/srv/irclogs.ubuntu.com/2014/05/06/#ubuntu-ci-eng.txt

imgbot=== trainguard: IMAGE 13 building (started: 20140506 02:10) ===02:09
imgbot=== trainguard: IMAGE 13 DONE (finished: 20140506 03:45) ===03:44
imgbot=== changelog: http://people.canonical.com/~ogra/touch-image-stats/13.changes ===03:44
Mirvmorning04:02
Mirvdidrocks: there'd be autopilot python 3 ports needing packaging ack https://ci-train.ubuntu.com/job/landing-015-2-publish/14/artifact/packaging_changes_camera-app_2.9.1+14.10.20140505-0ubuntu1.diff + https://ci-train.ubuntu.com/job/landing-015-2-publish/14/artifact/packaging_changes_notes-app_1.4+14.10.20140505-0ubuntu1.diff06:26
Mirvthey don't change the packaging identically, but also the one not changed by x_nox was approved by him06:27
Mirvnamely the more thorough one uses dh-python and the other one doesn't06:28
didrocksMirv: so, you think they did test it?06:29
didrocks(if so, +1)06:30
Mirvdidrocks: yes they marked it as being tested an hour ago, while it was built 15h ago06:32
Mirvthanks06:32
MirvI was wondering if there was any discussion about the new file manager's AP failures07:30
=== psivaa changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: psivaa | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: spreadsheet problems, do not trust CI-SNCF bot
Mirvpsivaa: was there some status update on the file manager failure? testsuite name ubuntu_filemanager_app -> filemanager08:03
psivaaMirv: is this in relation to the smoke failure?08:03
Mirvpsivaa: regarding http://ci.ubuntu.com/smokeng/utopic/touch/mako/13:20140506:20140506/7883/ubuntu_filemanager_app/ and I found out after updating my click test setup that the suite changed name apparently08:04
psivaaMirv: i see'could not import package ubuntu_filemanager_app: No module named ubuntu_filemanager_app' and remember that we had to change that in ci/autolanding a week or so ago. it could be the reason08:06
psivaaMirv: let me re-run that test with 'filemanager'08:06
Mirvpsivaa: image #9 saw update of filemanager to .169 and bzr166 renamed ubuntu_filemanager_app -> filemanager08:07
sil2100I see a commit for that in trunk indeed08:08
sil2100Too bad no one coordinated this with anyone08:08
psivaayea, 10,11,12 & 13 have this issue. running it with 'filemanager' now and tests are running08:12
Mirvpsivaa: thanks! 9 would have had it also, but it seems the whole test suite was not run there :S08:13
MirvI did get 3 failing tests myself locall, let's see if it's flaky or happens also on test infra08:13
psivaaMirv: yea. not sure what happened there.. just working my way backwords :)08:13
Mirv:)08:13
ToyKeeper... and after dealing with a surprise critical bug, I'm out of time for the day.  Lots of interrupts over the past few days.08:23
ogra_ToyKeeper, hopefully a desktop bug ?08:33
ogra_(or do we have to worry ? )08:33
ToyKeeperNo, and I don't think so.08:35
ogra_phew :)08:38
popeyOn #12 (I can't upgrade to 13 for some reason - probably network) if I open system settings I get a blank screen until I switch away or rotate the screen. anyone else see that?08:50
ogra_popey, i saw it going blank during download yesterday (on flo)08:56
popeynot the thing where the content disappears08:57
popeythe entire screen (indicator aside) is blank, looks like it didnt start08:57
ogra_ah, havent seen that yet09:05
=== vrruiz_ is now known as rvr
davmor2Morning all09:58
davmor2popey: I had to reset the networking on 12 this morning and then the upgrade to 13 was fine09:59
popeyreset networking? as in flip the switch off/on?10:00
ogra_davmor2, http://paste.ubuntu.com/7376744/ ... put that file in place (make it executable) ... that is the proposed workaround for the routing bug10:00
davmor2popey: yes10:01
ogra_mount -o remount,rw / ... put file in place ... mount -o remount,ro /10:01
ogra_i'm running with that here and have not seen routing issues since10:01
davmor2ogra_: let me do the dogfooding for 13 first so I can see how impacting it is10:02
ogra_ok10:02
ogra_popey, ^^^ for you too ... we need as much testing as possible ...10:02
davmor2popey, didrocks: So I'll be off tomorrow it looks like Sue has made appointments for nearly everything tomorrow so her Son and I only lose a day for arrangements and then one for the funeral.10:08
didrocksdavmor2: ok man, take care of her10:09
davmor2Yeah I am :)ssss10:09
Mirvdbarth: how's the plan on getting HTML5 apps using Oxide instead of QtWebKit? I'm wondering when I could allow the 5.2 QtWebKit in without fearing for regressions on that front.11:10
davmor2Mirv: has anyone let the app developers know?11:23
Mirvdavmor2: I don't know, a good question. the assumption is probably that nothing changes for worse at least when moving to Oxide.11:29
davmor2Mirv: ah nice okay11:30
Mirvapps using webview and 14.04 framework already use Oxide from what I understood, and after the SDK HTML5 apps switch the only remaining qtwebkit user would be those apps using 13.10 framework11:30
Saviqdidrocks, hey, any idea why https://ci-train.ubuntu.com/job/landing-002-1-build/49/console is waiting for arm64?11:53
davmor2popey: if you send a message to your phone while it sleeps do you get a notification tone for it?11:54
