/srv/irclogs.ubuntu.com/2016/02/05/#ubuntu-devel.txt

cjwatsonpitti: Well, I was pretty sure it would be fine to sync it with clock-setup, where that code came from in the first place anyway00:17
=== mnepton is now known as mneptok
=== alecu_ is now known as alecu
xnoxpitti, tah.04:00
xnoxmwhudson, infinity src:golang-1.5 =)04:02
Mirvpitti: \o/05:27
Mirvthat was very fast feature bring up05:27
xnoxpitti, cpr is needed for ppc64el on autopkgtest. Queues are piling up, and none are running.05:31
xnoxpitti, oh, they seem to be back alive06:09
=== cpaelzer_afk is now known as cpaelzer_
pittiGood morning07:26
pittixnox: yep, looking; utlemming fixed the ppc64el images yesterday, but apparently that didn't make it into yesterday's07:28
pittixnox: at 0600 UTC the janitor cronjob runs which cleans up orphaned instances and restarts stuff, but it can't fix the reason for the ssh timeouts07:29
* pitti prods the armhf runners too07:29
xnoxfun =/07:30
pittithere is some lxcfs lockup on armhf07:30
=== rbasak_ is now known as rbasak
pittiutlemming: hm, apparently something changed in the underlying environment -- I can't boot wily ppc64el images any more either :/07:59
pittiutlemming: can you please trigger new ones for wily, vivid, and trusty too?07:59
pittiutlemming: with yesterday's live-build fix, I mean07:59
pittiutlemming: hm, actually, console-log doesn't show me *any* detected network card, so maybe that's unrelated08:00
pittiPSA: disabling autopkgtesting on ppc64el, this is just broken08:08
pittiLaney, wgrant, infinity ^ FYI -- collecting logs and filing RT08:08
pittiapw: ^ affects your kernel tests, too08:08
rbasakrharper: for Debian bug/patch submissions from your Ubuntu hat, please use https://wiki.ubuntu.com/Debian/Usertagging - it helps when people complain that we don't contribute to Debian.08:14
rharperrbasak: thanks08:15
dholbachgood morning08:17
pittiutlemming: ok, unping -- eth0 exists, but it doesn't get a DHCP address, this is an IS problem08:19
roaksoax-brbpitti: howdy! I'm seeing stuff like "maas-proxy.service is a disabled or a static unit, not starting it." all over for things that used to work last week.. any ideas why? my debian/rules: http://pastebin.ubuntu.com/14886037/08:22
pittiroaksoax-brb: no [Install] section?08:24
* pitti needs to leave for doctor appointment, back in ~ 1 h08:25
dokomitya57, building metacity I see:09:06
dokodh_autoreconf09:06
dokom4/glib-gettext.m4:39: error: m4_copy: won't overwrite defined macro: glib_DEFUN09:06
dokom4/glib-gettext.m4:39: the top level09:06
dokoLaney, ^^^09:07
Laneydoko: m4/, sounds like a toolchain problem :) :) :) :)09:13
dokoLaney, sure, it's called glib ;-P09:13
Laneydoko: I think you can delete that file since aclocal should get the system one09:15
Laneyit must be doing some kind of deduplication and that breaks since glib changed /u/s/aclocal/glib-gettext.m4 in a recent release09:16
Laneypreviously m4/glib-gettext.m4 and the system one would have been identical09:16
Laneyyou can copy one to the other to confirm this09:16
dokothat works09:22
Laneysee, toolchain problem!09:22
dokostarts with g* ...09:23
Laneythe macro was always defined twice09:23
Laneyaclocal was hiding this from us before09:24
Laneyhappy fix anyway09:24
Laneycjwatson: good work09:26
Mirvpitti: otherwise going fine, but interestingly https://autopkgtest.ubuntu.com/retry.cgi?release=xenial&arch=amd64&package=qtcreator-plugin-ubuntu&trigger=qtdeclarative-opensource-src-gles%2F5.5.1-2ubuntu4&ppa=ci-train-ppa-service%2Fstable-phone-overlay&ppa=ci-train-ppa-service%2Flanding-023 says "You are not allowed to upload qtcreator-plugin-ubuntu or qtdeclarative-opensource-src-gles to Ubuntu, thus y09:41
