/srv/irclogs.ubuntu.com/2017/11/20/#ubuntu-release.txt

=== maclin1 is now known as maclin
LocutusOfBorguim is fixed, rebuilds issued06:59
LocutusOfBorgthanks doko for accepting them :)06:59
LocutusOfBorgcan we please have freecad ignored on ppc64el and s390x? it is regressed in release since long time autopkgtest for freecad/0.16.6712+dfsg1-1build1: amd64: Pass, armhf: Always failed, i386: Pass, ppc64el: Regression ♻ , s390x: Regression ♻07:01
LocutusOfBorgapw, good morning, I syncd python-pathspec (from Debian). The ubuntu package was providing python-pathspec and python3-pathspec, the Debian one just provides the python3 binding. Do we need some NBS cleanup?08:33
apwLocutusOfBorg, perhaps indeed08:43
LocutusOfBorgthanks for having a look :)08:49
LocutusOfBorghurray! yamllint finally fixed :)08:52
* LocutusOfBorg looks hard at the transition :)08:52
-queuebot:#ubuntu-release- New binary: foolscap [amd64] (bionic-proposed/universe) [0.13.0-0ubuntu1] (no packageset)10:06
-queuebot:#ubuntu-release- New: accepted foolscap [amd64] (bionic-proposed) [0.13.0-0ubuntu1]10:07
Laneyapw: slangasek: infinity: FYI autopkgtest-cloud-worker/3 is the new controller10:10
apwLaney, ack10:11
Laneyif you want to rescue something off /0, do it today please because I'm going to delete it tomorrow10:11
apwLaney, have nothing on it that i know of10:11
Laneyapw-banking-passwords.txt10:12
xnoxlol10:12
* Laney should extract some things from bash history10:12
ginggsapw would you bump the freecad hints for ppc64el and s390x to 0.16.6712+dfsg1-1build1 please?10:17
ginggsalso, would you please consider ' force-badtest python-scipy/0.19.1-1ubuntu1/i386' it looks like precisions errors, but the tests pass for me in an i386 VM running bionic-proposed so i cannot test a patch10:23
xnoxinfinity, was https://launchpad.net/ubuntu/+source/diaspora-installer/+publishinghistory demoted to proposed incorrectly, managed to migrate this morning, blocking up postgresql/icu transition now. Could it be RM from the archive again?10:29
LocutusOfBorginfinity, https://launchpad.net/~costamagnagianfranco/+archive/ubuntu/locutusofborg-ppa/+sourcepub/8503388/+listing-archive-extra if you want to take it (I'm interested in maven and qt5 fixes)10:32
LocutusOfBorgsigh libxml2...10:33
LocutusOfBorgCVE fixes out there but a lot of tests10:33
apwginggs, hints kicked down the road some more, not had a achnage to look ath python-scipy10:39
ginggsapw: thanks10:39
dokoapw: can pglogical be made expected to fail on armhf too? it's currently the only architecture where tests are not ignored10:41
LocutusOfBorgapw, as said, i386 is not a thing, and tests are just too flaky there, we never caught a regression on i386, but always patched tests to make it pass (by relaxing the bounds)10:44
LocutusOfBorgso, ignoring here is really something nice to do, and probably the new release is fine again, even if we don't want it right now (because it triggers a ton of stuff)10:45
LocutusOfBorghaving them candidate will make more readable the britney output10:45
LaneyLocutusOfBorg: If you're not going to use the results, why bother running the tests?10:50
LocutusOfBorgLaney, because we don't want to diverge more from Debian10:52
LocutusOfBorgand yes, we use them, and we decide if they are a bug in testsuite or a bug in code10:52
LocutusOfBorgI'm asking to ignore this specific version, not it forever10:52
dokoxnox: diaspora-installer removed again10:53
xnoxdoko, thanks.10:53
-queuebot:#ubuntu-release- Unapproved: accepted libreoffice-l10n [source] (artful-proposed) [1:5.4.2-0ubuntu0.17.10.1]10:53
-queuebot:#ubuntu-release- Unapproved: accepted libreoffice [source] (artful-proposed) [1:5.4.2-0ubuntu0.17.10.1]10:53
LaneyLocutusOfBorg: That last statement doesn't sound the same as "i386 is not a thing"10:54
