/srv/irclogs.ubuntu.com/2015/08/03/#ubuntu-devel.txt

codepython777has anyone recently bought a laptop here? I'm looking to buy a good ubuntu laptop. Had my eye on dell xps 13 - but that has been discontinued for now.01:56
=== FJKong_afk is now known as FJKong
=== BinLi is now known as BinLi_afk
pittiGood morning04:26
=== marcusto_ is now known as marcustomlinson
dholbachgood morning07:04
=== ara is now known as Guest14068
dholbach@pilot in07:16
=== udevbot changed the topic of #ubuntu-devel to: Archive: wily open | Devel of Ubuntu (not support or app devel) | build failures -> http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of precise -> vivid | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: dholbach
seb128hey dholbach, enjoy ;-)07:17
dholbachthanks07:17
Noskcajdholbach, 3.8.3 is the newest non-bluez5 version07:29
dholbachah ok07:29
NoskcajIt just brings in a few bug fixes, but it would be nice to have07:29
dholbachright07:29
NoskcajWe're discussing bluez5 in -desktop now, it will happen later this cycle if we can get the manpower finally07:29
=== dbarth__ is now known as dbarth
=== oSoMoN_ is now known as oSoMoN
dholbachNoskcaj, do you have a link to the source package of 3.8.0-2?09:04
NoskcajI thought i got the merge up to 3.8.3-109:05
Noskcajhttps://download.gnome.org/sources/gnome-user-share/3.8/gnome-user-share-3.8.3.tar.xz09:05
Noskcajhttps://download.gnome.org/sources/gnome-user-share/3.8/gnome-user-share-3.8.0.tar.xz if not09:05
dholbachthe source package09:05
dholbachlike the .dsc file and everything09:05
Noskcajoh, no. I just got it from the pkg-gnome tags09:06
dholbachmh09:06
dholbachnevermind... http://snapshot.debian.org/package/gnome-user-share/3.8.0-2/09:07
dokoseb128, I saw you didn't forward libproxy to debian. are there other uploads which are not yet forwarded?09:35
seb128doko, yes, I didn't manage to go through all the diffs on thursday and I was on friday, I'm going to do the remaining ones after lunch09:37
dokothanks09:45
hron85Hi, anyone can help me with working with quilt? I made an upgrade for a package and now I have a non-applying patch. However, I cannot refresh it, since it does not apply and quilt do not want to move to this patch until it does not apply09:55
hron85quilt refresh <patchname> does not work09:55
pittiLaney: in case you had some config or other files: I just redeployed the entire autopkgtest cloud stuff from scratch09:57
Laneypitti: nope, maybe stuff in shell history that'll be gone though. :P09:58
pittiLaney: I did some work this morning to add basenode/ksplice/landscape/two nova configs, and wanted to re-test the whole thing before we switch to production09:58
Laneywhat is basenode?09:59
* Laney can't quite google it up09:59
cjwatsonhron85: quilt push -f, resolve conflicts (it should tell you about where they are, and IIRC there'll be the usual patch-conflict <<< === >>> markers in the conflicted files), quilt refresh10:02
cjwatsonhron85: assuming that your starting point is a consistent tree whose quilt state is at the patch immediately below the non-applying one10:03
hron85cjwatson: yep, it solved my problem, thanks10:14
hron85cjwatson: yes, I knew the not-applying patch will be the next, but i did not found any way to move to it, since quilt refresh <patchname> rejects to work if the mentioned patch is not applied yet.10:15
jhenkedoes anybody know if someone is working on the proposed gnupg 2.1.x as default? I did not see anything since the original message and I wondering if I missed some message10:34
ricotzStevenK, hi, a gcc5 rebuild of aptitude please :)10:38
ricotzah sorry, slangasek ^10:39
cjwatsonhron85: right, in general I think it's best to avoid quilt refresh's mode where it takes a patch name - it's conceptually very confusing10:41
Laneyseb128: doko: do you want help with rebuilds/uploads? if so, where?10:50
dokoricotz, current ftbfs10:51
dokoLaney, see slangasek's message on -release10:51