MirvSaviq: it looks like it's currently available on arm64: https://launchpad.net/ubuntu/+source/unity-system-compositor/0.0.2+14.10.20140430.1-0ubuntu111:54
sil2100Saviq: hi, it seems u-s-c is built for arm64 in utopic11:54
didrocksyeah, it's built on it11:55
Saviqhmm ok /me looks why the dep wait11:55
Saviquh oh11:55
* Saviq got firefox 2911:55
ogra_rounded tabs11:57
Saviqyeah, doesn't look nice at all :|11:57
* ogra_ doesnt care as long as it renders websites correctly 11:57
SaviqMirv, u-s-c in that silo depends on libandroid-properties-dev, which are not available on arm64 :|11:58
davmor2didrocks: there is some odd behaviour with the dialer app and recalling a number.  Let me reboot and see if I can reproduce it11:59
ogra_i thought i saw a bug about that ...12:00
MirvSaviq: could the dependency be avoided, isn't u-s-c supposed to be used on desktop? now all the arm64 desktoppers would lose Mir goodness! (yes, we seem to build also Mir on arm64 which is nice)12:00
davmor2Saviq: don't look at the tabs then ;)12:01
ogra_bug 130836512:01
ubot5bug 1308365 in dialer-app "press-call-twice-to-redial doesn't work" [Undecided,Confirmed] https://launchpad.net/bugs/130836512:01
SaviqMirv, good question, I'm not really sure where that dependency comes from...12:01
davmor2ogra_: no not that one :)  That is a feature request12:01
=== cjohnston changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cjohnston | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: spreadsheet problems, do not trust CI-SNCF bot
* ogra_ wonders why pad.lv doesnt seem to load for him12:02
dbarth_Mirv: not there yet, more blocked on webapps mtaintenance right now12:02
dbarth_Mirv: maybe we can list that as a sprint goal12:03
=== alan_g is now known as alan_g|lunch
Mirvdbarth_: ok12:04
Mirvdbarth_: osomon would have a new webbrower-app request again. but there's still "Add UA override feature, per webapps" landing there, should you land it first?12:04
dbarth_Mirv: i can put this one on hold, nw12:05
* ogra_ thinks this is less impartant since the UA seems to work properly with oxide now ... 12:05
Mirvdbarth_: so postponing landing-003 again? ok.12:05
ogra_that feature will only be used in corner cases now ... while it was more important with webkit12:06
dbarth_ogra_: this is to let each app tweak it's own, but agreed, less important12:06
dbarth_its12:06
ogra_dbarth_, i knwo what it does, its my bug ;)12:06
ogra_(or is supposed to do)12:06
SaviqMirv, oh, found the culprit, https://code.launchpad.net/~unity-team/unity-system-compositor/new-gl-screen/+merge/210466, which is included as a prerequisite - it's about the boot splash... will follow up with mterry and Mirco12:07
ogra_dbarth_, btw, is anyone looking at browser activity fixing ? tabs/bookmarks etc dont work at all12:07
MirvoSoMoN_: landing-005 it is, unless you want to join the two landings into one12:07
ogra_(i assume that waits for the new bottom edge stuff ?)12:07
oSoMoN_Mirv, thanks, two separate landings will do12:07
Mirvthen it's good.12:07
davmor2ogra_: knock your phone onto standby, ring it, and hang up so you get a missed call notification in incoming, hit the call back button dialer app seems to go to the calling dialogue then when you answer it shows the call log here12:08
davmor2so you can't hang up12:08
oSoMoN_Mirv, by the way, I keep pinging you or sil2100 randomly about landings, is that ok or do you guys have a schedule that I should respect to not interrupt you all the time?12:09
davmor2you have to close the app and then reopen it12:09
ogra_davmor2, wont work with devel ... the event bock bug is still present12:09
ogra_*block12:09
dbarth_ogra_: oSoMoN_ is, but not sure that can be made ready; the top header redesign may come first12:09
* ogra_ hates the new top header with passion already :P 12:10
dbarth_we'll have our oxide catch up later today, so i'll ask12:10
ogra_dbarth_, oSoMoN_ well, if we cant fix it before going to market we should probably remove the activities button12:10
* ogra_ guesses we have lless than two months before we have to take the archive snapshot 12:11
oSoMoN_ogra_, can you elaborate on "don’t work at all"?12:12
ogra_oSoMoN_, tap the activities button ... nothing happens12:12
davmor2ogra_: should work12:12
oSoMoN_ogra_, if that didn’t work, then we would be seeing autopilot failures in the daily runs12:12
ogra_oSoMoN_, oh, trying it now it got even worse ... s/nothing happens/browser freezes/12:13
davmor2ogra_: what app has the new topbar header in it12:13
ogra_oSoMoN_, feel free to try it :)12:13
oSoMoN_ogra_, trying now12:13
ogra_davmor2, only the scopes yet, but it gives you an impresssion about the back button placement ... and if the header is permanent it will steal your screen space for content12:14
ogra_oSoMoN_, stopped for me when we switched to oxide ... its quite an old bug12:15
davmor2ogra_: and in most webapps they already have a header so then you get 2 so top half of the screen is headers right12:15
oSoMoN_ogra_, did you report it? it sounds like a pretty bad regression, if it was confirmed12:15
ogra_davmor2, right, and even in native apps we will steal 10-20% of the screen space from the content12:15
ogra_oSoMoN_, no, i thought it was in the works still and just a brawback of the switch to oxide ... i have the bad tendency to just turn my bookmarks into webapps anyway, so i stopped using bookmarks much  :)12:17
