[01:01] <cjwatson> dobey: We have the hardware to scalingstackify that and may almost be at the point where IS have time to start configuring that, so hopefully won't have to limp along for too much longer ...
=== chihchun_afk is now known as chihchun
=== chihchun is now known as chihchun_afk
[04:48] <Mirv> mzanetti: thanks!
=== chihchun_afk is now known as chihchun
=== chihchun is now known as chihchun_afk
=== chihchun_afk is now known as chihchun
=== chihchun is now known as chihchun_afk
=== chihchun_afk is now known as chihchun
[05:36] <bzoltan_> Mirv: I have flipped the silo3's tested bit ... let's see the QA validation. But the AP tests are getting worst and worst...
[05:45] <Mirv> bzoltan_: :( that has been the trend I've felt too.
[07:59] <bzoltan_> jibel: Here are the results of the silo3 tests http://people.canonical.com/~bzoltan/ap-2015_07_06-DUAL-SILO03-KRILLIN/ No sign of regression. But the reference tests of the archive are really ugly.
[08:01] <jibel> bzoltan_, Hey, thanks.
[08:48] <jibel> sil2100, popey there'll be new core apps with updated icons for OTA5?
[08:49] <sil2100> jibel: yes
[08:50] <sil2100> We're waiting for the proper timing to land those, no-change uploads with just icon changed
[08:50] <sil2100> *icons
[08:51] <jibel> sil2100, when is a "proper timing" ?
[08:55] <sil2100> Didn't get an answer from Pat yet - the two possible windows are: right before building the first candidate and right before building the image with additional missing translations
[08:56] <sil2100> Considering the second one is required
[08:58] <morphis> sil2100: wily landings doesn't require a QA sign-off, right?
[08:59] <sil2100> morphis: no
[08:59] <morphis> sil2100: so just leaving the field in column QA sign-off at "N/A nickname" is fine, or do I have to put my nickname in there?
[09:06] <sil2100> morphis: in this case it's fine :)
[09:06] <sil2100> morphis: oh, and we need to do a watch_only build on the silo
[09:07] <sil2100> Since it didn't notice your packages there
[09:07] <morphis> ?
[09:08] <sil2100> morphis: the train is a rather simple tool, when you upload something to the silo PPA, the train is not automatically triggered to check what that was as it's a set of jenkins jobs
[09:08] <sil2100> morphis: so whenever you upload something to the PPA and want the train to register it, you need to go to the silo build job, check 'watch_only' and run the job
[09:08] <sil2100> I did that just now ^
[09:09] <morphis> sil2100: ah I see
[09:09] <morphis> sil2100: so whenever I have such a case again, I just check the WATCH_ONLY parameter and leave everything else unchecked/unfilled?
[09:10] <sil2100> morphis: yes :)
[09:10] <popey> jibel: yeah, the idea we discussed in the call yesterday was for me to just re-upload the same revision of the clicks, but with new icons, and set the new icon in the store
[09:10] <sil2100> WATCH_ONLY basically means: "don't build/rebuild anything, just look at the PPA contents and check if everything built correctly"
[09:10] <morphis> sil2100: good
[09:10]  * sil2100 adds that to the docs
[09:18] <popey> jibel: agreed with sil2100  that we should upload them on Thursday
[09:19] <sil2100> Makes most sense to me, but still want Pat to say if he's ok with that
[09:22] <sil2100> morphis: added a mention of this to https://wiki.ubuntu.com/citrain/LandingProcess
[09:23] <morphis> sil2100: thanks
[09:28] <Laney> fginther: any clue about https://jenkins.qa.ubuntu.com/view/wily/view/All/job/wily-boottest-powerd/10/console or http://d-jenkins.ubuntu-ci:8080/view/Wily/view/BootTest/job/wily-boottest-qtubuntu-sensors/6/console or http://d-jenkins.ubuntu-ci:8080/view/Wily/view/BootTest/job/wily-boottest-qtubuntu-gles/9/console ?
[09:29] <Laney> oops, sorry for private link
[09:29] <jibel> sil2100, click of the camera-app has not been published?
[09:29] <jibel> sil2100, version 3.0.0.573
[09:43] <sil2100> jibel: I see that the publishing of that had some issues
[09:43] <sil2100> Need to check what that was
[09:44] <sil2100> jibel: yeah, so robru wanted to publish that but someone pushed new revisions to the camera-app branch there
[10:31] <abeato> trainguards, can I have a silo for line 75?