Mirvou are not allowed to use this service." even those are both universe packages and I'm a MOTU09:41
Mirvpitti: I was able to retry the same qtcreator-plugin-ubuntu tests under the non-gles qtdeclarative-opensource-src though09:42
xnoxcjwatson, i'm trying to figure out where and how built-using should be processed in germinate... It's a relationship declared on binary packages, but the values are source packages. I'm pondering if those sources' binaries should end up in extra.09:46
xnoxe.g. store them somewhere, and if binaries from those sources did not end up anywhere, add them into extra.09:46
xnoxor some such.09:46
pittiMirv: can you please file a bug against auto-package-testing? No off-hand idea right now, I'm afraid09:47
Mirvpitti: ok, I was just filing a bug against bileto so there it is, bug #1542239.09:49
ubottubug 1542239 in Bileto "Unable to run some retries" [Undecided,New] https://launchpad.net/bugs/154223909:49
roaksoax-brbpitti: this is my debian/rules http://paste.ubuntu.com/14886246/ I'm definitely missing something then09:50
pittiroaksoax-brb: what's the output of "systemctl status maas-proxy.service"?09:51
pittiroaksoax-brb: but I suspect these overrides are wrong -- can you just drop them? dh_systemd_* should already enable all units in that package, and --name doesn't exist as an option AFAIK09:53
pittiroaksoax-brb: sorry, --name does exist for dh_systemd_enable09:53
wgrantpitti: ppc64el vbuilders are still working fine, so I guess the neutron2 dnsmasqs, or possibly just your dnsmasq, is/are dead.09:54
wgrantWe've been seeing some related issues on vbuilders over the past two weeks, but they only hit the occasional boot.09:54
pittiwgrant: for me, I sometimes get bursts of failure mails where all of them failed, then it works again for half a day, etc.09:55
pittiright now it's again not working at all09:55
xnoxcjwatson, failing to grasp what "build_tree" flag is09:55
xnoxtoo09:55
xnox=(09:55
=== cpaelzer is now known as cpaelzer_afk
mitya57doko, Laney, so what should I do in metacity? dh_clean that file?10:07
Laneymitya57: I guess, not that sure why it is in m4/10:08
Laneyaclocal --install I guess10:10
Laneythis is annoying10:10
dokomitya57, for now, I just updated it10:12
cjwatsonxnox: build_tree basically means that germinate is processing something that it got to via a Build-Depends10:13
mitya57doko, I see, thanks10:14
Laneymitya57: updating it is a bad idea for unstable atm, since it only changed in exp10:15
xnoxcjwatson, aha. right, i see that in adding all the things read from e.g. Build-Depends-Indep et.al.10:15
cjwatsonxnox: Built-Using presumably wants to be specifically not build_tree though, otherwise what's the point, it wouldn't change anything10:16
xnoxi'm now wondering, if i should add all sources, as if it was the source package of a given binary. Or, like, find intersction of binaries(produced by Built-Using sources) && binaries(build-depends)10:16
xnoxand then seed those binaries.10:17
cjwatsonI feel like adding all binaries would probably cause a bit too much explosion10:17
cjwatsonBut getting the intersection is going to be really really really hard10:17
xnoxyeah.10:17
cjwatsonBecause it might be transitive10:17
xnoxespecially thanks to transitive build-deps.10:17
xnoxyou type faster =)10:17
xnox#not fair10:17
mitya57Laney, I hope new release will use upstream gettext, so there won't be m4/glib-gettext.m4 in tarball at all10:17
xnoxcjwatson, i shall study existing built-using for a bit.10:18
cjwatsonxnox: Maybe we just need to make sure it's annotated somewhere in germinate's output and then people can manually hike those up to seeds :-/10:18