Laneyhmm, I don't have one10:52
Laneynor on the archive10:52
Laneylink?10:52
dokoLaney, #ubuntu-release10:58
Laneystill unclear11:00
LaneyI see the individual transitions - should I take one and work on it?11:00
Laney+ checking your build logs for ABI changes11:00
ricotzdoko, oh, I see11:09
=== henrix_ is now known as henrix
=== utlemming is now known as utlemming_away
=== utlemming_away is now known as utlemming
=== utlemming is now known as utlemming_away
=== psivaa is now known as psivaa-lunch
=== utlemming_away is now known as utlemming
alkisgmitya57: hi, I'm trying to look into the gnome-flashback keyboard layout applet issue, is the problem there that the existing applet doesn't respond to layout changes reported by xorg?13:01
dholbach@pilot out13:39
=== udevbot changed the topic of #ubuntu-devel to: Archive: wily open | Devel of Ubuntu (not support or app devel) | build failures -&amp;gt; http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of precise -&amp;gt; vivid | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots:
dokoseb128, Laney: just updated to wily. the handles for the terminals don't exist. is this expected?13:50
dokoalso when trying to update to wily-proposed:13:50
dokoThe following packages will be REMOVED:13:50
doko  aptitude compiz compiz-gnome compizconfig-settings-manager libasprintf0c2 libcompizconfig0 libdouble-conversion1 libllvm3.6 libmirclient8 libprotobuf9 libproxy1 libtag1-vanilla libxapian2213:50
doko  mumble python-compizconfig python-xapian signon-ui signon-ui-x11 software-center tasksel tasksel-data ubuntu-desktop unity unity-control-center-signon unity-tweak-tool13:50
doko  webaccounts-extension-common xul-ext-webaccounts13:50
seb128doko, the scrollbar issue is known I think, Laney has the details13:53
seb128doko, compiz/unity rebuild/fixes for gcc5 are in silo43 it seems, check with bregma when he plans to land it I guess13:54
seb128dunno about aptitude mumble etc13:55
dokobregma, ^^^ ???13:55
bregmadoko, there's an issue with one of the branches landing that still needs to be retested, then I'm landing that silo as soon as I can13:56
dokota13:57
bregmait's been a game of whack-a-mole getting everything to build13:57
dokobregma, are there any other silos needed?14:00
bregmadoko, that silo installs OK as long as -proposed is enabled, so not for the Unity stack as far as I know14:01
pittiLaney: FYI: https://code.launchpad.net/~ubuntu-release/+git/autopkgtest-cloud (moved from ~pitti)14:09
dokojamespage, zul, roaksoax, smoser: there doesn't seem to be any progress on the ruby dep-waits ...14:12
pittiapw: not sure if you actually still need this, but http://autopkgtest.ubuntu.com/data/packages/wily/amd64/l/linux/latest.json (and all other JSONs) have URLs to the swift artifacts nwo14:13
apwthat looks useful, even if i don't use it, though i suspect i will14:14
cjwatsondoko: which ruby dep-waits are those, the ones on gem2deb?14:15
cjwatsondoko: because if so (and they quite probably involve gem2deb somewhere), don't hassle the server team about it, needs the next launchpad-buildd rollout because gem2deb started making use of build profiles14:16
dokocjwatson, ruby-hashie, ruby-rspec, ruby-safe-yaml, ruby-test-unit14:16
dokoahh, ok14:16
cjwatsondoko: there might be some MIRs buried in there too14:17
dokodpkg-gencontrol: warning: can't parse dependency libxapian22 (>= 1.2.21)v514:17
dokohrm,14:17
cjwatsonah, no, just misleadingly wrong dep-waits I think14:17
cjwatsondoko: certainly at least ruby-hashie and ruby-rspec will be ultimately fixed by new buildd, haven't checked them all14:18
cjwatsonhopefully I'll get a review of https://code.launchpad.net/~cjwatson/launchpad-buildd/snapcraft/+merge/266541 soon and then I can start the release process14:18
dokook, good to know ... still we should address the dep-waits too14:18
cjwatsondoko: what kind of addressing are you talking about?14:19