[10:36] <sil2100> abeato: on it
[10:36] <abeato> sil2100, thx
[10:44] <mzanetti> trainguards, please drop unity8 from this ppa: https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-035/+packages
[10:44] <mzanetti> the branch in there wasn't needed any more, so I re-used the silo for something else
[10:48] <sil2100> mzanetti: ok
[10:48] <sil2100> mzanetti: removed
[10:48] <mzanetti> thanks
[10:51] <mzanetti> sil2100, this looks like a temporary failure to me: http://d-jenkins.ubuntu-ci:8080/job/wily-adt-thunderbird/ARCH=i386,label=adt/49/console
[10:51] <mzanetti> it passed on the other archs
[10:51] <mzanetti> can we re-run that somehow?
[10:51] <mzanetti> it's in the excuses for http://people.canonical.com/~platform/citrain_dashboard/#?q=ubuntu%2Flanding-051
[10:52] <sil2100> I'm sure we can, let's poke cihelp for that ^
=== karni is now known as karni-afk
=== karni-afk is now known as karni
[11:17] <jibel> mzanetti, with the new theme the dots on the greeter are gone, is it ok?
[11:17] <jibel> ah, nm it's back
[11:17] <jibel> weird
[11:17] <mzanetti> not sure what dots you mean
[11:18] <jibel> mzanetti, there are small dot close to the border of the circle on the greeter to indicate the day of the month
[11:18] <jibel> sometimes there is nothing
[11:19] <mzanetti> ah ok...
[11:19] <jibel> I thought it was the theme I updated from the silo but apparently it's something else
[11:19] <ogra_> moar races !
[11:20] <ogra_> ubuntu is the fastest phone OS ever ... with so many races you have to be fast :)
[11:23] <jibel> jamesh, Hey, I tried the new thumbnailer in silo 10 and it fails to generate thumbnails of videos recorded with the camera
[11:23] <jibel> it's on arale, I'll try on krillin
[11:29]  * sil2100 gently pokes ogra_ 
[11:29] <sil2100> :D
[11:38]  * sil2100 off to lunch
[12:16] <jibel> jamesh, silo 10 again, with the new thumbnailer and lot of pictures in the gallery apps it triggers oom killer
[12:17] <cwayne> cihelp ping
[12:18] <jibel> jamesh, like a hundred pictures, not thousands
[12:23] <jibel> rvr, davmor2 Hey, when you have time I need a second opinion on silo 10. on krillin with 100 pictures or so and the device runs out of memory when I switch to the photo view in the gallery, and on arale there is no thumbnail for videos recorded from the camera or not
[12:24] <rvr> jibel: Ok
[12:27] <rvr> jibel: Flashing the krillin, I am using arale with my silo
=== _salem is now known as salem_
[13:06] <rvr> jibel: Ok
[13:07] <rvr> jibel: So, it's weird. No thumbnails, but also, I see a music icon
[13:07] <jibel> rvr, for photos?
[13:07] <jibel> or videos?
[13:07] <rvr> jibel: For photos
[13:07] <rvr> Uploading screenshot
[13:08] <jibel> uh
[13:08] <fginther> mzanetti, sil2100, http://d-jenkins.ubuntu-ci:8080/job/wily-adt-thunderbird/50/ has been rerun and passed.
[13:08] <mzanetti> fginther, yep, I saw. thanks a lot
[13:09] <cwayne> fginther: ping
[13:09] <fginther> cwayne, morning
[13:09] <rvr> jibel: http://people.canonical.com/~vrruiz/gallery-app-100-photos.png
[13:09] <fginther> cwayne, was in the middle of unwinding the help request stack
[13:09] <cwayne> fginther: mornin' :)  we've got some failing jobs, like this one: http://s-jenkins.ubuntu-ci:8080/view/cambridge/job/BQ%20International%20Tarball%20-%20Gated/  seems to be a java heap error?
[13:09] <sil2100> fginther: thanks :)
[13:09] <jibel> rvr, same pictures work fine without the silo
[13:09] <jibel> ?
[13:10] <cwayne> fginther: oops, sorry, feel free to skip me til it's my turn :)
[13:10] <rvr> jibel: Reflashing
[13:10] <fginther> cwayne, nope, it's ok
[13:14] <kenvandine> jibel, why did you bump silo 2 out of ready for testing?