xnoxcjwatson, given my vision for the new world order. I would have them show up in components missmatches, as source-only promotion to main, with .dot graphs indicating that these are built-using src packages.10:19
xnoxcjwatson, and that way e.g. security team would know -> it's in main, source only -> it must be the crazy statically linked in thing into _everywhere_10:19
pahello10:21
cjwatsonxnox: mm, causing Built-Using to include just the source but no binaries is probably quite a good way to handle this in germinate, indeed10:21
cjwatson(well, no binaries unless something else includes them, of course)10:21
xnoxyeah. and transitive build-deps are not broken, because new-world-order.10:21
paare you guys still supporting the decision of having a windows7 alike ctrl-tab on ubuntu touch?10:21
xnoxpa, most people here have no idea about windows7, nor ctrl-tab =) and i think touch development is done in another channel, like #ubuntu-touch (?!) i can't recall.10:22
cjwatsonxnox: well, even then, Built-Using is roughly always a subset of Build-Depends10:22
paok i try to ask there10:23
xnoxpa, do you have screenshots on the internet of this ctrl-tab thing?10:23
pai understand that in ubuntu desktop you already have plenty of problems ;)10:23
payes i can find10:23
pasec10:23
paxnox, this counterproductive thing: http://i1-news.softpedia-static.com/images/news2/Ubuntu-Touch-RTM-Update-10-Important-Milestone-Achieved-Screenshot-Tour-466165-8.jpg10:24
mitya57On phones it's not ctrl-tab, but rather some gesture (I guess) :)10:25
payes10:25
=== cpaelzer_afk is now known as cpaelzer
pabut on win7 it is :)10:25
xnoxpa, i totally misunderstood your question, i thought you were asking us to include something like that.10:26
pano i actually hope that to be dropped asap :)10:26
xnoxpa, anyway, such discussion is offtopic. And i think behaviour of those things is different on different form factors. I think on tablet/desktop things fly in expose mode, rather than in a carousel. Better ask in #ubuntu-touch.10:27
pai wish you guys won't end up like another firefoxOS :)10:27
paok i tyr10:27
patry10:27
xnoxpa, this carousel design seems reasonabl-ish.10:27
paxnox, it10:27
pait's not just useless10:27
pait's counterproductive10:27
paa bloody simple matrix is 10.000 times better10:27
xnoxbut it's *pretty* =)10:27
payes it seems pretty when you arent using it10:27
xnoxi use terminal, emacs in the terminal and a web-browser. I don't see myself having more than two windows like ever =010:28
pawell on a phone you might have multiple apps open at the same time10:28
pain particular because the app launch time on ubuntu phones is not small10:29
palike text app, mail app, calendar, browser, etc.10:29
pabut i'll ask there10:29
xnoxpa, i don't use, have or develop ubuntu touch UX / UI at all. try #ubuntu-touch, it should have more people who are actually involved in that.10:32
xnoxpeople here are scewed towards core-platform development.10:33
rbasakcjwatson: IIRC you once told me that in an Ubuntu merge Vcs-Browser should technically be changed to XSBC-Debian-Vcs or something, but I can't remember exactly what and my IRC log grep fails to find it. I've just received a merge request with this in the diff. Is there a normative thing that Ubuntu does here please?11:01
* rbasak found https://wiki.ubuntu.com/XS-Vcs but that doesn't have the answer.11:04
=== FourDollars_ is now known as FourDollars
cjwatsonrbasak: that links to a thread which ends up at https://lists.ubuntu.com/archives/ubuntu-devel/2007-March/023379.html, i.e. XS-Debian-Vcs-*, which is what I've always done11:19
cjwatsonrbasak: I don't remember if that was written down elsewhere11:19