ogra_if you can confirm it i'll file it right now12:17
oSoMoN_ogra_, it’s definitely a bug, and not an expected regression12:17
davmor2ogra_: I could see it being better if it disappeared like the bottom bar and then just a tap in empty space or drag the app down brought it back maybe12:17
oSoMoN_I’m updating to the latest image, will confirm in a moment12:17
ogra_if you cant, there must have something gone wrong when transitioning my settings from webkit to oxide ...12:17
ogra_which is a bug as well, but probably less severe12:18
ogra_davmor2, right, but that wont be possible ... dragging from the top is reserved for indicators ...12:18
ogra_so i assume th only option is to have it permanently visible12:19
davmor2ogra_: no I mean drag down on the app like the gallery app when you scroll up the header disappear scroll back down and it comes back12:19
ogra_ah12:19
davmor2ogra_: so thumb in the middle of the screen and drag down slightly reveals the window chrome once you tap it or scroll up it disappears again12:21
oSoMoN_ogra_, I can’t reproduce here, would you mind inspecting the logs for webbrowser-app to see if there is anything relevant when you try to open the activity view?12:21
ogra_ah, yeah, that might work12:21
ogra_oSoMoN_, nothing at all ... as if i had not tapped the button12:23
ogra_but the browser is hard locked again12:24
oSoMoN_and no logs?12:24
ogra_nope12:24
oSoMoN_huh, that’s weird12:24
ogra_http://paste.ubuntu.com/7404267/12:24
MirvoSoMoN_: pinging is fine, I just stop answering soon and then sil2100 will switch to robru/cyphermox later on12:24
ogra_thats what it rpints from start of the app to the point where i tap12:25
oSoMoN_Mirv, ok, thanks12:25
oSoMoN_ogra_, could it be that you have a huge history db + bookmarks db and that it fails to populate the model because of too much data, maybe?12:25
oSoMoN_ogra_, can you share them with me, by any chance?12:26
oSoMoN_~/.local/share/webbrowser-app/[bookmarks|history].sqlite12:26
ogra_oSoMoN_, where do i find them ... this phone runs devel since and of saucy ... could be that a few lines of history and bookmarks piled up over that year :)12:26
ogra_ogra@styx:~/Devel/apps/book-app$ adb shell ls -l ~/.local/share/webbrowser-app/bookmarks.sqlite12:27
ogra_ls: cannot access /home/ogra/.local/share/webbrowser-app/bookmarks.sqlite: No such file or directory12:27
ogra_ogra@styx:~/Devel/apps/book-app$ adb shell ls -l ~/.local/share/webbrowser-app/history.sqlite12:27
ogra_ls: cannot access /home/ogra/.local/share/webbrowser-app/history.sqlite: No such file or directory12:27
ogra_ogra@styx:~/Devel/apps/book-app$12:27
ogra_hmm12:27
ogra_are you sure that is the right path ?12:28
ogra_err12:28
ogra_lol12:28
ogra_funny ... so ~ gets expanded12:28
ogra_never had that with adb before12:28
ogra_ogra@styx:~/Devel/apps/book-app$ adb shell ls -lh /home/phablet/.local/share/webbrowser-app/history.sqlite12:29
ogra_-rw-r--r-- 1 phablet phablet 149K May  6 14:23 /home/phablet/.local/share/webbrowser-app/history.sqlite12:29
ogra_ogra@styx:~/Devel/apps/book-app$ adb shell ls -lh /home/phablet/.local/share/webbrowser-app/bookmarks.sqlite12:29
ogra_-rw-r--r-- 1 phablet phablet 2.0K Dec 19 17:09 /home/phablet/.local/share/webbrowser-app/bookmarks.sqlite12:29
ogra_ogra@styx:~/Devel/apps/book-app$12:29
ogra_not actually big12:29
ogra_let me pull them12:29
ogra_oSoMoN_, http://people.canonical.com/~ogra/browser/12:31
oSoMoN_ogra_, thanks, I’ll run some tests with them after lunch12:31
ogra_let me try to move them away and see if that changes behavior here12:32
ogra_oh !12:32
ogra_it came up ... the phone was lying there for like 10min and now it just popped up the activities page12:33
ogra_so it works ... just takes ages12:34
ogra_and the app is completely unresponsive during that12:34
ogra_oSoMoN_, moving the dir away makes it work12:36
cyphermoxogra_: good morning! how's the network? :)12:42
ogra_cyphermox, steady :)12:44
ogra_cyphermox, i'm waiting for davmor2 to test too since he is more roaming around etc :)12:44
ogra_from my pov it is fixed12:44
ogra_but i dont swithc much between 3G and WLAN12:44
cyphermoxright12:45
* didrocks goes for a run12:47
popeyogra_: gonna go for a walk, will test12:55
ogra_popey, thanks !12:55
ogra_i actually need to mow the lawn today ... i should perhaps just do that now so it roams a bit12:56
=== alan_g|lunch is now known as alan_g
davmor2didrocks: so 13 looks to be on par with most of the other recent images I'm just writing up one bug for the dialer app on returning a missed call13:09
sil2100davmor2: you think that's a blocker?13:09
davmor2sil2100: I think it should be high to get it fixed asap.  However I'm going to install R2 and see if it exists there and was just missed previously, before it is classed as a blocker.13:12
davmor2sil2100: however I'm going to file the bug first :)13:12
sil2100davmor2: ok, thanks ;)13:14
popeyogra_: seems way better13:21
ogra_popey, way better or fixed ?13:21
popeywell, fixed13:22
ogra_yay13:22
ogra_cyphermox, ^^^13:22
cyphermoxcool13:22
cyphermoxship it!13:22
ogra_yeah13:22
oSoMoN_ogra_, right, so the issue is that it’s slow and unresponsive, which is not surprising given how it’s implemented, I’ll keep an eye on performance when I implement the new activity view13:23
