/srv/irclogs.ubuntu.com/2016/11/22/#ubuntu-devel.txt

=== Foxtrot is now known as foxtrot
smosernacc, hey.02:21
smoserusd import -v --directory=isc-dhcp --lp-user=smoser --lp-owner=usd-import-team isc-dhcp02:21
smoserKeyError: "Cannot describe - No tags can describe '02c1acc925e9b52df3cdb807c525d56464102d26'."02:21
smosernever mind. seems better now (after a git pull)02:23
smoserwell, now i get02:23
smoser NameError: name 'remote_namespace' is not defined02:23
smosernacc, i pushed fix02:35
=== JanC is now known as Guest78373
=== JanC_ is now known as JanC
=== salem_ is now known as _salem
pittiGood morning06:15
cpaelzer_good morning06:40
=== cpaelzer_ is now known as cpaelzer
cpaelzerhi pitti, thanks for pushing gnutls through06:40
pitticpaelzer: no worries06:41
cpaelzertjaalton: hi, since you have the required "powers" and were related to bug 1640805 in the past, could you take a look at reviewing and uploading the fix I attached?07:54
ubottubug 1640805 in sssd (Ubuntu) "SSSD authentication fails after upgrade to 1.11.8-0ubuntu0.2" [High,Fix released] https://launchpad.net/bugs/164080507:54
cpaelzerThat is the Trusty SRU portion of it which needs a sponsor07:54
tjaaltoncpaelzer: yep, uploaded08:01
cpaelzerthank you tjaalton08:01
tjaaltonand committed to git08:01
cpaelzertjaalton: sorry I didn't check on the VCS - thank you08:01
tjaaltonno worries, easy to apply the debdiff08:01
cpaelzertjaalton: would that have been a LP git or is the Ubuntu branch on alioth08:02
* cpaelzer is looking up the pkg08:02
tjaaltonon alioth08:02
cpaelzerah ok08:02
tjaaltoni'm not sure how to best integrate with lp08:02
mwhudsonpitti: hello09:03
pittihey mwhudson, how are you?09:06
mwhudsonpitti: fine, wondering about autopkgtests again09:07
mwhudsonpitti: is the process by which the tests to run for a given upload to -proposed documented somewhere?09:07
pittimwhudson: the top-level page is https://wiki.ubuntu.com/ProposedMigration; it links to various details09:08
mwhudsonpitti: in particular, does a depends on $foo in $bar's debian/tests/control mean that an upload of $foo will run $bar's tests?09:08
pittimwhudson: yes, britney has done reverse test dependency triggering for a few months now09:08
mwhudsonpitti: great, thanks09:09
mwhudsonpitti: the wiki is slightly out of date then i think :-)09:09
smbgood to know :)09:09
pittiwe hadn't had that for many years, but it led to hard-to-diagnose regressiosn09:09
pittimwhudson: I don't think "out of date", but I'm happy to clarify "reverse dependencies" to ".. binary and test .."09:10
mwhudsonpitti: ok, i have the edit page open anyway :)09:10
mwhudsonhttps://wiki.ubuntu.com/ProposedMigration?action=diff&rev2=25&rev1=2409:11
mwhudsonpitti: anyway, thanks09:11
pittithanks09:11
=== _salem is now known as salem_
=== hikiko is now known as hikiko|ln
tinococpaelzer: https://bugs.launchpad.net/qemu/+bug/162697212:09
ubottuLaunchpad bug 1626972 in qemu (Ubuntu Zesty) "QEMU memfd_create fallback mechanism change for security drivers" [Undecided,In progress]12:09
tinocois done for you. waiting on sponsors now12:09
tinocoill continue working with vhost log upstream in parallel12:10
tinococpaelzer: btw, are u a sponsor by any chance ;) if you are, feel free to test/upload it :o)12:10
cpaelzertinoco: hi12:13
tinococpaelzer: o/12:13
cpaelzertinoco: I just uploaded a minor fix to zesty which means one more debdiff for you - but not more I hope12:14
tinococpaelzer: i didn't do zesty12:14
tinocosince it was behind xenial12:14
cpaelzerbehind yakkety I know12:14
tinocooops, yakkety12:14
cpaelzerbut I did the upload while you were writing this :-)12:15
cpaelzerjust replied to you via mail as well12:15
tinococpaelzer: ok, let em check then. i can provide the debdiff, no problem12:15
cpaelzertinoco: ok, I will throw all these at a ppa of mine and run some more tests against it12:17
cpaelzertinoco: opening a query to discuss what you already tested ...12:17
tinococpaelzer: awesome!12:17
=== hikiko|ln is now known as hikiko