cjwatson(XSBC- is silly because this doesn't belong in .deb control files or .changes, but XS- is reasonable)11:20
rbasakThat's great, thanks.11:20
roaksoax-brbpitti: tmaas-proxy.service: unrecognized service , the weird thing though, is that the overrides were working last week11:36
=== roaksoax-brb is now known as roaksoax
=== _salem is now known as salem_
=== cpaelzer is now known as cpaelzer_afk
marlincHow 'breaking' would it be to run 16.04 as daily driver until its released as actual stable12:11
marlincI'll be using on a ZFS root so I can make snapshot to go back when something does go wrong12:12
marlincdasjoe, ^ how bad would you say it is?12:16
marlincI can just try it12:16
=== kickinz1|afk is now known as kickinz1
xnoxmarlinc, well, you will always get to keep both pieces =)12:31
marlincGlueing them together would definitely work12:33
dasjoemarlinc: I try to stay on LTS releases, so you'll be on your own. Also, as of 15.10 I still prefer dajhorn's PPA to the official packages provided by cking, as dajhorn's PPA is at 0.6.5.4 and I'm not sure the official packages support rpools12:42
=== dholbach_ is now known as dholbach
marlincI've tried a rpool using the default packages in 16.04 on my USB. That appeared to work12:44
marlincAlthough I am unsure about GRUB2's ZFS feature flag support. I'm not sure if its in yet12:45
marlincAh, yes. I was looking for https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/153045712:46
ubottuLaunchpad bug 1530457 in grub2 (Ubuntu) "grub2: cherry-pick support for ZFS pool feature flags, bugfixes" [High,Fix released]12:46
seb128doko, http://launchpadlibrarian.net/236385287/libcmis_0.5.0-4ubuntu2_0.5.0-4ubuntu3.diff.gz ... did you forgot the actual change?12:49
seb128the diff only has a changelog entry12:49
dokoseb128, it's a permission change12:49
seb128oh ok12:49
seb128misleading changelog text :-)12:49
marlincWhen's the feature freeze?12:50
=== marlinc changed the topic of #ubuntu-devel to: Wily (15.10) Released! | Archive: open | Devel of Ubuntu (not support or app devel) | build failures: http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of precise-wily | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: darkxst
marlincWut, woop12:50
seb128marlinc, https://wiki.ubuntu.com/XenialXerus/ReleaseSchedule12:51
=== marlinc changed the topic of #ubuntu-devel to: Wily (15.10) Released! | Archive: open | Devel of Ubuntu (not support or app devel) | build failures: http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of precise-wily | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: darkxst
marlincThanks!12:51
pittiinfinity, cjwatson: any known problem with teh ppc64el builders?  I got a weird "chroot problem" complaining about some invalid zlib stuff on  unpack, and now https://launchpadlibrarian.net/236387344/buildlog_ubuntu-xenial-ppc64el.mbr_1.1.11-5ubuntu1_BUILDING.txt.gz12:56
pittisegfault during dpkg12:56
pitti(retry worked)12:59
=== didrocks1 is now known as didrocks
cjwatsonpitti: happens every now and again, incidence is low enough that it's tolerable13:03
pitticjwatson: ok; I got it three times in 15 minutes, so it felt like something broke13:03
cjwatsonpitti: can you see why qtcreator-plugin-ubuntu on https://requests.ci-train.ubuntu.com/static/britney/xenial/landing-046/excuses.html is sad?  it's installable in xenial13:07
smbpitti, Speaking of odd things... have you ever had "TZ=UTC date" not print the date in UTC? Asking because somehow the Xenial image I seem to pull with adt-buildvm-ubuntu-cloud does exactly that and I think that causes the build-tests of libvirt which I try to run through adt to fail. Felling somewhat dumbfound.13:10