=== barry` is now known as barry_
=== barry_ is now known as barry
[13:18] <jibel> kenvandine, I prioritized features and bugs targeted to ota5
[13:18] <kenvandine> jibel, ok, thx
[13:18] <jibel> I'll move it to ready if we have enough bandwidth
[13:20] <fginther> cwayne, something has gone very bad on s-jenkins. It needs a restart
=== fginther 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: http://s-jenkin.ubuntu-ci:8080/ needs to be restarted, it's hung
[13:21] <seb128> kenvandine, where do you see what is ready for testing or not?
[13:22] <fginther> josepht, psivaa-afk, I'm restarting s-jenkins. It's essentially dead right now
[13:22] <rvr> jibel: Without the silo I see the thumbnails
[13:23] <rvr> seb128: In trello
=== chihchun is now known as chihchun_afk
[13:23] <josepht> fginther: ack, let me know if I can help
[13:23] <rvr> seb128: https://trello.com/b/AE3swczu/qa-testing-requests-for-questions-ping-eu-jibel-us-jfunk-or-ubuntu-qa-on-ubuntu-ci-eng
[13:23] <seb128> rvr, thanks
[13:24] <jibel> rvr, thanks for the confirmation. Moved to failed
[13:24] <seb128> rvr, jibel, can we get 034 in the "ready for testing"? one of the bugs would be nice to get in ota5 and was acked by pmcgowan (dialpad respecting silent mode)
[13:26] <jibel> seb128, bug 1384274 ?
[13:26] <ubot5> bug 1384274 in telephony-service (Ubuntu) "silent mode doesn't silence keyboard or dialpad" [High,In progress] https://launchpad.net/bugs/1384274
[13:27] <seb128> jibel, yes
[13:27] <jibel> seb128, it is not targeted
[13:27] <seb128> jibel, well, pat said it was ok to add on IRC, that's not good enough?
[13:27] <seb128> pmcgowan, ^ can you target the bug?
[13:36] <rvr> pstolowski: Approving silo 42
[13:37] <pstolowski> rvr, great, thank you!
[13:39] <pmcgowan> seb128, jibel sure, but targeting s more priority not exclusivity to me
[13:39] <pmcgowan> as long as we have time to test
[13:56] <fginther> camako, the default channel in use for MP tests is "ubuntu-touch/devel-proposed/ubuntu"
[13:57] <camako> thanks fginther
=== chihchun_afk is now known as chihchun
=== Olli_ is now known as olli
[14:36] <Laney> fginther: did you see my Q?
[14:41] <fginther> Laney, yes, sorry about that. Was hoping to get to that next
[14:46] <fginther> cwayne, that job is passing now
[14:47] <AlbertA> cihelp: need help again on platform-api rdeps failing boottest
[14:47] <fginther> AlbertA, powerd, qtubuntu-sensors and qtubuntu-gles ?
[14:47] <Laney> fginther: no worries, thanks
[14:47] <AlbertA> fginther: yes and ubuntu-touch-meta
=== psivaa-afk is now known as psivaa
[14:48] <Laney> If we could make the boottest output more useful (e.g. 'sh -x' and saving the apt-get stderr) then we'd be a long way to being able to understand problems
[14:50] <fginther> AlbertA, I'll retry ubuntu-touch-meta and look at the others
[14:53] <AlbertA> fginther: thanks
[14:53] <fginther> Laney, part of the problem is that the test has to be retried multiple times and it's only saving the last set of results. But it's not an impossible problem to fix
[14:56] <Laney> fginther: For apt-get-install.stderr you could put the package name in there, like xargs -I{} ... 2>apt-get-install-{}.stderr (maybe with '>>' if there are indeed many runs)
[14:56] <Laney> and sh -x should go to the console so be logged anyway, AFAICS
[15:00] <fginther> Laney, thanks
[15:02] <cwayne> fginther: <3 thanks
[15:13] <greyback> do silos have dbg symbols saved anywhere?
[15:14] <greyback> oh, they're in the silo, ignore me
[15:18] <cjwatson> they sure are :)
[15:29] <dobey> sil2100: you're manually uploading translations for the overlay ppa? where do the .po files come from in that case?
[15:31] <sil2100> dobey: currently? For all that it's possible it comes from wily, that's the only sane way
[15:32] <dobey> ok
[15:34] <cjwatson> dobey: next LP deploy will stuff those into ubuntu-rtm/15.04
[15:34] <cjwatson> (all being well)
[15:37] <pmcgowan> jibel, did you test the thumbnailer with the new gallery in 36?
