for sure you know that? [08:17] yes [08:18] Yeah... sorry for that, I only took a look at the two other components that were universe-related [08:18] i.e. unity-system-compositor and media-hub changes [08:34] sil2100: We've just been talking about this, I'll do one to move those to suggests [08:34] i-power handles the case when they're not present [08:37] it's also slightly disturbing that this mp is being landed while not approved yet [08:38] rsalveti, kgunn ^ shouldn't the mp be top approved before being put up for landing? [08:43] okay I proposed https://code.launchpad.net/~laney/indicator-power/depends-to-suggests/+merge/224246, please add that if approved [08:43] could anybody configure silo 7 (I cannot as we added a new component) [08:48] on http://ci.ubuntu.com/smokeng/utopic/touch/mako/94:20140623:20140530/8685/gallery_app/ why are some tests listed several times? For example test_camera_button_visible appears 3 times [08:50] ah, nm, it's because the dashboard only shows the name of the test, not the full name with the testsuite [08:52] thostr_: sure [08:54] (I added it myself) [08:55] brendand, could you look why TestPhotosView.test_camera_button_visible passed while it's impossible to reveal the toolbar? [08:56] brendand, I think the test should wait until the toolbar autohides, then reveal it and continue. [08:56] jibel, probably, yes [08:57] sil2100, my morning is packed, so i'll have answers on those failures sometime around lunch [08:58] brendand, gallery app tests should be almost completely red. [08:59] jibel, well if that was the test method then yes they would be i guess [09:00] jibel, it's only because the test sometimes catches it when it's auto-revealed that any pass === vila changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: vila | CI Train Status: #87 promoted | CI Train Sheriff: barry, robru | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. [09:03] brendand: sure :) [09:05] thostr_: reconfigured - when you have some CI train business it's best to say 'trainguards', as all CI Train people will get a ping then ;) [09:06] sil2100: thanks and good to know [09:06] bzoltan: hey! Any possibility we'll have the new UITK sometime soon? The silo is already waiting for a really long time ;) [09:07] sil2100: the silo16 should be about a day old. [09:07] sil2100: the previous one landed as far as I know. [09:08] bzoltan: uh, I don't think it landed, we still get the UITK failure as before which was supposed to be fixed with the UITK landing [09:08] Let me check the commit logs [09:09] bzoltan: hm, you might be right, I guess it got landed during my holidays [09:09] Strange that we still get that failure then [09:09] Maybe it's a different one then [09:11] bzoltan: ok, ignore my doubts, it seems it's a new failure... we'll fill in a bug an poke you later ;) [09:11] Sorry for the confusion [09:11] brendand, ogra_: so, it seems we have a new UITK failure [09:11] huh ? [09:12] oh, true ! [09:13] woah [09:13] OSError: [Errno 8] Exec format error [09:13] that looks like it tries to exec x86 binaries or some such [09:21] sil2100: this one? https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1332162 [09:21] Ubuntu bug 1332162 in Ubuntu UI Toolkit "test_get_unity_top_container failed on mako#89" [Undecided,In progress] [09:23] yeah, looks like it [09:33] Yeah [09:37] feel free to reconfigure/rebuild silo 20 now, indicator-power's fix got approved [09:37] and the original mp too [09:41] vila: can you help? "cp: error writing 'debian/mir-test-tools//usr/bin/mir_unit_tests': No space left on device" - https://jenkins.qa.ubuntu.com/job/generic-mediumtests-builder-utopic-armhf/1964/console [09:41] Laney: thanks! [09:42] np [09:44] alan_g: urgh [09:44] alan_g: let me look [09:58] alan_g: 400MB only for /tmp that's probably where things break, I'll need retoaded to remaster all cyclops nodes, can you reproduce locally to estimate how much you need ? [09:59] vila: mo... [10:01] vila: it's probably an overestimate (as not all are shipped), but the executables and libraries in my local build add up to 450Mb [10:05] alan_g: ack, I'll need to check with fginther and retoaded if bumping the tmpfs to 512MB is reasonable. I'm worried though, those nodes are configured to run up to 2 jobs at the same time so this may not be enough anyway. You may want to investigate if/when you started consuming that much space ? [10:07] sil2100, ping [10:07] vila: better estimate (still an upper bound): 327M [10:08] alan_g: hold on, I read the wrong