smb(note that another Xenial VM which I manually installed works just like expected)13:12
=== cpaelzer_afk is now known as cpaelzer
=== cpaelzer is now known as cpaelzer_afk
=== cpaelzer_afk is now known as cpaelzer
menacehi, how big's the chance, that kernel 4.5 comes to 16.04?15:16
jrwreneventually, very likely ;]15:18
menacei meant on release (16.04.0) :-D15:18
michael-vbHello.  Bad moment to ask, as I have to leave almost at once, but could someone take a look at this (development) question re unity-settings-daemon?15:18
michael-vbhttps://answers.launchpad.net/unity-settings-daemon/+question/28446115:18
michael-vbThanks in advance.15:18
roaksoaxpitti: this is weird, postinst scripts do end up with: # Automatically added by dh_systemd_start -> which actually tells the service to start15:19
roaksoaxpitti: but maas-rackd.service is a disabled or a static unit, not starting it. -> still shown in install15:20
roaksoaxpitti: and the service ends ujp being stopped15:20
jrwrenmenace: i don't know, but i'd guess not likely15:21
menacejrwren: i hear that from several sources... but if one wants to support an ubuntu lts edition and has to write own externel (3rd party) kernel modules for that release, a little bite more reliability would be nice =) or even factors where i can recognize, if it gets more likely.... but okay then, thanks for your time.15:31
jrwrenmenace: I don't see the issue. gl hf15:34
marlinchttps://bugs.launchpad.net/ubuntu/+source/grub2/+bug/154235815:37
ubottuLaunchpad bug 1542358 in grub2 (Ubuntu) "GRUB does not include ZFS modules in EFI partition" [Undecided,New]15:37
cjwatsonmarlinc: ah yes, fair point, will fix15:41
cyphermoxcjwatson: there's another bug for this15:43
cyphermoxok, maybe not quite the same thing15:44
cyphermoxcjwatson: do you have anything against applying the changes from Chad in https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/1527727 too?15:45
ubottuLaunchpad bug 1527727 in grub2 (Ubuntu) "grub-probe for zfs assumes all devices prefix with /dev, ignoring /dev/disk/..." [Medium,Confirmed]15:45
cjwatsonthere's a different one about ZFS probing which will take a bit more thought15:45
cjwatsonI want to think about that patch :)15:45
cyphermoxyeah, that's really what I'm asking :)15:45
marlincAh thanks for taking a look cjwatson :) Any idea what's causing it or what I can do to help?15:47
cjwatsonmarlinc: just missing entries for those modules, I already committed a fix and uploaded to Debian15:48
marlincVery nice15:48
marlincI was told by someone that I had to test ZFS stuff on Xenial before feature freeze15:49
marlincI'm going all in by trying a rpool on Xenial15:50
marlincCan I follow it somewhere cjwatson?15:51
cjwatsonmarlinc: follow what, sorry?15:52
marlincThe Debian package, to see when it gets imported into Xenial for example15:53
cjwatsonmarlinc: well it'll show up in xenial in a day or so, once everything gets processed and published and synced and built and published etc.15:53
marlincThis is the first time I'm actually testing stuff for a new release so I've got no idea how those packages get handled between Debian and Ubuntu15:53
cjwatsonI'm not sure there's much that's desperately interesting to follow en route there :)15:54
marlincOkay :p15:54
marlincIs the patch available somewhere?15:54
cjwatsonbut it's https://anonscm.debian.org/cgit/pkg-grub/grub.git/commit/?id=a293f3b3349327b4e25f5e4998753f05aa7144c3, it'll show up on https://tracker.debian.org/pkg/grub2 in a little bit, then on https://launchpad.net/debian/+source/grub2, then eventually on https://launchpad.net/ubuntu/+source/grub215:55
marlincAh interesting, thanks!15:55