[15:38] <dobey> oh gah, forgot that the template needs to be updated
[15:50] <dobey> trainguards: What happens if a new revision is pushed to an MP branch for something that's already in -proposed pocket?
[15:50] <jibel> pmcgowan, no, but I can try
[15:52] <pmcgowan> jibel, seems it was marked as a pre-req
[15:52] <jibel> pmcgowan, sorry I missed that. Where is it marked as a pre-req
[15:52] <jibel> ?
[15:53] <pmcgowan> only in the doc
[15:53] <sil2100> dobey: not good... this means that LP will have problems marking the MP as merged after it migrates
[15:53] <pmcgowan> jibel, see bills pointer to that bug
[15:54] <dobey> sil2100: will ci train commit anything?
[15:54] <jibel> pmcgowan, yeah, not obvious. I'll give it a try
[15:54] <sil2100> dobey: to make sure that everything merges correctly we usually propose a bzr push --overwrite without the commit that didn't get built
[15:55] <sil2100> dobey: it will commit that what it built, the rest will stay unmerged
[15:55] <sil2100> dobey: the MP will not be marked as merged as well because of that ,as it will only be partially-merged in LP's view
[15:56] <dobey> hmm
[16:00] <fginther> AlbertA, Laney, The boottest for qtubuntu-sensors fails because the unity greeter never starts. I'm able to reproduce on a local device
[16:00] <fginther> It just displays the ubuntu spinner
[16:01] <fginther> AlbertA, ubuntu-touch-meta is now passing
[16:03] <rvr> mzanetti: ping
[16:03] <mzanetti> rvr, hey ho
[16:03] <rvr> mzanetti: Silo 17. Is there an app to test that?
[16:04] <mzanetti> rvr, http://notyetthere.org/data/untitled5.mzanetti_0.1_all.click
[16:04] <mzanetti> rvr, however, the text in the app does NOT work
[16:04] <mzanetti> rvr, watch it's debug prints in ~./cache/upstart/application-click-untitled5.zanetti.log
[16:05] <AlbertA> fginther: ack
[16:05] <mzanetti> rvr, when it prints "disabled" the screensaver should be disabled, meaning it will keep the display on
[16:06] <rvr> qml: enabling/disabling: false
[16:06] <rvr> qml: enabling/disabling: true
[16:07] <rvr> mzanetti: enabling/disabling... which one is? :D
[16:07] <mzanetti> haha
[16:07] <mzanetti> rvr, "false" means screenSaver not enabled :D
[16:07] <rvr> mzanetti: Ok
[16:07] <mzanetti> rvr, it prints print("enabling/disabling:", screensaver.enabled)
[16:08] <mzanetti> I didn't think that app became so famos :D
[16:08] <mzanetti> otherwise I would have put a little more efforts into it
[16:09] <rvr> mzanetti: lol
[16:09] <rvr> mzanetti: Seems to work
[16:09] <mzanetti> nice
[16:09] <rvr> Doing more checks
=== chihchun is now known as chihchun_afk
[16:28] <jibel> pmcgowan, same result, gallery is killed http://paste.ubuntu.com/11836572/. That's the only app running
[16:28] <jibel> pmcgowan, and I've only a hundred photos
[16:29] <pmcgowan> jibel, rats
[16:29] <pmcgowan> jibel, could land the gallery change though if its working
[16:29] <jibel> pmcgowan, sure, I'll try gallery without the new thumbnailer
[16:32] <oSoMoN> jibel, hey, why did the card for silo 22 go back to "Needs QA sign-off" ?
[16:38] <jibel> oSoMoN, I kept in "ready for testing" bug fixes or features for ota5 to make sure testers land them first
[16:38] <oSoMoN> jibel, ok, so nothing wrong with the silo itself, right?
[16:39] <jibel> oSoMoN, absolutely nothing. Maybe it'll land in this release if we have enough bandwidth
[16:41] <oSoMoN> fair enough
=== zsombi_ is now known as zsombi
[17:51] <popey> sil2100: when do you want these icon-only changed clicks  uploading?