LocutusOfBorgI mean, what is the purpose of fixing a testsuite that fails from time to time, only on i386 by relaxing test bounds10:57
LaneyI know what you mean, so why run it on i386 at all?10:57
LocutusOfBorgask Debian, I hate deltas :)10:58
Laneyplease do10:58
LocutusOfBorgI don't even remember how to ignore it in one specific architecture lol10:58
LaneyI hate multiple people having to spend their scarce time unproductively10:59
LocutusOfBorgLaney, https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=75802211:01
ubot5Debian bug 758022 in python-scipy "test_interpolate.TestPPoly.test_roots test fails on i386" [Important,Open]11:01
LocutusOfBorgseems the bug is there since only 3 years11:01
xnoxLocutusOfBorg, still fresh =)11:01
LocutusOfBorgfor debian ages, it is like "yesterday" :)11:02
* LocutusOfBorg tries to disable testsuite on i38611:02
dokosil2100: "Grouped with PPA ~ci-train-ppa-service/+archive/ubuntu/3033" does this have any effect on migrations?11:06
apwdoko, yes, it means that all of the things from that place need to get ready together before they migrate11:21
dokoapw: does this include exact versions?11:22
apwdoko, now i don't think i know the answer to that ... it could easily be floppy11:23
dokoLaney: ^^^11:26
-queuebot:#ubuntu-release- New: rejected calamares-settings-ubuntu [source] (bionic-proposed) [1]11:26
dokojust asking because some of these packages were updated again11:26
dokotsimonq2: ^^^ rejected, missing copyright holder11:27
dokotsimonq2: also some of the files are at least GPLv3, not v211:27
Laneydoko: I'm not sure, that was a robru thing11:41
Laneyhttps://git.launchpad.net/~ubuntu-release/britney/+git/britney2-ubuntu/tree/britney2/policies/sourceppa.py11:42
LaneyI think that probably you can search for the grouped PPA and any packages listed need to be candidates at the same time11:43
Laneyin excuses11:43
dokooutput?11:43
Laneywhat?11:44
dokoupdate_output?11:44
Laneywhat about it?11:44
dokolibreoffice/1:5.4.1-0ubuntu311:45
dokobahh, pasting not working11:45
dokowhat you said above: "in excused"  ?11:45
Laneythat's where you see those Grouped by xxxxxxx things11:46
sil2100doko: I don't have much knowledge about this, never used it or worked on it11:46
slashdo/ sil2100 good Monday, could you please release cups in artful-updates for (LP: #1729910) ?12:09
ubot5Launchpad bug 1729910 in cups (Ubuntu Artful) "lp ignores ~/.cups/lpoptions in 17.10" [Medium,Fix committed] https://launchpad.net/bugs/172991012:09
slashddgadomski, ^12:09
sil2100slashd: hey! Sure, let me take a look at it once I'm done with kernels12:11
slashdsil2100, ok thanks ;)12:11
LocutusOfBorghow to disable testsuite on a specific architecture? Laney ^12:13
LaneyLocutusOfBorg: I don't know of a better way than looking at the architecture using dpkg-architecture or something12:15
cjwatsonI'd use dpkg --print-architecture to avoid the dpkg-dev dependency, but yes12:16
acheronukdpkg-architecture -e i386 && exit 012:16
acheronukI think?12:17
cjwatsonagain, that depends on dpkg-dev which isn't necessarily sensible in a testsuite12:18
LaneySure12:18
cjwatson[ "$(dpkg --print-architecture)" = i386 ]12:18
acheronukpoint taken12:18
-queuebot:#ubuntu-release- New: rejected hibagent [source] (bionic-proposed) [1.0.1-0ubuntu1]12:21
doko<ginggs> doko: hint just needs updating? https://bazaar.launchpad.net/~ubuntu-release/britney/hints-ubuntu/revision/265212:57
dokoLaney: any reason about the limited set of versions?12:57
LocutusOfBorgso, you mean inside the test, not in the test/control file13:12
LocutusOfBorgginggs, maybe you want to try the patch? I'm not quite confortable in blindly uploading it and triggering a a lot of autopkgtests13:18
ginggsLocutusOfBorg: i thought blindly uploading was your thing :p  what patch do you want me to try?13:22