line it seems, it's not 400MB, it's 2GB already [10:09] * alan_g denies using all that [10:11] alan_g: right, it may be more involved than it looks, those 2GB may be shared across several uses [10:11] alan_g: rebuild triggered at http://s-jenkins.ubuntu-ci:8080/job/generic-mediumtests-builder-utopic-armhf/1970/console [10:12] alan_g: it went on a different node, let see if we can reproduce there [10:12] vila: ack [10:13] alan_g: pfew, previous run took ~2h ? Does that match your expectations ? [10:14] vila: sadly, yes [10:17] If folks must insist on building on feeble kit they could at least use cccache === vila changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | CI Train Status: #87 promoted | CI Train Sheriff: barry, robru | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. === psivaa changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: psivaa | CI Train Status: #87 promoted | CI Train Sheriff: barry, robru | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. [10:21] Our primary builders tend not to benefit much from ccache because they build far too many different things on too many different nodes; I suspect similar is true for Jenkins-based builds [10:22] Maybe not to quite the same extent [10:27] cjwatson: I know there are scaling issues (although there are probably solutions). But it can be frustrating waiting five hours to autoland something that just built in CI. [10:47] mandel: hi! Looking at your u-d-m landing now - did you make sure no one uses the common.h include? === MacSlow is now known as MacSlow|lunch [11:11] sil2100, yes, no one does [11:11] sil2100, 100% sure since it is a not exposed header [11:11] sil2100, nad I recompiled a couple of project to make sure [11:14] mandel: ok, thanks :) [11:22] * sil2100 tries to get his system back to an usable state === pete-woods is now known as pete-woods-lunch === cprov changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cprov | CI Train Status: #87 promoted | CI Train Sheriff: barry, robru | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. === MacSlow|lunch is now known as MacSlow [12:04] ogra_: as most Tuesdays I might not make it for todays evening meeting due to practice [12:04] So could you lead it if I wouldn't be around? === alan_g is now known as alan_g|lunch [12:11] sil2100, sure [12:11] fginther: sil2100 hey guys, gatox is building the first pure click package we ever saw; [12:12] ci is completely busted with this scenario [12:12] fginther: can you make https://code.launchpad.net/~diegosarmentero/pay-ui/pay-backend/+merge/223642 just build clicks and not debs? [12:14] ;) [12:15] sergiusens: ok, that's good to know... [12:15] this one is the latest: https://code.launchpad.net/~diegosarmentero/pay-ui/pay-backend/+merge/224196 [12:17] sil2100, I'm back on the internets again, what was your concern with my libgrip packaging? [12:24] bregma: hi! So, first of all most of the concerns are really minor packaging things but would like to get those fixed before publishing: first of all, you removed some comments and trailing commas [12:25] bregma: the other thing is also that there were build dependencies changed but not mentioned in the changelog [12:26] bregma: same for the removal of python-grip binary package - this should be mentioned in the changelog (i.e. in the respective commit message) as otherwise we will not know easily when it has disappeared [12:29] sil2100, I'm not aware of any build dependency changes, I simply did a wrap-and-sort so I could more easily diff against the Debian packaging (ci-train is boken when it comes to working as a downstream) [12:29] but I can update the changelog to me more clear about deprecating the python- package [12:30] sil2100: hey, could I get a silo for line 27 ? [12:35] ricmm: sure, am having lunch right now [12:36] alright [12:37] ricmm, hey! [12:38] ricmm, we need you, see -touch, the qtubuntu-sensors made unity8 desktop unhappy [12:40] gah. where's the new ci-train spreadsheet? I've only got links to the old one [12:46] https://wiki.ubuntu.com/citrain [12:47] cjwatson: meta-thanks [12:48] sergiusens, I have work for that in progress. [12:49] fginther: I setup something to sync on a hangout for later today [12:49] sergiusens, thanks [12:51] fginther: of interest would be to be able to build fats [12:51] :-) [12:51] it's winter here, so fat is good :-P [12:52] sergiusens, :-) === renato_ is now known as Guest77133 === barry changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cprov | CI Train