[18:00] <fginther> AlbertA, Laney, powerd was failing boottest because it needs to do a cross partition link during install which isn't going to work. It's been manually passed for now. A slightly improved log output is here: http://d-jenkins.ubuntu-ci:8080/job/fjg-boottest/63/artifact/adt-run-stdout.0/*view*/
[18:01] <AlbertA> fginther: yeah I have to do "sudo umount /usr/share/powerd/device_configs/config-default.xml" before powerd install
[18:01] <AlbertA> fginther: thanks btw
[18:02] <fginther> AlbertA, you're welcome
=== dpm is now known as dpm-afk
[18:27] <cyphermox> cihelp: I'd like to login to jenkins and investigate directly why the wily-desktop-amd64-smoke-default job keeps failing. There's no apparent reason for the debconf warning in the log, nor for the late command to wait for network
[18:46] <boiko> kenvandine: hey, with gallery-app silo 36 being published, do you know if any extra step is needed to get the click package updated too?
[18:48] <kenvandine> boiko, there is something that has to be done
[18:48] <kenvandine> but i'm not familiar
[18:49] <boiko> kenvandine: me neither, bfiller asked me to help with the landing, but it is my first click package landing :)
[18:49] <boiko> robru: maybe you know something about it? ^
[18:49] <robru> boiko: uh yeah, you need popey to upload the click package to the store
[18:49] <popey> hello
[18:50] <boiko> popey: hey! :) could you please get http://people.canonical.com/~bfiller/com.ubuntu.gallery_2.9.1.1209_armhf.click uploaded to the store?
[18:50] <popey> sure
[18:50] <boiko> popey: thanks!
[18:50] <popey> has it been QA'ed?
[18:50] <boiko> popey: yep, silo 36, the deb is in the proposed pocket already
[18:51] <kenvandine> cool
[18:51] <popey> got a chanelog?
[18:51] <boiko> popey: let me get that, just a sec
[18:51] <popey> store currently has 1208
[18:51] <popey> so looks like translations only?
[18:52] <boiko> popey: http://paste.ubuntu.com/11837368/
[18:53] <boiko> popey:  * Disable unused GalleryPhotoComponent so we stop doing unnecessary
[18:53] <boiko>     request to thumbnailer (LP: #1460993)
[18:53] <ubot5> Launchpad bug 1460993 in gallery-app (Ubuntu) "Multiple thumbnail requests for a single image" [High,In progress] https://launchpad.net/bugs/1460993
[18:54] <popey> you sure that click has that change?
[18:54] <popey> https://bugs.launchpad.net/gallery-app/+bug/1460993 doesn't look like that merge has landed, approved, not landed yet
[18:54] <ubot5> Launchpad bug 1460993 in gallery-app (Ubuntu) "Multiple thumbnail requests for a single image" [High,In progress]
[18:54] <popey> https://code.launchpad.net/~phablet-team/gallery-app/trunk also
[18:54] <boiko> popey: in the proposed pocket
[18:54] <popey> this is a click though
[18:54] <popey> built from trunk 1209 by the filename
[18:55] <popey> so that click doesn't match the deb
[18:55] <popey> IMO?
[18:55] <boiko> popey: that's the click bfiller built for testing
[18:55] <boiko> popey: it for sure has the fix for the bug
[18:55] <charles> yay
[18:56] <boiko> popey: I am not familiar with click packages, sorry, just helping out as bfiller is off today
[18:56] <popey> ok
[18:56] <popey> just making sure :)
[18:57] <boiko> popey: yep, if you can wait a couple minutes, I am double checking with arthur (who made the changes) and then I let you know
[18:57] <boiko> artmello: hey, so popey is helping out with the click package publication
[18:57] <boiko> artmello: the one from bfiller's people.c.c contains the fix for the bug, right?
[18:58] <boiko> artmello: that's the one that should be published, is that correct?
[18:58] <artmello> boiko: yes, it does contains the change
[18:58] <popey> great
[18:58] <robru> boiko: the merge that built the deb doesn't get merged to trunk until after it migrates through proposed, so if the click was just built from trunk it won't have the fix. unless bill built it specifically from the MP.
[18:59] <popey> it wasn't built from trunk
[18:59] <robru> alright
[18:59] <popey> jenkins does that and I diff'ed the jenkins one and bills one
[18:59] <popey> not the same
[18:59] <boiko> robru: artmello confirmed it has the fix, so it wasn't built from trunk
[18:59] <robru> oh, great
[18:59] <robru> boiko: ok no worries, just trying to offer some insight into train process.