dokowriting MIR's, dropping b-d's etc14:20
cjwatsondoko: well, some of the dep-waits you see are bogus14:20
cjwatsonso it's probably not worth chasing any of that down until the buildds are fixed14:20
dokosure, but I doubt that the perl team uses these too14:21
cjwatsondoko: ruby-test-unit -> ruby-power-assert looks like a real MIR case14:21
cjwatsondoko: perl example?14:21
dokosee http://qa.ubuntuwire.com/ftbfs/14:22
cjwatsonok, sure, there's probably a bunch of MIR handling to be done there14:25
=== psivaa-lunch is now known as psivaa
Laneydoko: handles? we changed the scrollbars and there's a problem with the terminal's widget but the thumb is gone if that's what you mean14:59
Laneyyou can still scroll it but it's annoying to make it go big15:00
Laneypitti: ah, cool15:00
Laneypitti: do you plan to document the architecture? :)15:01
pittiLaney: yes, of course15:01
dokoLaney, indeed very ugly15:04
Laneypitti: great - I'm currently still a bit sketchy and would be scared if I had to (e.g.) redeploy it :P15:05
Laneydoko: we will try to tweak more before release15:05
pittiLaney: heh, sure; indeed there is one remaining bug for complete re-deployment15:06
pitti(bug 1480962)15:08
ubottuError: Could not gather data from Launchpad for bug #1480962 (https://launchpad.net/bugs/1480962). The error has been logged15:08
pittiLaney: once that's fixed, it will become really simple; I'll write documentation of all of that once it's all in place15:08
Laneyyay15:08
pittitseliot: hmm, seems something broke the fglrx driver? http://autopkgtest.ubuntu.com/packages/u/ubuntu-drivers-common/wily/amd64/15:43
tseliotpitti: having an actual build log would help. I'll try it here, thanks15:46
pittitseliot: yeah, we should fix those tests to cat the build log on failure15:46
tseliotpitti: that would be nice15:46
smoserbdmurray, i gues you're not awake...15:50
smoserhttps://code.launchpad.net/~smoser/ubuntu-reports/trunk.fix-server-merges/+merge/26674515:50
bdmurraysmoser: I am actually and in a meeting. I'll have a look shortly.15:51
* pitti waves good night15:52
smoserbdmurray, ok. thanks. and then kick the report generation if you could.15:52
tseliotpitti: the module built but I see two architectures in my chroot: dkms status15:53
tseliotfglrx-core, 15.200.1, 4.1.0-3-generic, amd64: installed15:53
tseliotfglrx-core, 15.200.1, 4.1.0-3-generic, x86_64: installed15:53
pittitseliot: did you enable -proposed?15:59
tseliotpitti: not really15:59
tseliotpitti: this is all I have http://paste.ubuntu.com/11993942/16:00
pittitseliot: i. e. it's plausible that it fails due to gcc 5; /me tries in a chroot with -proposed16:00
tseliotpitti: gcc -v reports "gcc version 4.9.2 (Ubuntu 4.9.2-17ubuntu1)"16:01
pittitseliot: right, and in -proposed we switched to 5.16:01
tseliotpitti: right, but I don't have -proposed enabled16:02
tseliotso it must be something else16:02
tseliotor maybe the failure is a separate issue and/or a false positive16:03
pitti(wily-amd64)root@donald:/home/martin# modprobe fglrx16:05
pittimodprobe: ERROR: could not insert 'fglrx': Unknown symbol in module, or unknown parameter (see dmesg)16:05
pitti[  404.772855] fglrx: Unknown symbol amd_iommu_bind_pasid (err 0)16:05
pitti[  404.772925] fglrx: Unknown symbol amd_iommu_set_invalid_ppr_cb (err 0)16:05
pitti[  404.773082] fglrx: Unknown symbol amd_iommu_set_invalidate_ctx_cb (err 0)16:05
pitti[  404.773170] fglrx: Unknown symbol amd_iommu_unbind_pasid (err 0)16:05
pitti[  404.773265] fglrx: Unknown symbol amd_iommu_init_device (err 0)16:05
pitti[  404.773372] fglrx: Unknown symbol amd_iommu_free_device (err 0)16:05
pittitseliot: ^ I get that with -proposed16:05
tseliotpitti: ok, so it's not a build error. It looks like fun. I'll fix it, thanks16:06
pittitseliot: hm, I get the same without -proposed16:06
pittiit started failing on Aug 116:07
tseliotmaybe the ABI broke somehow?16:08