ogra_right, it works fine with a new/empty DB13:23
popeygot bug 1309556 again while I was out ☹13:24
ubot5bug 1309556 in camera-app (Ubuntu) "camera crashed and took phone with it." [Undecided,New] https://launchpad.net/bugs/130955613:24
ogra_i assume we should probably create some bookmarks for the test runs to catch such issues13:24
ogra_and ship a pre-filled history DB in the autopilot package or so13:24
oSoMoN_ogra_, sounds like a good idea, noted down13:25
oSoMoN_ogra_, running webbrowser-app on a beefy desktop with your db, the activity view takes 6secs to show up, and it’s not smooth when flicking, so I think we have our culprit13:29
ogra_yeah13:29
ogra_its quite snappy here with the removed db13:29
oSoMoN_even if we optimize the model loading, we should also probably limit the number of entries, who cares about their browsing history a year ago?13:29
=== oSoMoN_ is now known as oSoMoN
ogra_yeah13:31
ogra_a week should be enough ... or even less13:31
cyphermoxogra_: which ppa for the fix?13:37
ogra_cyphermox, a silo i'd say13:37
ogra_i'll need to update the mms silo again then ...13:38
cyphermoxwell yeah, that's what I meant13:38
ogra_(since that has another lxc-android-version already)13:38
cyphermoxdon't we want to ship those separate?13:38
sil2100Is there a silo with the route workaround already?13:41
oSoMoNsil2100, would you mind publishing silo 5? I just completed testing for it13:49
sil2100oSoMoN: sure, just doing lunch right now13:56
sil2100But let me take a look13:56
oSoMoNsil2100, there’s no urgency, enjoy your lunch and have a look afterwards :)13:56
sil2100didrocks: I would need 2 packaging ACK from you :)14:01
sil2100Both look rather good14:01
sil2100didrocks: first -> https://ci-train.ubuntu.com/job/landing-005-2-publish/lastSuccessfulBuild/artifact/packaging_changes_webbrowser-app_0.23+14.10.20140506-0ubuntu1.diff <- python3 transition14:01
oSoMoNah, I was about to ask who could ack the packaging changes, sil2100 has been faster :)14:01
sil2100didrocks: second -> https://ci-train.ubuntu.com/job/landing-014-2-publish/12/artifact/packaging_changes_gallery-app_0.0.67+14.10.20140505-0ubuntu1.diff <- python3 transition ;p14:02
sil2100oSoMoN: ;) No worries, I simply have to pre-scan it for anything strange before passing it on to didrocks or some other core-dev14:02
oSoMoNit makes sense14:02
didrockssil2100: the transition has been tested server-side as well?14:18
didrockssil2100: if so, +114:19
didrocks(for both)14:19
sil2100didrocks: by server-side, you mean on CI infra?14:19
didrocksyeah14:19
sil2100didrocks: let me ask around about that14:19
didrocksok14:19
didrockson packaging itself, +1 though14:19
sil2100oSoMoN: ^ did you coordinate it with the CI guys and the CI infrastructure, if it won't suddenly break every test from that application when running, for instance, smoketesting?14:20
sil2100oSoMoN: phablet-test-run still works and such, yes?14:20
sil2100bfiller: ^ same regarding your landing14:21
bfillersil2100: QA verified the AP tests all worked on the device14:21
oSoMoNsil2100, yes, I’ve run the tests on my device and desktop, and CI runs pass in the MR itself, so I think we’re fine14:22
oSoMoNalso, there are other apps that have already migrated to python3 for their autopilot tests, and they seem to work14:23
oSoMoNsil2100, this is a CI run for the py3-autopilot branch on mako: https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner-mako/6547/?14:25
sil2100plars: hi! Are you around?14:27
plarssil2100: no14:27
sil2100:D14:27
plars:)14:27
plarssil2100: what's up?14:27
sil2100plars: just want to make double sure - there are some apps migrating to py3 now, dropping py2 support - do you know if we already run some apps as py3 in smoketesting?14:28
plarssil2100: I saw that, there are a few that there might be concerns with14:28
plarssil2100: we use phablet-test-run for all the autopilot tests, which aiui handles that automagically.14:29
plarssil2100: so if the current autopilot tests have been ported to py3, then it should already be running them with it14:29
plarssil2100: there's a hack in place that handles making sure the custom tests get run with autopilot3 (even though they apparently don't even need to be autopilot tests)14:29
sil2100Oh14:30
sil2100;)14:30
plarssil2100: as long as we continue to have an autopilot3 binary, those won't break14:30
plarssil2100:  but we can drop the hack when autopilot == autopilot314:30
sil2100I guess that's still not the case, but anyway, thanks! I guess it's safe to land those14:30
plarssil2100: the only things I'm concerned about are the ones around the edges - the smem/eventstat/memevent/etc14:31
plarssil2100: I was going to try to check on those today14:31
plarssil2100: when are you looking to pull py2 from the image?14:31
sil2100plars: I don't know the exact ETA, but we're getting to that slowly...14:32
plarssil2100: there are some minor (I hope) cleanups to do, and I'll be taking a look at those shortly14:33
balloonssergiusens, have some time today to give feedback on my mp's for phablet-tools? Landing the 13.10 -> 14.04 change should be really painless for instance14:37