dokois anybody able to start quassel-client on a Unity desktop? the UI doesn't come up. on all of xenial, y and z. happened with a recent update ...13:37
=== Son_Goku is now known as Conan_Kudo
=== Conan_Kudo is now known as Son_Goku
dokopython-vmware-nsxlib/amd64 unsatisfiable Depends: python-netaddr (>= 0.17.3)14:43
dokopython3-vmware-nsxlib/amd64 unsatisfiable Depends: python-netaddr (>= 0.17.3)14:43
dokozul: ^^^14:43
zuldoko: cool thanks ill have a look today14:43
naccsmoser: thanks, i was just about to fix the same way, thanks!16:31
naccsmoser: sorry for the breakage16:31
naccrbasak: i think i found another attribute case, if a .gitattributes has specific files listed with eol, then '* -text' doesn't help :/16:34
slangasekinfinity: stgraber: kees: mdeslaur: TB meeting in 18minutes16:42
mdeslaurack16:43
slangasekrbasak: oh, you can't assign bugs to us?  well that's not good16:59
slangasekrbasak: on Gunnar's bug I ran the commands you requested, but that doesn't put any packages in that packageset; is that part something the DMB can take care of or should I do that too?16:59
naccsmoser: just pushed a small fix for gbp import-orig, fyi17:18
naccsmoser: to use the new tag namespace correctly on import17:18
hjdHow can transition status be unknown? (http://people.canonical.com/~ubuntu-archive/transitions/html/html/glew.html)17:48
hjdAlso, if I'm reading http://people.canonical.com/~ubuntu-archive/proposed-migration/update_output.txt correctly, the on-going glew transition is why for instance https://launchpad.net/ubuntu/+source/widelands hasn't migrated to zesty-release, right?17:49
Laneyhjd: Affected but neither good nor bad17:49
rbasakslangasek: assign bugs> yeah, that was surprising to me. I guess I need to be a member of the community team or something.18:01
rbasakslangasek: yes, I can take care of populating the packageset. Thanks!18:01
Laneyrbasak: FYI edit-acl has a 'copy' action18:01
LaneyWhich creates and copies the packages18:02
LaneyMight want to advise the TB to use that next time18:02
rbasakOh, neat. Thanks!18:02
rbasak"<slangasek> I suggested using the bug tracker"18:08
rbasakslangasek: I don't see an email. Did I miss a note somewhere?18:08
rbasakslangasek: also I'd be quite happy if the process was manual - I file a bug *and* send to the ML. As long as I can forget once fired :-)18:10
=== JanC_ is now known as JanC
naccsmoser: fyi, another fix pushed!18:27
slangasekrbasak: I did cc: you on my email reply18:32
naccsmoser: what would be the ideal way (given our usd.run) to invoke `git tag -a` for creating an annotated tag (using the git-tag spawned editor)?18:52
naccsmoser: nm, i think i figured it out18:54
smoser run.run(['git', 'tag', '--annotate', 'foo'], stdin=None, stdout=None, stderr=None)18:56
smosernacc, i think that works18:56
naccsmoser: well, then how would you get input to the tag command? it'll just sit forever, afaict18:56
naccsmoser: i think run() doesn't actually work without a string as an argument if shell=True18:56
naccsmoser: it ends up calling 'sh -c git tag -a upload/1%9.10.3.dfsg.P4-10.1ubuntu2 HEAD'18:57
naccsmoser: note the missing "" around the command18:57
naccsmoser: if i put "" around that command (e.g., sh -c "git tag -a upload/1%9.10.3.dfsg.P4-10.1ubuntu2 HEAD"), it runs from the shell; without it, git just spits out its usage18:58
smoserhm. it worked forme18:58
smoserin ipython18:58
smoserit brought up an editor18:58
naccsmoser: strange, i'll keep testing then, might be something elese18:58
smosernacc, i just tagged the usd repository with 'foo', by doing:18:59
smoserpython3 -c 'from usd import run; run.run(["git", "tag", "--annotate", "foo"], stdin=None, stdout=None, stderr=None)'18:59
naccsmoser: oh! i see what i'm doing wrong19:01
naccsmoser: thanks!19:01
smoserand, fwiw, both:19:04
smoser python3 -c 'from usd import run; run.run(["ls"], stdin=None, stdout=None, stderr=None, shell=True)'19:04
smoserand19:04
smoser python3 -c 'from usd import run; run.run("ls", stdin=None, stdout=None, stderr=None, shell=True)'19:04