pittitseliot: hm, maybe this is normal in a chroot; that's not what the test fails on16:08
pittiso this needs a closer look16:08
mitya57slangasek, ScottK is right — only the plugin(s?) are built against default python3, everything else is built against all supported versions16:08
mitya57(sorry for the delay btw)16:08
tseliotpitti: yes, I'll test it on a real system16:08
pittitseliot: cheers16:08
mitya57alkisg, hi, if by "existing applet" you mean indicator-keyboard, then it doesn't support xkb switching. It listens to gsettings changes on org.gnome.input-sources (that are done by g-s-d or u-s-d).16:12
mitya57And that should work fine in Ubuntu (unlike Debian or upstream).16:13
alkisgmitya57: it works here in 14.04, but not in 15.1016:13
alkisgI can use indicator-keyboard with the mouse to change the language, but I can't use any keyboard shortcut to change it16:14
alkisgEven though gsettings are the same as in 14.04 when it worked16:14
alkisgmitya57: so the plan is for someone to write such an applet, and then use it in all variants, upstream, debian and ubuntu?16:16
alkisgOr ubuntu will keep using indicator-keyboard?16:16
mitya57alkisg, it's in our "patches welcome" list upstream :)16:18
mitya57let me look at the code a bit16:18
alkisgIf you accept one that listens for *xkb* changes as well, instead of only for gsettings changes, it would be interesting... :)16:19
mitya57alkisg, why do you still need xkb changes? GNOME upstream deprecated that and gsettings is much easier to use.16:25
mitya57alkisg, also, do you have gnome-settings-daemon or unity-settings-daemon running?16:25
alkisgmitya57: unity-settings16:25
mitya57that should work fine…16:26
alkisgIt doesn't... clean 15.10 installation16:26
alkisgThat was basically what I was worried about, I thought that it would be broken in ubuntu 15.10+, 16.04 etc like it is upstream...16:26
mitya57Are you using modifiers-only keybinding or not?16:26
alkisgThe defaults, win+space16:27
alkisgI also tried alt+shift which worked in 14.04, it doesn't work either16:27
mitya57Does it list in gsettings get org.gnome.desktop.wm.keybindings switch-input-source?16:27
alkisg<Super>space16:28
mitya57Looks fine.16:28
mitya57Is there a bug open in LP for it, and if no, can you please file one?16:28
alkisgI'll check + file a bug, thanks,16:29
alkisgabout xkb changes, unfortunately gnome upstream breaks sdl apps, and we're trying to make things work as they were with xorg without gnome breaking them...16:29
mitya57Can you also attach "gsettings list-recursively org.gnome.desktop.input-sources" to the bug, please?16:30
alkisgE.g. in tuxpaint, tuxtype, we cannot type greek at all now with recent gnome versions, the last one to work fine for us was 12.0416:30
alkisgmitya57: another issue that might need resolving,16:30
alkisgI install ubuntu with greek layout in ubiquity etc. Then I create a user, and /home/user is empty16:31
andolThere is something off with the http://releases.ubuntu.com/vivid/ FOOSUMS files, and their corresponding FOOSUMS.gpg - http://paste.ubuntu.com/11994117/. In addition to gpg reporting BAD signatures there is also the oddness that the FOOSUMS files appear to have newer mtimes than their corresponding FOOFUMS.gpg.16:31
alkisgIf that user logs in, he has nothing in the layouts list, while it should be "us,gr"16:31
andol(No idea where the proper place to report such an issue is)16:31
alkisgSo all users need to go to keyboard settings and add the layouts manually16:31
alkisgI think this happens in all ubuntu gnome variants, while it doesn't happen in e.g. fedora... or of course in ubuntu <= 12.0416:32
mitya57alkisg, please report that against the installer (i.e. if you use the graphic one, then it's ubiquity)16:32
alkisgmitya57: I think the problem is that gnome doesn't respect the global xorg settings anymore16:33
alkisgIt's not about ubiquity... e.g. when one run `adduser x` after the installation, that "x" user doesn't get the xorg system default layout16:33