sergiusensballoons: yeah, sorry, holidays made me forget to recheck :-)14:37
sergiusenssil2100: most of the core apps tests are py3 (if not all)14:38
sergiusensplars: sil2100 yeah, you could use autopilot as a replacement for nose if you wanted (wrt to not needing to be an autopilot test)14:39
balloonssergiusens, ty14:46
rsalvetiogra_: when are we building a new image? :-)14:49
plarssome of the 2to3 tool recommendations seem like nonsense14:49
rsalvetiwant a new build to test the x86 image14:49
ogra_rsalveti, i think didrocks wanted to build a new one today ... i would pretty much like to get lxc-android-conifg with the routing fix in before though14:50
didrocksogra_: yeah14:51
didrockswould be nice14:51
rsalvetiogra_: what is the fix for that?14:51
ogra_we had 4 perople test it and it worked for all of us14:51
rsalvetido we have a silo for it already?14:51
rsalvetithen land it :-)14:51
ogra_rsalveti, not sure and for a standloanr lxc-android-config upload i always find a silo overkill (since you cant really install that package easily anyway)14:51
rsalvetiogra_: where is the debdiff then?14:52
rsalvetiogra_: want to see what fixed it :-)14:53
ogra_rsalveti, a hack :P14:53
rsalvetihahah14:53
ogra_rsalveti, http://paste.ubuntu.com/7376744/14:53
rsalvetia dirty one for sure14:53
* ogra_ likes the name :P14:53
sil2100hoho14:54
rsalvetihahah14:54
rsalvetievil14:54
ogra_it just flushes the default routes every time an interface change is detected14:54
sil2100Is that the name we'll be shipping? That's evil indeed ;)14:54
ogra_we'll probably add _muhahaha at the end of the filename14:54
ogra_:P14:54
rsalvetilol14:54
sil2100Yeah, that makes it all much better ;p14:54
rsalvetiogra_: please land this then, and let's trigger a new image14:54
ogra_cyphermox, ok with you ? i'd just do a direct upload of lxc-android-config for that14:55
cyphermoxyes, that's always fine by me14:55
ogra_didrocks, objections ?14:55
didrocks"objection!"14:56
cyphermoxlet's not waste any more time than necessary14:56
didrockssorry, too many phoenix wright (ace attorney) :)14:56
didrocksso yeah, please land it!14:56
ogra_didrocks, testing it from a silo means you unpack the package any copy the file in place ... since lxc-android-config isnt installable :)14:56
ogra_great14:56
sil2100http://www.name.com/blog/wp-content/uploads/2013/04/objection.jpg14:57
didrocksogra_: http://upload.wikimedia.org/wikipedia/en/9/96/Ace_Attorney_objection.png14:57
ogra_(well, it is installable from recovery mode with a lot hackery)14:57
ogra_lol14:57
davmor2ogra_: cyphermox: right I am just waiting on some data transfers a reboot and some testing of missed calls and then I'm going to trial the proposed fix and see if I can break it14:58
cyphermoxbrb, need juice14:58
=== doanac changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: doanac | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: spreadsheet problems, do not trust CI-SNCF bot
ogra_uploaded15:03
sil2100\o/15:05
rsalvetisil2100: mind allocating a silo for line 34?15:10
popeydoanac: hello - https://code.launchpad.net/~popey/reminders-app/fix-vcs-manifest/+merge/218444 doesn't seem to have picked up my change?15:10
sil2100Just got the ping15:10
sil2100rsalveti: assigned!15:11
doanacpopey: let me poke around15:11
rsalvetisil2100: thanks15:12
popeydoanac: thanks, also https://code.launchpad.net/~popey/ubuntu-clock-app/fix-1315318/+merge/218067 seems to be failing and I cant see why15:12
sil2100:)15:13
doanacpopey: as per your clock-app. do you not think these are real failures: https://jenkins.qa.ubuntu.com/job/generic-click-autopilot-runner-mako/137/testReport/ ?15:20
popeydoanac: looking at http://ci.ubuntu.com/smokeng/utopic/touch/mako/13:20140506:20140506/7883/ubuntu_clock_app/ - it seems they're flaky..?15:21
doanacpopey: i'm not sure what the precedence is on getting an MP through that depends on flaky tests. do we have to get those tests un-flaky first?15:24
popeyhm15:24
popeydidrocks: was someone else looking at the clock app failures on CI this morning?15:26
didrockspopey: nothing more than on the ML AFAIK15:26
=== gatox is now known as gatox_lunch
doanacpopey: as per the reminders-app. The jenkins master is down that needs to kick that job off. fginther is going to get it back online15:33
popeythanks15:34
davmor2sil2100, didrocks: okay so the behaviour of callback didn't work on image 2 either but was actually broken worse so to that end I think it is safe to promote image 13 and add https://bugs.launchpad.net/ubuntu/+source/dialer-app/+bug/1316662 as a new blocker15:38
ubot5Ubuntu bug 1316662 in dialer-app (Ubuntu) "dialer app goes a bit funny when using the return missed call feature" [Undecided,New]15:38
didrocksdavmor2: \o/15:40
didrocksogra_: promotion :)15:40
popeyrenato: any ideas on https://bugs.launchpad.net/ubuntu-clock-app/+bug/131453315:41
ubot5Ubuntu bug 1314533 in Ubuntu Clock App "Test fails on mako #3 utopic" [High,Confirmed]15:41
popeyi just ran it locally here and it seems to take a while to save but not 15 seconds15:41
popeyand I'm not sure saving is the issue anyway, I don't see the toolbar come up at all15:42
popeyso I don't think it actually gets as far as trying to save15:43
popeyballoons: ^^15:43
balloonspopey, heh, I was just looking at those clock bugs15:43