smoserwork for me19:04
naccsmoser: yeah, my problem was not passing all of the None arguments19:04
naccsmoser: thanks for the pointers!19:04
jgrimmsmoser, stgraber, slangasek: any idea why '::1' line in /etc/hosts differs between cloud-images and ISO d-i installed?   the latter, includes 'localhost', the former does not.20:14
stgrabernot sure. it does make sense having localhost listed for ::1 though, otherwise "ping6 -n localhost" wouldn't work20:15
smoserjgrimm, i suspect that the process that writes that file is older than writing ipv6 localhost20:16
jgrimmstgraber, indeed20:16
jgrimmsmoser, agreed.. though my desktop is missing it too20:17
slangasekmy desktop is not missing it fwiw20:17
jgrimmslangasek, interesting20:18
slangasekbut my desktop was installed with d-i, not ubiquity, because lvm+crypt20:18
jgrimmah20:18
slangasekI'm trying to remember if this was a bug we fixed in ubiquity20:18
jgrimmmine was ubiquity installed, circa wily20:18
jgrimmslangasek, ryan found an old one claiming it was fixed20:18
slangasekso, I think the right answer is to include localhost for the ipv6 address, and we should fix the cloud image build scripts to match d-i here20:19
jgrimmslangasek, cool. that's was my gut feel, except to wonder if someone intentionally backed it out at some point??20:19
slangasekjgrimm: checking the current /var/lib/dpkg/info/netbase.postinst, which should be considered authoritative, we are including localhost for ::1.20:19
smoserwell, if that were the case then hopefully said process has version control on its /etc/hosts writing.20:19
smoserslangasek, is it 'localhost' ?20:20
slangasekthings that deviate from that probably do so out of ignorance rather than by design20:20
smoseri have ip6-localhost and ip6-loopback20:20
slangaseksmoser: in /var/lib/dpkg/info/netbase.postinst?20:20
smoserno, in /etc/hsots20:20
smoseryeah, i have no 'localhost' entry for ::120:21
jgrimmsmoser, then presumably you were a cloud-image, or ubiquity installation20:21
smoserubiquity, yeah.20:22
smoseri didnt realize we were speciically speaking of 'localhost'20:23
jgrimmno worries20:23
jgrimmslangasek, given responses i'll just open bugs against the offenders20:24
slangasekjgrimm: I suspect livecd-rootfs for both20:25
jgrimmslangasek, cool. thanks20:26
jgrimm164400920:41
jgrimmbug 164400920:41
ubottubug 1644009 in livecd-rootfs (Ubuntu) "/etc/hosts ::1 line should additionally include 'localhost'" [Undecided,New] https://launchpad.net/bugs/164400920:41
naccrbasak: jgrimm: well, in pleasant news, even with all the changes to the importer, continuing the php-imagick import to re-merge worked and the rebase went through straightforwardly. In that particular case, we may be able to drop parts of the delta (testing those cases now), but it's nice to see.21:00
jgrimmnacc, nice!21:00
jgrimmnacc, did you see my poke about usd merge error I was getting this morning?21:01
=== salem_ is now known as _salem
naccjgrimm: no, sorry!21:44
naccjgrimm: too many overnight pings, i expect21:44
jgrimm:)21:45
naccjgrimm: can you resend?21:45
jgrimmack21:46
jgrimmnacc: pretty simple, i was just trying to do a usd merge, with a fresh usd, and fresh clone21:46
jgrimmnacc, ERROR:Failed to create local branch (debian/sid). Does it already exist (pass -f)?21:47
naccjgrimm: ok, i just pushed those fixes before lunch21:47
jgrimmnacc, ok i'll retry, thanks21:47
naccjgrimm: can you try it again and let me know, and if it stilld oesn't work, send me the src pkg name and i'll reproduce21:47
jgrimmnacc, different behavior at least.  fyi, i did a fresh clone again too21:49
jgrimmnacc, ERROR:tag is not a defined object in this git repository.21:49
jgrimmnacc, the package is at21:49
naccjgrimm: ah, sorry, i just changed behavior! :)21:50
naccjgrimm: just run `usd merge ubuntu/devel`21:50
jgrimmnacc, ah... :)21:50
naccjgrimm: the subcommands were unnecessary -- in the general case you'll do both back to back21:50
naccjgrimm: that succeeded for me locally21:50