Status: #87 promoted | CI Train Sheriff: robru | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. === alan_g|lunch is now known as alan_g [13:06] sil2100, around? [13:06] tvoss: sure [13:07] sil2100, just letting you know I've updated my libgrip d/changelog, I'm going to rebuild silo 006 to pick it up [13:08] hiya peeps. Could I have a silo for row 28 please? [13:09] davmor2, wow ... i know what you meant yesterday with the google logo ... the first boot on 95 takes several centuries ... [13:09] i wonder why [13:09] ogra_: :) [13:09] ogra_: the bit after the google logo took about the same amount of time [13:10] ogra_: It was just the waiting for the google logo to go that took forever [13:11] davmor2, the second boot is fine ... i assume the apparmor profile generation on first boot got slower [13:12] stopwatching the second boot i get 10-12sec with the google logo and 11-13 for the boot animation ... latest after 25 sec i have the UI on screen [13:12] so i guess we can blame apparmor here [13:14] ogra_: :) [13:15] mterry, nice first boot wizard ! [13:15] ogra_, yay! [13:15] mterry: not so nice if you have a different back drop though [13:16] davmor2, backdrop? You mean if you've changed the greeter wallpaper? [13:17] mterry: Yes if you upgraded to the image and had change the greeter to one that is mostly black and white the text is really hard to see [13:17] change to greeter wallpaper even [13:17] davmor2, ah fair. That's a temporary issue during upgrades though. Most users will only see this the first time they boot [13:18] sil2100: hey, any chance to get the silo for27 ? [13:18] mterry: but your assumming that oem's won't change the backdrop though ;) [13:18] davmor2, no, but if they do, they'll see the effect on the wizard [13:18] true [13:18] * jdstrand notes that server side generation of profiles on the default image should be landing this week [13:19] (as I'm told) [13:19] ricmm: let me take a look, one moment [13:19] thank you [13:19] jdstrand, oh, cool !! [13:20] sil2100: hey there, so we tested silo 20 & marked it "Yes" for tested so it would be released...i just checked this morning [13:20] ricmm, tvoss: ok, so both of you have platform-api and qtubuntu-sensors [13:20] and looks like someone marked it "no" [13:21] any ideas? [13:21] kgunn: yeah, so... [13:21] kgunn: I wanted to publish that but then it was -1'ed for release [13:21] sil2100: which line is tvoss' [13:21] ? [13:21] ricmm, 11 [13:21] ricmm: silo 11, line 22 - it's a 'test silo', but tvoss mentioned it lanind this week [13:21] robru: does you being sheriff mean requests for silos go to you? [13:21] kgunn: ok, let me continue my explaination [13:22] * kgunn was being patient :) [13:22] tvoss: do you mind letting me through? need to fix a crasher for unity8 on desktop [13:22] ricmm, fine with me [13:22] Chipaca, it means you can shoot him and make a song about it [13:22] ;) [13:22] tvoss: my silo is ready to land as soon as someone helps me test [13:22] tvoss: thanks [13:22] sil2100: ^ [13:22] kgunn: so, during packaging ACKing we noticed that some main packages suddenly depended on universe packages [13:22] kgunn: which is trouble [13:23] kgunn: so Laney devised a fix, we reconfigured and rebuilt ;p [13:23] ricmm: ACK [13:23] kgunn, worse was that the MP wasnt top approved yet though ... [13:23] (indicator-power) [13:23] it is all sorted now ... but i guess you want to re-test [13:24] ricmm: assigned! [13:24] sil2100, CI should really refuse to land stuff if there is no top approval [13:24] ogra_: https://www.youtube.com/watch?v=GCY_GKX09AE [13:24] sil2100: ogra_ ok, will retest....and i'll follow up on the mp approval [13:24] ogra_: hmmm... good idea! [13:25] Chipaca, hah [13:26] sil2100: thanks! [13:26] ogra_: will work on that today I guess :) [13:26] sil2100: back to good old silo 5 :) [13:26] sil2100, ++ [13:31] sil2100: robru: I'm unsure who i'm supposed to ping to ask for a silo, but i'd like one for row 28, pretty please [13:31] Chipaca: hi! Let me take a look [13:31] sil2100: hi :) [13:31] Chipaca: ah! Ok, so, if you want a silo you need to switch the 'Ready' field to Done [13:31] Otherwise we don't consider a landing ;) [13:31] So it could just sit there forever ;p [13:34] sil2100: done! [13:34] sil2100: I have to ask for it anyway, once it's ready? or does setting it to ready get me a silo automagically? [13:35] Chipaca: we have to give a silo ourselves, but once it's set to ready we either see it ourselves when scouting the spreadsheet or just get