pete-woodsmdeslaur: hi. I've been looking at getting gnome-keyring running on the phone for password storage, but initially it crashed in some of the ARM asm inside libgcrypt15:57
pete-woodsand as the only canonical person I recognise on the changelog for that project, was hoping you'd take a look at my MR15:57
pete-woodshttps://code.launchpad.net/~pete-woods/ubuntu/wily/libgcrypt20/disable-arm-asm-rijndael/+merge/28489315:57
pete-woodswhich basically just disables the ARM assembler for AES15:58
pete-woodsafter this change, gnome-keyring functions correctly15:58
pete-woodsI'm happy to revise the patch to follow patch conventions (as you can see from the MR message, I was initially unsure if this change would even help :) )15:59
mdeslaurpete-woods: hrm, sure, one se16:01
pete-woodsthanks!16:06
mdeslaurpete-woods: I think that makes sense, but please file a bug upstream about it16:09
mdeslaurI don't see any further fixes in the git tree16:09
pete-woodsmdeslaur: okay, cool, will report the upstream bug16:12
=== sil2100_ is now known as sil2100
marlinccjwatson_, do you know of a way to fix my grub image so that the workaround can be avoided?16:19
=== cjwatson_ is now known as cjwatson
cjwatsonmarlinc: since you're already running xenial, I don't think it's worth expending energy on thinking about workarounds16:25
cjwatsonmarlinc: upgrade in a day or two instead :)16:25
marlincOkay :p16:29
smbpitti, Ok, to answer my question about the weird date output in the adt test image... Something (not sure what and when) screwed up the zoneinfo, so /usr/share/zoneinfo/UTC pointed to something wrong. Forcing a reinstall of tzdata in the disk image fixes the problem.16:30
=== popey_ is now known as popey
bdmurrayDoes anybody know what the lxc error is here? http://pastebin.ubuntu.com/14888364/16:37
dannfdoko: did you have anything queued for a gccgo-4.9 trusty SRU? you cool w/ me uploading for LP: #1542080?16:40
ubottuLaunchpad bug 1542080 in gccgo-4.9 (Ubuntu Trusty) "Needs tar/xattr support to build docker" [Undecided,Triaged] https://launchpad.net/bugs/154208016:40
dannfdoko_: ^16:40
doko_dannf, sure, go ahead16:41
dannfdoko_: thx16:41
dasjoecking: cjwatson: Thanks again for your work on our favorite file systemâ„¢! Now, let's get it into snappy ;)16:45
ckingwere moving forward ;-)16:46
=== chrisccoulson_ is now known as chrisccoulson
pete-woodsmdeslaur: FYI, here is the upstream bug report for the libgcrypt crash: https://bugs.gnupg.org/gnupg/issue224217:14
pete-woodsokay, I gotta EOD and pick up kids17:16
pete-woodsthanks for your time!17:16
=== marcoceppi_ is now known as marcoceppi
=== tvoss is now known as tvoss|dinner
jtayloris util-linux a package one can safely backport?18:31
=== doko_ is now known as doko
=== tvoss|dinner is now known as tvoss
marlincAnyone running into issues with Rhythmbox and GStreamer? I'm getting the following error when trying to play a mp3 file. (21:21:14) [0x1dc1c40] [rb_shell_player_error] rb-shell-player.c:2443: playback error while playing: Problem occurred without error being set. This is a bug in Rhythmbox or GStreamer.20:21
marlincPlayback does however work in the Videos app and in VLC20:21
marlincI've installed the codec as Video's asked for it20:21
marlinchttps://bugs.launchpad.net/ubuntu/+source/gstreamer1.0/+bug/154247120:30
ubottuLaunchpad bug 1542471 in gstreamer1.0 (Ubuntu) "Playing mp3 files causes: Problem occurred without error being set. This is a bug in Rhythmbox or GStreamer." [Undecided,New]20:30
=== salem_ is now known as _salem
marlincInteresting, when I now Google the issue I'm having with Rhythmbox I actually get my bug report as first result23:26

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