[19:00] <boiko> robru: sure, always welcome, I know 0 about click landings, so all the help is appreciated :)
[19:00] <robru> boiko: might be beneficial to ask the QA person if they tested the click or the deb or both ;-)
[19:00] <boiko> om26er: ^
[19:00] <boiko> robru: btw, is there an excuses page for the stable phone overlay? or a place where I can check the migration from proposed to the overlay?
[19:01] <robru> boiko: there is no "migration" to the overlay, packages are binary copied immediately.
[19:01] <AlbertA> fginther: any idea on qtubuntu-sensors? installing the packages here locally works fine (unity comes up just fine)
[19:01] <om26er> robru, boiko I tested the click,  from bill
[19:01] <fginther> AlbertA, do you happen to have other packages installed from that silo?
[19:01] <robru> om26er: sweet, thanks
[19:02] <psivaa> cyphermox: the VMs for those desktop tests are in aldebaran.ubuntu-ci
[19:02] <fginther> AlbertA, maybe theres a dependency?
[19:02] <psivaa> cyphermox: you should be able to login there now
[19:02] <boiko> robru: ah, gallery is a dual landing, ok, any idea why it is not showing up in the excuses page then?
[19:02] <AlbertA> fginther: yeah all the packages from the silo (citrain device-upgrade 13 <passwd>)
[19:03] <popey> boiko: robru 1209 in the store
[19:03] <robru> boiko: excuses page only updates every 15 minutes or so, give it some time. packages will be in vivid overlay already, and will migrate to wily soon
[19:03] <robru> popey: thanks!
[19:04] <boiko> robru: ok, thanks for the info :)
[19:04] <boiko> popey: thanks!
[19:04] <boiko> artmello: ^
[19:04] <robru> boiko: you're welcome
[19:04] <artmello> boiko, popey , robru: thx guys
[19:05] <robru> artmello: you're welcome!
[19:06] <awe> robru, I created a line for NM ( 79 ), but it didn't get a silo assigned.  At this point, we're not going to land for OTA5, but just wanted to give you a heads-up.  As we're doing a binary sync from the telephony PPA, I'm doing my testing on packages pulled from there
[19:06] <awe> no rush on it though
[19:06] <awe> just wanted to give you a heads-up
[19:06] <robru> awe: you want a silo?
[19:07] <awe> yes, but it'll be for post OTA5 at this point, so if your're running low, I can wait
[19:07] <boiko> jibel: jfunk: I had marked silo 26 as ready for QA, but after talking to salem_ and pmcgowan we decided to put one more messaging-app bugfix in there, so I will have to re-test the silo
[19:07] <awe> robru, ditto if you have other higher prioritiesd
[19:07] <fginther> AlbertA, boottest only installs binaries from the source package under test (and their dependencies). I suspect if you perform a clean install and just update qtubuntu-sensors, you should be able to reproduce the problem
[19:07] <robru> awe: 13 free, I think it's ok ;-)
[19:08] <AlbertA> fginther: well it should pull libubuntu-application-api3
[19:08] <fginther> AlbertA, I'll check the logs for that
[19:09] <robru> awe: ok, binaries copied, should be available in silo 6 shortly: https://ci-train.ubuntu.com/job/ubuntu-landing-006-1-build/261/console
[19:09] <AlbertA> fginther: I'll try to reproduce here though thanks!
[19:09] <fginther> AlbertA, I'm trying again too, with some improved logging
[19:10] <fginther> (at least I hope it's improved)
[19:14] <awe> thanks robru!
[19:14] <robru> awe: you're welcome
[19:40] <cjwatson> robru,sil2100: can you let me know next time something's published to the vivid overlay PPA?  I'd like to make sure that the new translations redirection is working
[19:41] <robru> cjwatson: oh I just did a couple, is it too late for those? eg gallery-app was just published within the last hour
[19:41] <cjwatson> robru: our deployment might have been in time, let me check
[19:42] <cjwatson> scratch that, toddler just woke up.  will check later
[19:43] <robru> cjwatson: ok no worries, will ping if I publish anything
[19:43] <robru> brb
[19:52] <AlbertA> fginther: indeed, so the issue is qtubuntu-sensors depends indirectly on ubuntu-application-api3-touch or ubuntu-application-api3-desktop... my guess is since you can't describe that cleanly with debian deps
[19:52] <AlbertA> the dependency is actually in qtubuntu
[19:53] <AlbertA> so qtubuntu-android depends on ubuntu-application-api3-touch and qtubuntu-desktop depends on ubuntu-application-api3-desktop
[19:54] <fginther> AlbertA, can this be solved for this set of packages, or does it need some exception to make forward progress?