nacclet me update the wiki too21:51
jgrimmnacc, thanks sir21:51
jgrimmnacc, indeed verified that works21:52
naccjgrimm: note that if you are doing a re-merge21:52
naccjgrimm: meaning there is an upload tag earlier21:52
jgrimmits not a remerge21:52
naccjgrimm: then you might want `usd merge --tag-only`, and then actually do a git-rebase21:52
naccjgrimm: ok21:52
jgrimmgood to know tho!21:53
naccjgrimm: presuming no new delta added, that is; (meaning your last upload/ tag matches ubuntu/devel)21:54
naccjgrimm: i'll try and document these cases on the wiki21:55
jgrimmnacc, sounds good!  fyi <cmd> -h output on wiki page are stale too21:56
naccjgrimm: yep, i was trying to stabilize the tooling and then c&p again21:56
jgrimmthanks sir21:56
naccjgrimm: will update by EOD21:56
naccmdeslaur: i think your imagemagick upload clobbered the fixes that were in 8:6.8.9.9-7ubuntu4 (and thus it regresses php-imagick)22:16
naccmdeslaur: this is because (and i'm pretty sure i reported this to the debian maintainer who ignored it), the debian backports of certain upstream fixes are wrong/incomplete22:17
rbasakslangasek: that's odd. I think I'm missing it. Did you Cc the list too? https://lists.ubuntu.com/archives/technical-board/2016-November/thread.html doesn't have it either.22:17
rbasakjgrimm, nacc: that's odd. "ERROR:Failed to create local branch (debian/sid)" was what I complained about the other day too, but it since worked for me since before today.22:18
naccrbasak: it was the importer -> lpusip change22:19
LStrangercould anyone tell me why openbox-3.6.1-3 is still in proposed? gilir put it there a month ago and it's still there.22:19
rbasakAh22:19
naccrbasak: it should work now on a fresh clone and merge22:19
rbasakOK, thanks.22:19
naccLStranger: http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#openbox22:19
naccLStranger: it regressed plasma-framework's autopkgtests22:19
LStrangernacc: ah, I see, thank you very much22:20
LStrangerprobably I should poke debian maintainer to add the dependency in debian too22:20
LStrangeror may be just open a RC bug instead :D22:21
LStrangerif I get it right, the openbox-kde session would not work in current state unless kde session is installed22:23
tarpmanLStranger: are we looking at the same log? the failure linked there looks like bug 1590956 to me... doesn't look like openbox's fault22:25
ubottubug 1590956 in plasma-framework (Ubuntu) "plasma-framework 5.22.0a-0ubuntu1 fails tests on s390x" [Undecided,Confirmed] https://launchpad.net/bugs/159095622:25
mdeslaurnacc: hrm, I'll take a look tomorrow. Is there a bug for php-imagick being broken now?22:25
mdeslaurnacc: I have another imagemagick update coming out in a few days I believe, so I can fix it at the same time22:25
slangasekrbasak: oh wow, my local dns is broken and mail not going out, thanks for the prod :)22:26
naccmdeslaur: i think what's missing is that the deb8u5 did refresh one bug and pick up the patch, but the original backport that was refresh is still wrong (see my messages in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=811308)22:29
ubottuDebian bug 811308 in imagemagick "Multiple minor security issues" [Normal,Fixed]22:29
naccmdeslaur: that has the three upstream backports that are missing, and cause php-imagick to fail22:29
naccmdeslaur: well, imagemagick is stuck in proposed22:29
naccmdeslaur: so no bug, but i can file one :)22:29
LStrangertarpman: well, I thought it was about missing dependency on plasma-workspace there, probably I'm wrong22:33
naccmdeslaur: so the issue is ubuntu9's 0070-Fix-PixelColor-off-by-one-on-i386.patch and ubuntu10's are different. I'm guessing that is due to taking Debian's version22:34
naccmdeslaur: but our version had add-on fixes, i apologize for my poor changelog not indicating that more clearly (i've since learned better!)22:34
naccmdeslaur: would it be more productive to just merge imagemagick?22:35
mdeslaurnacc: I have to see if the cves are fixed in unstable, but sure we should be merging it22:42