popeywhen i run it here, it looks more like it's trying to scroll the flickable15:44
popeyand failing, then giving up15:44
balloonspopey, perhaps it's not as nik90 described then15:45
balloonspopey, we need to release shorts15:45
popeygiven I'm blocked on releasing reminders and clock, sure!15:45
popeyI'll test now15:46
balloonsI'm using r221.. those victor's changes won't land :-(15:46
balloonsbizzarre, the upload is failing :-(15:49
popeyballoons: 221 shorts pass, please upload to store15:57
balloonspopey, love to, but store uploads return server error 50015:57
popeybueno ^15:57
balloonspopey, they are looking at it :-)15:58
rsalvetiogra_: lxc-android-config is in, can we trigger a new image?15:58
ogra_didrocks, ^^^^15:59
didrocksogra_: please do, but don't image my promotion request! :)15:59
didrocksrsalveti: yes ;)16:00
ogra_already running16:00
rsalvetigreat16:00
didrockssweeeeeeet :)16:00
ogra_"Waiting for other process to release the global lock"16:00
* ogra_ sits in front of that since a while 16:00
ogra_oh, it moved now16:00
ogra_yup, 13 promoted16:00
ogra_build triggered16:01
didrocks\o/16:01
didrockssil2100: coming?16:01
rsalvetididrocks: hey, we have a newer hud in trusty, that is not yet in utopic:16:05
rsalvetihttps://launchpad.net/ubuntu/+source/hud16:05
rsalvetishouldn't we copy it over to utopic as well?16:05
didrocksrsalveti: yeah, cjwatson is doing it regularly at the start of the cycle16:06
cjwatsonIt's not in trusty-updates yet16:07
rsalvetiright, needs to be promoted first16:07
cjwatsonI don't generally copy things to utopic until we know we aren't going to get into a tangle with failed SRU verification16:07
cjwatsonThat said, in this case there's already a branched version in utopic-proposed16:07
cjwatsonSo I would consider this package out of bounds for the usual copying-up - please arrange for it to be uploaded separately16:08
rsalvetithat one was just a no-change rebuild, but failed for ppc64el16:08
cjwatsonOh, right, I guess that would be OK16:08
cjwatsonDid you try lobbing that ppc64el build back a few times?  It worked before, apparently ...16:09
rsalveticjohnston: it seems it's not arch related actually, did a new bump in my ppa and it failed similarly for both amd64 and i38616:09
rsalvetisorry, cjwatson ^16:09
balloonsping doanac16:09
cjwatsonWell, why don't I try retrying it then16:09
imgbot=== trainguard: IMAGE 14 building (started: 20140506 16:10) ===16:09
doanacballoons: hey16:10
cjwatsonJust in case it's a race16:10
rsalvetiright16:10
balloonsdoanac, it looks like maybe a jenkins builder is offline. The core apps autolanding jobs don't appear to be running16:10
doanacballoons: yes. fginther is aware and is looking into it16:10
doanacsorry16:11
balloonsdoanac, brillant thanks. I thought he might know, just wanted to confirm ;-)16:11
fgintherballoons, doanac, I think I have it fixed now. just restarting to clear out some old crud16:11
doanacold crud - i can't believe that would be possible16:11
balloonspopey, shorts is ready for you to approve16:20
Mirvok re: landing meeting filed a bug for file manager bug #131667716:20
ubot5bug 1316677 in Ubuntu File Manager App "flaky test test_cut_directory" [Undecided,New] https://launchpad.net/bugs/131667716:20
popeyballoons: done16:21
balloonsMirv, I didn't see that on the dash16:22
Mirvballoons: yes I got the link to http://q-jenkins.ubuntu-ci:8080/job/utopic-touch-mako-smoke-daily/38/ from plars16:24
balloonsMirv, ahh :-)16:24
Mirvso after the ubuntu_filemanager_app -> filemanager suite name was fixed, first rerun yielded that flaky test and the next one succeeded16:24
balloonsMirv, yea I kind of caught on that you had some fallout from that16:25
balloonsI'll have to remember fro renames it affects the dashboard also16:25
Mirvand then as I'm around today at this hour I can ping bfiller about getting this AP fix into gallery-app: https://code.launchpad.net/~chris.gagnon/gallery-app/autopilot-fix-flakyness-and-make-work-on-desktop/+merge/21812916:26
Mirv(ping done ^)16:26
cjwatsonrsalveti: Good, it built16:27
Mirv...except that it was just released already as a .deb, so only needs a click update16:28
Mirvgone for 2h and you go and fix things!16:28
rsalveticjwatson: awesome, did a local rebuild and it failed similarly on amd64, might be a flaky test16:29
bfillerMirv: it's landed16:33
=== gatox_lunch is now known as gatox
bfillersergiusens: gallery, notes and camera trunks now ready to be released as click16:39
* ogra_ reads todays landing team mail and wonders what the heck the "sync monitor toolbar" is 17:10
dpmhi doanac, we've got a couple of top-approved MPs from a few hours ago here that haven't had any Jenkins run - https://code.launchpad.net/reminders-app/+activereviews - is there anything going on with the Jenkins jobs for lp:reminders-app? Do they need to be retriggered?17:21
doanacdpm: i think they should get kicked off soon. we had a jenkins server offline this morning.17:22
sergiusensbfiller: ack17:22
bfillersergiusens: thanks17:23
doanacdpm: there are a couple of reminders-app MP being worked on it seems: http://91.189.93.70:8080/17:23
dpmdoanac, ok, so I guess that should get taken care of soon automatically. Thanks!17:26