[19:55] <balloons> plars, fginther did you guys setup any cards or other actions from the meeting yesterday afternoon on core apps jenkins? i'd like to keep track of the status of everything :-)
[19:55] <AlbertA> fginther: the issue is if we add a dependency in qtubuntu-sensors, it would have to be both ubuntu-application-api3-touch and ubuntu-application-api3-desktop...so you would end up with
[19:55] <AlbertA> unnecessary binaries in one platform vs the other
[19:56] <AlbertA> fginther: so exception in this case...
[19:56] <AlbertA> fginther: the qtubuntu test should be enough proof as that pulls qtubuntu-sensors
[19:58] <fginther> AlbertA, I can confirm that upgrading "qtubuntu-android" solves the greeter problem
[19:58] <popey> robru: is the goal to built ota5 tonight?
[19:58] <AlbertA> fginther: cool
[19:58] <fginther> balloons, yeah, we created some cards for our work, I'll send them to you in a moment
[19:59] <plars> balloons: we're actively looking at expanding the hardware to support this, but it's not going to happen overnight, until then there's a transition plan around the jenkaas stuf
[20:00] <fginther> AlbertA, I'll get that package unblocked
[20:00] <AlbertA> fginther: thanks
[20:04] <robru> popey: dunno
[20:04] <robru> fginther: psivaa: did you guys make any progress on the jenkins persistent volume issue? I might take a look at that today
[20:07] <fginther> robru, we're making progress, but the problem isn't solved yet as we keep uncovering new issues. We're now trying to solve it a different way by attaching the persistent volume before the charm does anything else
[20:07] <fginther> robru, which is difficult because, the nova volume can't be attached until after the instance has been deployed
[20:08] <robru> fginther: are you hacking on the jenkins charm itself to fix this?
[20:08] <fginther> robru, yes.
[20:08] <robru> hmmmm
[20:09] <robru> fginther: if you fix it, will you upstream it, or just leave it as a special ci-owned branch?
[20:10] <fginther> robru, it's currently under a team branch as we work on it. And yes we plan on upstreaming what we can. It will at least be available to future ci-train deployments
[20:10] <robru> fginther: I feel like the easiest solution would be "find whatever file in /var/lib/jenkins that jenkins pukes without, then copy that file to the persistent volume after mounting"
[20:11] <robru> fginther: I was considering handling this in my jenkins subordinate charm that I own, but it'd be really nice if jenkins just stopped crapping itself when you mount a volume in it ;-)
[20:12] <fginther> robru, the approach we're taking should actually treat the persistent volume as a backup, and not install anything over it on a new deployment.
[20:13] <fginther> robru, but we'll have to see how it works. Jenkins always provides surprises
[20:14] <robru> fginther: hmmmm so /var/lib/jenkins would stay as ephemeral but everything in it would be routinely backed up? how do you decide what parts to restore when mounting the volume?
[20:15] <robru> fginther: actually that seems a sensible approach, you just need a reliable way to decide what to restore, because if you clobber the wrong file you'll get the exact same issue.
[20:15] <robru> (I'm really annoyed that I never bothered to figure out specifically which file is the problematic one)
[20:15] <fginther> robru, /var/lib/jenkins would be on the persistent storage. But the charm would be setup so that it never installs jenkins until after /var/lib/jenkins is mounted. It eliminates the rsync step
[20:16] <fginther> nothing is restored or copied
[20:16] <robru> oh
[20:16] <robru> fginther: that would be tricky, what if you wanted a deployment without a persistent storage? it'd just fail to come up without the volume?
[20:17] <robru> fginther: (in ci train we use a persistent volume in production but not in staging precisely because of this bug)
[20:17] <fginther> robru, if you don't want persistent storage, don't use this charm :-)
[20:17] <fginther> robru, actually, even that should be possible if you set the right config options.
[20:18] <robru> fginther: yeah that's the problem, there's already half a dozen forks of the jenkins charm floating around. will be significantly better if you upstream a solution that works with and without persistent storage
[20:19] <cjwatson> robru: gallery-app was half an hour or so too early, it seems, but I'm sure there'll be more
[20:20] <cjwatson> AlbertA: can't you have qtubuntu-sensors depend on ubuntu-application-api3-touch | ubuntu-application-api3-desktop, and then have a platform-specific package depend specifically on one or the other
[20:20] <cjwatson> ?