mdeslaurnacc: sorry for overwriting your fix, I'll make sure to correct it in my next updates to the stable release22:42
naccmdeslaur: sorry, just to make sure i'm not crossing wires -- i'm only seeing the failure to migrate in z-p22:43
mdeslauroh! Does it just affect the autopkgtest?22:44
naccmdeslaur: we'd be able to drop my backports, at least, if we merge, as they are fixes from upstream that were mis-cherry-picked back. I'm happy to take a look at the merge, if you want22:44
naccmdeslaur: yeah, i think so -- i mean it's a functional bug in the package, that only php-imagick provokes :)22:44
mdeslaurnacc: sure, could you handle the merge?22:44
mdeslaurnacc: I can check tomorrow if all the cves are fixed in unstable22:45
naccmdeslaur: yep, i'll add it to my list, most of the updates to the package were by me :)22:46
mdeslaurcool, thanks!22:46
naccmdeslaur: thank you!22:47
slangasekpitti: so, systemd-resolved apparently stopped resolving ipv6 names for me, and I'm not sure how to debug22:47
juliankHmm, should I pass -v1.2.15 when building apt 1.2.17 which replaces the 1.2.16 in the xenial-proposed unapproved queue (1.2.15 is the one in -updates), or does that not matter at all?22:48
juliankCurrently it also has a weird test failure, not sure what's going on there...22:49
LStrangeranyway, I've filed an RC bug against Openbox in Debian, should it help to get openbox from proposed into zesty?22:52
LStrangerwhy I'm interested - my friends need #1336521 to be fixed in LTS releases.22:54
Unit193(Debian #845386)22:54
ubottuDebian bug 845386 in openbox-kde-session "openbox-kde-session package is unusable" [Grave,Open] http://bugs.debian.org/84538622:54
LStrangerfor myself I've created a patched package but I want others to have fix too. :)22:57
LStrangertarpman: so far, if that isn't openbox fault, what can be done to let openbox 3.6.1-3 enter zesty? this another bug will be fixed later then.23:04
juliankHmm, the arm64 build failure for trusty bash holding back bug 1422795  is not a regression of that upload (the failing file was not changed at all) Should I still generate a  regression-proposed bug report for that?23:04
ubottubug 1422795 in bash (Ubuntu Trusty) "bash crashes often if inputrc contains revert-all-at-newline" [High,Fix committed] https://launchpad.net/bugs/142279523:04
juliankThe fix is basically ready23:05
tarpmanLStranger: you have to find someone who has the ability to bypass that autopkgtest regression... I don't know who that would be. someone here, I'm sure23:05
juliankProposed bash fix: http://paste.ubuntu.com/23519397/23:06
LStrangerwell, I'm asking such someone then -- do it, please, bug 1336521 is pretty annoying for people23:06
ubottubug 1336521 in pcmanfm (Ubuntu) "Application Startup Notify fully ignored" [Undecided,Confirmed] https://launchpad.net/bugs/133652123:06
Unit193LStranger: They tend to want you to attach debdiffs, and sub sponsors.23:07
LStrangerUnit193: the matter is in migration Debian updated package into Ubuntu23:08
juliankHmm, it's even stranger.23:10
juliankThe same line causing the error on arm64 in trusty is present in zesty as well, but it works there.23:11
LStrangerUnit193: the difference in 3.6.1-1ubuntu2 was kde-workspace was replaced with plasma-desktop, which is also wrong but in 3.6.1-3 that kde-workspace-bin dependency is removed so no difference except missing dependency, which I filed as a grave bug so it will be fixed soon23:11
juliankdoko: You might want to look into bug 1644048 - there's a printf(ngettext(...)) line in bash's builtins/help.def that causes a FTBFS on arm64 in trusty, but not on other architectures, and not in newer releases.23:12
ubottubug 1644048 in bash (Ubuntu Trusty) "4.3-7ubuntu1.6 FTBFS on arm64" [Undecided,New] https://launchpad.net/bugs/164404823:12
LStrangerso far it fits into zesty so I hope someone would push a button here23:12
LStrangerwell, time to sleep for me :)23:14

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