pinged by the bot that someone added a ready landing [13:35] So no need to poke, unless there's something important to mention ;) [13:39] I'll get the hang of it someday [13:40] No worries! [13:51] sil2100, i can't reproduce either the weather or filemanager issue [13:51] sil2100, i'll file bugs for both of them to keep track [13:53] kgunn: ogra_: I can help landing 20 as well, not sure if everything got properly reviewed and top approved [13:53] but good that it was tested by qa already [13:53] rsalveti, its all fine now [13:53] well, we still need to land it :-) [13:54] yep ... but the other issues are fixed [13:57] seb128: it's fine to start a silo with mrs that are not yet top approved [13:57] seb128: the only problem is landing such silo in that state [13:57] rsalveti, right [13:58] rsalveti, well, somebody asked for packaging review on that one, so I assume it was about to land [13:58] rsalveti, not to mention that it was adding universe depends to a package in main, and things we don't want on the desktop (like powerd and the system compositor) [13:59] rsalveti, anyway, was an easy one, the code handle those not being there, Laney demoted them to suggest, so should be all good ;-) [13:59] brendand: ok, thanks! [13:59] seb128: right, but it wasn't ready for landing yet [14:00] seb128: the testing flag was set to yes [14:00] and that's why I believe sil2100 tried to land it [14:00] but not every mr was reviewed/approved yet [14:00] k [14:00] maybe the issue was that the testing was set to done when the work wasn't really done [14:00] yeah [14:00] lander assume they can land stuff that are "tested: yes" [14:00] they take it as "ready to land" [14:00] right [14:00] Right, it was set to tested -> done, as kgunn said he tested it and switched that to yes [14:01] right, that was the issue then [14:01] I'm changing CI train to not allow (without an override) to build unapproved branches [14:02] right [14:06] bregma: oh my, the changelog looks soo sassy! === pete-woods-lunch is now known as pete-woods === cprov changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: Ursinha | CI Train Status: #87 promoted | CI Train Sheriff: robru | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. [14:17] sil2100: sassy changelog? do share! [14:19] * kgunn wonders if changelogs can be sassy/ [14:20] sil2100: can i get a silo for line 29 ? ....all the mp's are approved ;P [14:21] kgunn: hehe ;) Sure, sorry, was inside some packaging work [14:22] i'm sorry about that mp slipping thru... [14:22] i think it might have been a late add....that was a rather dynamic list of mp's [14:22] kgunn: remember that you have another unity-mir silo assigned for testing of the trusted sessions [14:24] sil2100: yeah, that trusted sessions is only for testing right now...so [14:24] sil2100, can I get a silo for line 30? [14:24] they one i'm asking for is for landing [14:25] tvoss: set it to Ready yes! :) And assigning [15:00] sil2100: I think there is an issue with the osk and scopes on the latest flo I'm going to reboot and see what happens then [15:11] davmor2: what seems to be the problem? [15:12] sil2100: mterry: I think after a boot with the welcome screen the osk is screwed up a bit. It doesn't appear properly till you reboot the device then it works fine. I'll do a fresh install to confirm and grab some logs [15:14] davmor2, hrm [15:14] sil2100, hey, could we get a silo for line 4? I promise, we'll land it this time! [15:14] davmor2, thought I tested that -- we should be killing maliit-server when done with wizard [15:15] Saviq: ! [15:15] mterry: yeah that might be the issue it isn't coming back up :) [15:15] Saviq: ok, but I already see some of the projects there are locked currently [15:16] sil2100, I'll deal with that as needed [15:16] mterry: shame on youuu! Killing things is a dirty workaround ;) [15:17] sil2100, maliit-server is tied to the current MIR_SOCKET server, and we'd be swapping it out on them [15:17] Excuses! [15:18] Saviq: ok, just make sure to coordinate [15:18] sil2100, will do [15:19] sil2100: dunno if you know this, but on #95 music app is missing and i can't bring up osk in the app lens [15:20] i see davmor2 has confirmed this [15:20] popey, ugh [15:20] popey, we dropped mediascanner 1 last night [15:20] right, but music-app is a click [15:21] music-app shouldnt use it anymore though [15:21] it doesnt [15:21] phew [15:21] ogra_: popey I see the music app [15:21] * ogra_ isnt near his flo atm ... cant check [15:22] hm, i see it on my flo [15:23] * popey reboots both makos [15:23] popey: I see it on my mako [15:23] we got a new apparmor ... probably it missed to re-create the profile ? [15:23] hmmm its on my other mako [15:23] maybe yes [15:23] * popey reboots [15:23] nope not showing [15:23] * popey gets a cable [15:24] Saviq: uh oh [15:25] Saviq: ok, it took longer since there was an invalid MR in the MR list, fixed that and reassigned [15:29] phablet@ubuntu-phablet:~$ click list | grep music [15:29] com.ubuntu.music 1.3.496 [15:29] hmm [15:30] phablet@black-phablet:~$ click list | grep music [15:30] phablet@black-phablet:~$ [15:31] yeah, it doesn't show up on my other device [15:31] weird [15:31] it is in /opt/click.ubuntu.com/com.ubuntu.music tho [15:31] I'll uninstall it and see if i can install from store [15:31] oh, i cant can I ☻ [15:32] you should be able to uninstall from cmdline [15:32] how? click doesn't think i have it [15:32] oh, hmm [15:32] did you search for it ? [15:32] yes [15:32] it should offer it in the store if it really thinks that [15:34] mterry: also phone settings on a tablet doesn't sound right for the disabling of error reporting :) [15:35] * sil2100 prepares to drive to practice [15:35] o/ [15:35] davmor2, that screen/text is going to be changed [15:36] mterry: good to know :) [15:41] mterry: okay I have a keyboard but it took nearly a minute to appear [15:41] mterry: screen had dimmed so is that 45seconds [15:44] davmor2, Bug marked Invalid, user is too impatient ;) -- I've marked a note down for that [15:56] sergiusens, this is the error I'm currently getting: http://paste.ubuntu.com/7695713/ [15:56] sergiusens, missing package :-( [15:56] fginther: I think that's for xnox :-) [15:57] hey. [15:58] fginther: click from utopic is needed. utopic's version is compatible to install on previous releases (e.g. trusty) [15:58] ahh [15:58] xnox, thanks [15:58] fginther: and it has been copied from utopic into sdk ppa or some such. [15:58] bzoltan: ^ where did you publish fixed click? [15:58] ogra_, is there a reason we don't have two builds per day? [16:02] fginther: oh, click is update in the sdk ppa [16:02] xnox: should be the sdk ppa [16:02] sergiusens, ack [16:03] fginther: how are you calling click? I want to copy paste :-) [16:05] sergiusens, "sudo click chroot -aarmhf -fubuntu-sdk-14.10 create" [16:12] sil2100, ugh, sorry :| [16:16] Chipaca, oops, sorry. I forgot to take myself off the Sheriff topic when I EOD'd last night [16:16] stgraber, you around? it's your day to be sheriff [16:17] robru: I'm around, not in the call though due to the TB meeting and my lunch break both happening at the same time :) [16:18] stgraber, ok, no worries [16:19] we didnt have much to discuss in the call anyway ... you didnt miss anything === robru changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: Ursinha | CI Train Status: #87 promoted | CI Train Sheriff: stgraber | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. === Ursinha changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | CI Train Status: #87 promoted | CI Train Sheriff: stgraber | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. [16:25] stgraber, can you check the packaging in silo 6? big diff there https://ci-train.ubuntu.com/job/landing-006-2-publish/28/artifact/packaging_changes_libgrip_0.3.8+14.10.20140624-0ubuntu1.diff [16:27] robru: I should be able to look at this in about an hour [16:28] stgraber, alright, thanks === fginther changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: fginther | CI Train Status: #87 promoted | CI Train Sheriff: stgraber | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. [16:39] xnox: in the SDK PPA [16:43] sil2100, silo13 is good to go [16:45] stgraber, ^ [16:46] robru: sure, will look once I'm back from meetings and lunch. [16:47] stgraber, thanks [16:58] seb128: hey did you test the silo? [17:02] seb128: because I dont see it crashing, but its aborting somewhere in Mir for me === alan_g is now known as alan_g|EOD [17:24] seb128: I no longer see it crashing but its failing to connect to the mir server instance... not sure what could be a cause for that [17:28] robru, while you're around, seen bug #1333184 ? [17:28] bug 1333184 in phablet-tools (Ubuntu) "`citrain host-upgrade` removes the PPA after installation" [Undecided,New] https://launchpad.net/bugs/1333184 [17:38] Saviq, yeah, I put that in on purpose originally, didn't realize it broke ppa-purge. not sure, ppa-purge is weird, sometimes it can remove the ppa and sometimes it