alkisg*runs16:33
mitya57alkisg, are you installing Ubuntu GNOME?16:33
alkisgNo, plain ubuntu and then gnome-flashback16:33
alkisgIt's been that way since 14.04. It was working in 12.0416:34
alkisgI've been filling a lot of bug reports about these issues, but they didn't draw much attention...16:34
mitya57This is first time I hear about this issue.16:35
mitya57:)16:35
alkisg:)16:35
mitya57It looks like it's not specific to GNOME Flashback, so try sending a mail to ubuntu-desktop@lists.ubuntu.com16:35
mitya57There are people who should know what component to blame16:35
alkisgThe long story is that ubiquity, lightdm, accountsservice, gnome-settings-daemon, unity-settings-daemon, and a couple of other applications don't work with multiple keyboard layouts since 13.04 or so16:36
alkisgLet me try the indicator with unity...16:37
alkisgmitya57: I think that the "indicator doesn't change on layout changes by keyboard shortcuts" is specific to flashback, in unity it's working fine16:38
alkisgAnd it broke after 14.0416:38
alkisgThe 10+ rest issues about layouts are not specific to flashback, but they are specific to ubuntu though... :-/16:39
* alkisg checks + files a bug for that one...16:40
mitya57alkisg, does the gsettings key (i.e. org.gnome.desktop.input-sources current) change when you toggle the keybinding?16:40
alkisgunity => yes, moment for flashback...16:41
alkisgmetacity => no, moment for compiz...16:41
infinitypitti: Did I ever get those trusty langpacks?16:42
alkisgmitya57: gnome-flashback (compiz) gives me unity without the top panel... go figure :)16:44
mitya57alkisg, metacity is enough, that indicates that the problem is not in the indicator, but in unity-settings-daemon16:44
mitya57the fix for compiz will be in next compiz upload :)16:44
alkisgmitya57: hope you don't mind me asking, are you using the flashback+metacity session yourself in production?16:45
alkisgWe're using gnome-flashback in 1000+ schools here, but I was wondering if mate would be more appropriate, considering the old hardware...16:46
alkisggnome-flashback feels like the best choice, considering it's closer to upstream gnome, but it also feels understaffed... :)16:46
mitya57Yes, we are quite lacking resources16:50
mitya57Actually I do just packaging (don't have time for anything else), all upstream work is done by 1.5 people :)16:50
alkisgYes fortunately Alberts is doing the bulk of the work... but mitya57, are you using it in production somewhere? Or is it just your preferred DE?16:52
mitya57Many my friends use it :)16:53
alkisgNice :)16:53
mitya57Also I set it up in a school some time ago, not sure if they are still using it16:53
* mitya57 keeps accidentally pressing Ctrl+Q16:53
* alkisg wonders where "ibus" fits in all this... it's preinstalled, and previously it was not running by default, and now it is...16:54
* mitya57 doesn't know anything about ibus16:55
alkisgmitya57: should I file the bug against indicator-keyboard (Ubuntu), or against unity-settings-daemon?17:00
alkisgI.e. that it works in unity but not in flashback17:00
mitya57alkisg, unity-settings-daemon, and also add https://launchpad.net/ubuntu-gnome-flashback as a subtask17:05
alkisgThanks, trying...17:05
dokowhy is xapian-bindings ftbfs on the buildds ...17:08
dokobregma, so even with your landing PPA enabled, I can't upgrade17:14
alkisgmitya57: I've filed LP #148102517:20
ubottuLaunchpad bug 1481025 in unity-settings-daemon (Ubuntu) "Keyboard shortcut for layout switching works in Unity but not in Gnome-Flashback" [Undecided,New] https://launchpad.net/bugs/148102517:20
Mirvcyphermox: thanks for the new wpa! fwiw it does not seem there is stated delta anymore to Debian's debian/config/wpasupplicant/linux, other than the fact that Ubuntu has the settings specified twice + there is unmentioned Android config delta17:27
cyphermoxMirv: are you saying I forgot to re-add the android crap in changelog?17:35