bfillerfginther: could you setup CI for this branch: lp:~phablet-team/address-book-app/staging17:26
ogra_doanac, did you see my ping from yesterday ? the bootcharts you produce are not reproducable for me ... yours always take ~10sec longer, i dont get why ...17:27
doanacogra_: no. didn't see that.17:27
fgintherbfiller, yes, that can be done, should be ready a little later today17:27
doanacogra_: is there a specific phase where the CI lab is slower than what you see?17:28
ogra_doanac, well, i dont manage to get bootcharts with more than 28sec overall time here since a few images17:29
ogra_yours are far above 30 ... even close to 40 all the time17:29
ogra_http://people.canonical.com/~ogra/touch-bootcharts/ubuntu-phablet-utopic-11.png17:29
imgbot=== trainguard: IMAGE 14 DONE (finished: 20140506 17:30) ===17:29
ogra_http://people.canonical.com/~ogra/touch-bootcharts/ubuntu-phablet-utopic-13.png17:29
imgbot=== changelog: http://people.canonical.com/~ogra/touch-image-stats/14.changes ===17:29
ogra_and i have re-run both of these a few times when i saw yours17:30
ogra_they never take that long17:30
ogra_not sure how that can be17:30
doanacogra_: just a wild guess. we provision the phone with all the click-apps. could that slow things down any?17:31
ogra_doanac, i'll try to find what takes longer in yours ... (EOD here and i still have some things to do)17:31
ogra_doanac, you mean you dont use phablet-bootchart ?17:31
doanacogra_: we use phablet-bootchart, but skip its provisioning step i think17:32
doanacwe use the "--noinstall" flag17:32
ogra_more clicks can slow down the first boot massively ... but subsequent ones when the clicks are registered shouldnt slow down the boot17:32
ogra_hrm17:33
doanacogra_: probably not that then. because the phone should have already beent through the initial boot before we run the bootchart test17:33
ogra_you need three boots to even get the caches right ... thats why i added the install step to make sure it gets reliable numbers17:33
davmor2ogra_, cyphermox, popey: I can haz interwebz now to put it through it's paces :D17:33
ogra_davmor2, yay17:34
ogra_doanac, well, lets keep this as a topic for malta ... not sure i will find much time to dig deeper the next two weeks17:34
doanacogra_: ack. sounds good.17:34
ogra_(and i guess you are busy enough too :) )17:35
bfillerfginther: thank you17:36
fgintherbfiller, forgot to ask, you want the auto merger setup for that branch too, right?17:38
bfillerfginther: would be great17:38
fgintherbfiller, np17:39
rsalvetirobru: can you flush silo 9, don't need that landing anymore, need more work on it first17:50
=== bfiller is now known as bfiller_afk
robrursalveti, sure18:05
rsalveticjwatson: another rebuild that only failed for powerpc https://launchpad.net/ubuntu/+source/libusermetrics/1.1.1+14.04.20140305-0ubuntu318:07
cjwatsonrsalveti: Retried, although I don't know how hopeful I should be since it's failed at least twice now18:12
rsalvetiright18:13
rsalvetirobru: thanks!18:16
tvoss_sil2100, still around?18:17
robrutvoss_, I can help with landing things, what's up?18:19
sergiusensbfiller: who tested the clicks on the phone? did they notice 'Loading tests from: /home/phablet/autopilot/legacy-py2' at all?18:31
sergiusensthey don't seem to break on py3, but it's still py218:31
bfillersergiusens: ChrisCagnon tested them18:32
sergiusensxnox: do you know if camera, notes and gallery land that we would be python2 free? Just so we can get rid of the provisioning switch in phablet-click-test-setup18:32
xnoxsergiusens: as per our current list, that is the case.18:33
sil2100tvoss_: back now18:33
xnoxsergiusens: best way to check is to make sure $ click list --manifest has autopilot dir set for everything that has AP tests.18:33
sergiusensxnox: ok, because they migrated them to python3 but the manifest wasn't updated18:34
sergiusensI might consider an out of band update ;-)18:34
robotfuelsergiusens: what problems do you have with new click apps?18:34
xnoxsergiusens: yeah, some conversions were incomplete: missing manifest update, missing .deb updates, missing build-depends etc.18:34
robotfuelI tested them yesterday and it worked.18:34
sergiusensrobotfuel: they are provisioned to legacy-py2 since they don't have the click manifest switch18:34
xnoxsergiusens: i did clean up a few of them, but more sneaked in past.18:34
sergiusensrobotfuel: Loading tests from: /home/phablet/autopilot/legacy-py218:35
sergiusensrobotfuel: did you notice that?18:35
xnoxrobotfuel: sergiusens: where are the clicks?18:35
sergiusensxnox: s-jenkins, one sec18:35
xnoxor which branches / apps are we taking about ?18:35
sergiusensxnox: http://s-jenkins.ubuntu-ci:8080/view/click/job/camera-app-click/54/ http://s-jenkins.ubuntu-ci:8080/view/click/job/gallery-app-click/64/ http://s-jenkins.ubuntu-ci:8080/view/click/job/notes-app-click/64/18:35
xnoxsergiusens: i mean it's fine to land it, as long as one manually tests python3 code path, or a chaser merge-proposal is made to fix it up.18:36
sergiusensxnox: well I wanted to act like the translator bot that commits to trunk :-P18:36
sergiusensbut yeah, an MR requires a useless silo18:36
robotfuelsergiusens: I didn't notice that. I ran the tests manually with autopilot3 run <app-name> yesterday and they worked.18:36