can't. haven't figured it out yet [17:38] Saviq, although I guess it's just one line to drop from citrain tool to stop deleting the ppas after installation [17:50] seb128: I'll assign you a silo for that indicator-network landing, however note that in the future you can also just get one yourselves (as coredevs have the needed Jenkins ACL now) [17:50] *yourself === gatox is now known as gatox_lunch [17:59] robru, it will never be able to remove a ppa that it has no deb line for [17:59] robru, because it needs to find out which packages are there to be able to know to backport them [17:59] Saviq, maybe I'm confused but that code was working for me at the time that I wrote it ;-) [18:00] robru, it could probably read that from dpkg db or something, but obviously it doesn't :) [18:00] Saviq, yeah, but you tell it what ppa you want to remove, i thought it could query that ppa even if it wasn't currently enabled [18:00] robru, right, well, it doesn't :) === fginther changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | CI Train Status: #87 promoted | CI Train Sheriff: stgraber | Known issues: The spreadsheet is extremely slow, http://people.canonical.com/~rbpark/citrain/ knows the statuses before the spreadsheet does. [18:13] Saviq, https://code.launchpad.net/~robru/phablet-tools/dont-delete-silos/+merge/224340 if you want to review [18:13] sergiusens also if you're around ^ [18:14] robru: I'm around === gatox_lunch is now known as gatox [18:38] * sil2100 goes to prepare the e-mail [18:58] ogra_: anyway, I think we can get back to our previous approach of building 2 images a day [18:58] sil2100, yeah [18:59] no issue with that [19:02] ogra_: so, I have a family house-work emergency tomorrow, so I have to drive to my parents place and help out finishing some wall-painting, so sadly I'll be most of the time in and out ;/ [19:02] ogra_: could I ask you one last time to help out and take the lead of the meetings for me? [19:02] I promise it'll be the last day like this ;) [19:03] I'll be around on IRC when resting inbetween house-work [19:05] robru, can you please recon silo 4? I added dialer-app (yes, we know there's a bunch of conflicts, I'll coordinate) [19:06] Saviq, ok [19:09] robru, thanks [19:38] Saviq, you're welcome! [19:59] ricmm, sorry, it was not built and then I was away for a while, trying in a bit [20:00] stgraber, hey, thanks, I know I can technically assign silos, I just though the CI team was still supposed to do it because they have a better view of the current image status and know if things should flow in or be on hold for a bit [20:06] seb128: nah, so long as we're in regular landings mode, there's no problem with coredevs just doing everything themselves [20:12] stgraber, where did you get that? I saw no communication on that [20:13] stgraber, even the fact that coredev can assign silo was not communicated, Laney and I learnt it earlier this week by lurking on this channel where you or somebody mentioned it [20:14] seb128: it was one thing we discussed with slangasek and robru when barry and I went through the landing team training [20:14] seb128: basically the landing team is now under foundation and so we're changing a couple of small bits to the process here and there to get things go a bit smoother [20:14] well [20:15] if that's true, could that be communicated out of the foundation team? [20:15] I think it becomes a problem if all coredevs start claiming silos without coordination [20:15] right [20:15] I don't think it was ever said that coredevs should just claim their own silos [20:15] true, but you need to be more than coredev to have the necessary rights [20:15] which is why I was assuming the default was to let the CI lander do the "orchestration" [20:15] you need coredev+spreadsheet edit rights [20:15] well, quite a bunch of us have that [20:15] which leads to the same issue [20:16] slangasek: robru did say during that call that he didn't see any problem with coredevs just assigning their own silos [20:16] until we run out of available silo [20:16] right. if there are not enough silos, then you just have to wait [20:16] stgraber: right, and it's fine that he thinks that, but it's not something I had any intention of socializing more broadly because silos are a limited resource :) [20:16] if we do that we should have a proper and documented workflow [20:17] slangasek: sure, which is why I haven't sent an e-mail about it and instead just told people who'd benefit most from it and are known to be reasonable :) [20:17]