[20:20] <balloons> fginther, plars right. Since it won't be quick I want to make sure everyone understands what will be done, and have some idea about how long it may take to complete. I'll look at the cards, thanks
[20:21] <cjwatson> AlbertA: (maybe, I haven't checked the details)
[20:22] <AlbertA> cjwatson: is that possible?
[20:22] <cjwatson> which bit?
[20:23] <AlbertA> cjwatson: do Depends: ubuntu-application-api3-touch | ubuntu-application-api3-desktop
[20:23] <cjwatson> sure
[20:23] <cjwatson> you do have to pick a preferred alternative though
[20:24] <cjwatson> it might make more sense to have both ubuntu-application-api3-{touch,desktop} Provides: some virtual package which describes the interface depended on by qtubuntu-sensors
[20:24] <cjwatson> and then have qtubuntu-sensors Depends: that virtual package
[20:24] <cjwatson> that is probably what I'd do
[20:24] <AlbertA> cjwatson: I see
[20:35] <cyphermox> psivaa: thank you!
[20:38] <robru> cjwatson: alright i haven't published anything since but will let you know
[21:06] <popey> balloons: fginther did we have a conversation about enabling the overlay ppa in jenkins?
[21:06] <popey> I know I wanted to, but couldn't remember if we did.
[21:06] <fginther> popey, maybe you did and I don't remember it :-)
[21:07] <popey> ok, so the problem is we have merges landing which depend on that ppa being enabled, as it is on the phone
[21:07] <popey> kinda surprised nobody else has brought this up
[21:07] <popey> and I know balloons has talked to you guys about various jenkins issues, so didn't want top stamp on that conversation
[21:08] <fginther> popey, it was brought up for the touch packages, didn't know this might also apply to the core-apps (I assume this is about the core-apps)
[21:08] <balloons> popey, I was hoping I wouldn't hear the overlay ppa is required
[21:08] <fginther> popey, got an example?
[21:09] <popey> well, UI toolkit bugs for example are addressed in wily and land in overlay
[21:10] <popey> and apps which are building in vivid will fail
[21:11] <popey> https://code.launchpad.net/~nikwen/ubuntu-filemanager-app/remove-popover-workaround/+merge/261909
[21:11] <popey> there you go
[21:11] <popey> https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1395118
[21:11] <ubot5> Launchpad bug 1395118 in Ubuntu File Manager App "ActionSelectionPopover does not handle the 'visible' property of actions properly" [Undecided,In progress]
[21:15] <popey> http://tumbleweed.popey.com/
[21:15] <fginther> popey, sorry, it's summertime and the kids interrupt...
[21:15] <popey> haha :)
[21:18] <fginther> popey, I don't think this is going to help utopic builds. The overlay PPA is targeted to vivid and I would be surprised (very surprised) if that would install cleanly in an utopic environment
[21:18] <fginther> popey, but it would help vivid
[21:18] <popey> right, so the next question you can expect...
[21:18] <popey> when will we move to vivid
[21:19] <fginther> popey, balloons is already hounding me on that :-)
[21:19] <fginther> popey, we're trying to work out something, but it sounds like the urgency needs to be bumped now
[21:19] <fginther> so consider it bumped
[21:21] <popey> thanks
[21:38] <robru> AlbertA: got you silo 13, note conflict in silo 32 and 34
[21:41] <AlbertA> robru: ack, thanks
[21:41] <boiko> jfunk_: jibel: in the QA board, there are two cards for silo 26, one under column Ready for Testing and another under Need QA Sign-off
=== salem_ is now known as _salem
[23:12] <AlbertA> trainguards: landing-013 can be abandoned - didn't realize it papi was already synced...
[23:13] <robru> AlbertA: no worries
[23:15] <robru> cjwatson: ah, published telephony-service to the overlay ppa, can you check the translations? https://ci-train.ubuntu.com/view/2.%20Publish/job/ubuntu-landing-034-2-publish/lastSuccessfulBuild/console
[23:31] <cjwatson> robru: Thanks, it's worked to the extent that the translations are in the correct import queue.  I'll see what we need to do to get those imported and made available for translation.
[23:45] <robru> cjwatson: great