sergiusenssilo is useless as it's not click ready18:36
sergiusensrobotfuel: yeah, that's not what ci does18:37
xnoxsergiusens: let's not release these to store, and i'll do a merge proposal to correct above.18:37
xnoxsergiusens: then you can review those.18:37
xnoxsergiusens: .... or just commit to trunk.18:38
sergiusensxnox: do the MR, I'll look and we can manually merge18:38
robotfuelsergiusens: what's the needed click manifest switch?18:43
robotfuelsergiusens: so I know what to look for next time...18:43
xnoxrobotfuel: xtest-autopilot18:43
xnoxrobotfuel: see $ adb shell click list --manifest | grep -C 2 autopilot18:44
dbarth_o/ robru hiya, we've got silo 003 finally verified; i did rebuild it ~1h ago, so i think it's safe to land now18:48
=== bfiller is now known as bfiller-afk
xnoxsergiusens: camera & gallery fixes proposed -> didn't build locally, will wait for jenkins (build-deps not installable on utopic for me)18:55
xnoxsergiusens: notes-app looks correct and has autopilot key.18:55
sergiusensxnox: ah, I didn't try that one actually18:56
sergiusensxnox: I'm uploading notes now18:58
sergiusensI just assumed it would have the problem while waiting for the everlasting gallery tests to finish18:59
sergiusensrobru: rsalveti can you reconfigure silo 006?19:00
robruon it19:01
sergiusenspopey: https://myapps.developer.ubuntu.com/dev/click-apps/118/ for Happroval19:05
sergiusenspopey: I ran the tests myself as you will notice in the changelog19:05
popeysergiusens: done19:07
sergiusenspopey: great, thanks; the stock thing should be sorted by now btw19:07
popeythanks19:14
davmor2popey, cyphermox, ogra_: Good news the lift test passed woohoo!  (Lift being equivalent to standing in a Faraday cage surrounded by 3 feet of concrete and mesh.  That cuts out 3g and wifi, downstairs door opens 3g comes back, upstairs wifi comes back each time I was able to connect to the net and g+ app)19:18
Saviqrobru, please delete lightdm packages from 002 ppa when you have a moment19:19
robrusure thing19:19
robruSaviq, ok, done. might take a second to really disappear though19:20
Saviqrobru, yup, known19:20
davmor2popey: the 3 apps you just updated notes, shorts and calculator none seem to be downloading19:26
sergiusensrobru: was that on it wrt my request and was it completed? :-)19:29
sergiusensrobru: was that on it wrt my request and was it completed? :-)19:30
robruugh my IRC client just crashed, who just pinged me?19:30
sergiusensrobru: me :-)19:30
robrusergiusens, uhhh... yes? I don't remember. When I said "on it" it was in response to the message *immediately* before it, and whatever it was is done now19:30
sergiusensrobru: great19:31
robruand my scrollback is gone19:31
robru;-)19:31
robrutime to find a new IRC client, smuxi has crashed like 5 times today, losing scrollback each time.19:31
sergiusensrobru: if I look at https://ci-train.ubuntu.com/job/landing-006-0-reconfigure/6/ it doesn't seem to be reconfigured though19:32
robrusergiusens, oh yes, that's because there's two different jobs for that19:32
robrusergiusens, the "reconfigure" job is the one that you are allowed to run but will fail if the packages have changed. when you ping me to reconfigure, I actually re-run the prepare job19:33
sergiusensrobru: ok; just wondering if there's a way to figure this out in case I missed the bot19:33
robrusergiusens, https://ci-train.ubuntu.com/job/prepare-silo/335/console19:33
popeydavmor2: store problem?19:33
sergiusensrobru: thanks, we don't have links for that though :-)19:33
robrusergiusens, there is a link at http://people.canonical.com/~rbpark/citrain/ ;-)19:33
davmor2popey: no idea I'm just double checking things my end19:33
popeydavmor2: lemme check on my stable phoen19:34
popey(which is frozen)19:34
sergiusensrobru: ah, the prepare a silo at the top, thanks19:34
davmor2popey: see it's so stable it's static19:34
sergiusensrobru: I'm using your view btw; easier to use; not a huge fan of the hiding buttons, but much much better than the spreadsheet19:35
sergiusensrobru: hope you add editing in there this weekend ;-)19:35
robrusergiusens, yeah I would love to add editing but unfortunately it's a bit beyond what I'm able to do with that particular setup (there's no server-side listening to form submissions there, it's a purely client-side ajax html page served statically)19:36
robrusergiusens, also, an older view of that page had non-hiding buttons and it was massively cluttered. had to hide them just to keep the view sane19:37
sergiusensrobru: yeah, I figured as much from the url; this would need some django or fun golang learning project ;-)19:37
sergiusensrobru: yeah, I just get confused when scrolling, I am not that picky either having used enterprise intranets before I'm used to anything these days... but I avoid excel like things as much as I can :-)19:38
robrusergiusens, yeah me too, I've been campaigning against spreadsheets since forever, but there's too much inertia to kill it.19:40
popeydavmor2: worked here19:54
=== bfiller-afk is now known as bfiller
=== jhodapp is now known as jhodapp|afk
fgintherbfiller, FYI, the address-book-app/staging jobs you requested were deployed21:43
bfillerfginther: great thanks23:26

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!