cyphermoxlooks to me like it's there17:36
mitya57alkisg: thanks17:36
alkisgTthank _you_ :)17:38
dokoNoskcaj, qqwing accepted in -proposed17:46
hjdcyphermox: Thanks for packaging. :) I suppose it's too late to mention it in the changelog, but would you mind leaving a comment on bug 1370245?17:47
ubottubug 1370245 in wpasupplicant (Ubuntu) "wpa_supplicant version 2.2 available" [Undecided,Confirmed] https://launchpad.net/bugs/137024517:47
Mirvcyphermox: the section stating remaining changes in debian/config/wpasupplicant/linux from Debian mentions the two options that are double stated in that file but does not mention the CONFIG_ANDROID_HAL=y which is the only actual delta in that config18:01
cyphermox  - debian/config/wpasupplicant/linux: enable CONFIG_ANDROID_HAL.18:02
Mirvcyphermox: I mean, oh, right, it's up there. then the later mention is just unneeded + the config file should be cleaned since the delta is actually now this http://paste.ubuntu.com/11994721/18:03
cyphermoxnah18:03
cyphermox*bah18:03
Mirvso CONFIG_P2P and CONFIG_AP are repeated in Ubuntu's version18:04
cyphermoxMirv: how about that. Feel free to clean it up if you want to18:06
hjdjjohansen: Hi :)19:05
jjohansenhjd: hello19:06
hjdjjohansen: Sorry for the random ping, but I saw bug 1481039 and wondered whether the linux-lts-backport-maverick and -natty tasks are included for completeness, since they don't seem to be part of any current Ubuntu release.19:06
ubottubug 1481039 in linux-manta (Ubuntu Wily) "CVE-2015-5697" [Low,New] https://launchpad.net/bugs/148103919:06
jjohansenhjd: its actually because I need to do some updates to the tooling we use so we can drop those19:07
hjdok :)19:11
doko_desktop updated to wily-proposed + landing-043. seems to work again \o/19:44
infinitydoko_: Impressive.19:46
Laneyoh, they fixed the arm64 FTBFS of $package19:47
LaneySHIP IT.19:47
doko_apt-get dist-upgrade removes gnome-control-center, but you can install it again. only thing missing is aptitude19:48
NoskcajLogan, Would you mind checking the sponsors queue for stuff before you merge? I had a finished merge for light-locker waiting, so there was a bit of an overlap of work. That said, i should have asked you if i could merge it19:51
Laneydoko_: If you can get an easily understandable list of work to be done then I can help out tomorrow19:52
Laneylike 'rename these packages' or 'rebuild/fix for these transitions'19:52
doko_I'm trying, need to prepare for travel tomorrow as well19:54
doko_I mean, you could starting to rebuild stuff for icu and boost in main19:54
LaneyI also don't want to clash/duplicate19:55
Laneyso some pad/wiki would be nice19:55
doko_Laney, coordinate with slangasek; I think I won't do much tomorrow19:55
ari-tczewcjwatson: shouldn't the fakesyncs be synced automatically?20:28
ari-tczewcjwatson: nvm, due to mismatch tarballs isn't possible, but I think fakesyncs should be on MoM visible...20:33
ari-tczewcorrect me if I'm wrong20:33
doko_cyphermox, you still involved with libcolumbus?  could you have a look at the ftbfs?21:45
=== doko_ is now known as doko
dokoNoskcaj, pyicu 1.8 ftbfs with ICU 55. updated to 1.9.221:55
Noskcajdoko, I'll update it in debian python svn today22:09
dokota22:10
infinityrobert_ancell: Hey, do you know of anyone (perhaps you?) who could verify the SRU for https://bugs.launchpad.net/ubuntu/+source/ubuntu-drivers-common/+bug/1410801 ?22:25
ubottuLaunchpad bug 1410801 in HWE Next "[AMD Richland + Topaz XT ] gpu-manager generated xorg.conf is not compatible with switchable graphics" [Medium,Confirmed]22:25
robert_ancellinfinity, I don't have any AMD hardware, sorry22:29
mwhudsongo 1.5 prep feels like http://media.giphy.com/media/MRUj8R8SMWlIk/giphy.gif today23:39
infinitymwhudson: You keep getting attacked by a vicious cat?23:57
mwhudsonnot entirely what i was trying to convey, no23:57

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