acheronukcan someone please ignore test fails for libkf5eventviews on ppc64el and s390x? binaries to test have not been buildable on those architectures since May (QtWebEngine), so it makes no sense to suddenly have the tests try to run now!13:26
acheronuk^^^ same for akonadi-calendar13:28
acheronukthose 2 are spuriously blocking abi-compliance-checker13:28
Laneydoko: Probably because we expected that kernel regression to not be around for many months13:31
LaneyFeel free to change it13:31
Laneyacheronuk: Someone manually requested those tests13:31
Laneyand then people have been blindly retrying them for some reason13:31
dokoLaney: I can't13:31
Laneyok13:32
acheronukLaney: I think I may have retried one or two in error before I realised, but didn't make the initial wrong request as far as I can see. I recall a week ago meaning to mention they should not be rune, but with one thing and another I forgot to bring it up13:39
LocutusOfBorg[11:23:58] <ginggs> also, would you please consider ' force-badtest python-scipy/0.19.1-1ubuntu1/i386' it looks like precisions errors, but the tests pass for me in an i386 VM running bionic-proposed so i cannot test a patch13:43
LocutusOfBorgthe one mentioned here :)13:43
LocutusOfBorgbtw you can also run testsuite from ppa if you have one with the package published13:43
Laneyacheronuk: It doesn't matter about the delay - that someone requested it in the first place, and then people kept on retrying afterwards without reading the log ... yeah, to me that's a bit of a problem13:47
Laneylaney@snakefruit:/home/ubuntu-archive/proposed-migration/log/bionic$ zgrep -e "Requesting.*akonadi-calendar.*(s390x|ppc64el)" 2017-11-*/*.gz13:47
Laneylaney@snakefruit:/home/ubuntu-archive/proposed-migration/log/bionic$13:47
Laneyanyways, added the hint for the current versions13:52
dokoLaney: please could you ignore the python-scipy/i386 fail triggered by python-numpy? I'd like to avoid a new upload and triggering autopkg tests again13:54
LaneyI'm doing something else now, sorry, maybe someone else can help you (ap_w mentioned that one earlier I think)13:54
dokoapw: ^^^13:56
bdrungdoko, as you probably saw, i fixed the build failure of rdma-core on armhf14:02
dokobdrung: you probably saw that it needs promotion14:03
bdrungyes14:03
bdrunganything that should be added from by side?14:04
ginggsdoko: dropping freecad ppc64el and s390x binaries stills requires hinting :(14:12
ginggssame happens when debian stops running autopkgtests14:12
ginggsif they've passed once, they have to pass forever (or get hinted)14:13
-queuebot:#ubuntu-release- New source: hibagent (bionic-proposed/primary) [1.0.1-0ubuntu1]15:03
slashdthanks for cups sil210015:11
apwdoko, looking at python-scipy/i38615:16
doko<ginggs> doko: i *think* this patch to scipy will do it: https://paste.ubuntu.com/25998603/ - i can't test it because i can't reproduce the failure in my VM15:18
doko$ fgrep pglogical *15:18
dokoadconrad:force-badtest pglogical/2.0.0-115:18
dokoadd that one should be updated to 2.1.0-115:19
apwpython-scipy looks reasonable, applid15:20
xnoxdoko, yes please about pglogical.15:20
apwdoko, pglogical done ...15:22
dokoapw: and please remove the freecad ppc64el and s390x test results. the package is not built anymore on these archs15:23
apwtrying to work out if i know how to do that :)15:25
dokowell, or add an ignore for now15:25
xnoxtesting a fix for pgagent/ppc64el15:30
xnoxand after that we need to kick out postgresql-mysql-fdw/2.3.0-1 and all should be well.15:30
dokoxnox: is there a bug report for that?15:30
xnoxdoko, https://github.com/EnterpriseDB/mysql_fdw/issues/15015:31
xnoxthey added opportunistic support for 10.0, but it's not working with 10.115:31
* xnox opens a bug15:32
xnoxapw, could you please remove postgresql-mysql-fdw, but keep the one in bionic-proposed around?15:34
xnoxor like remove all of it all together? or hint a badtest?15:34
xnoxhttps://bugs.launchpad.net/ubuntu/+source/postgresql-mysql-fdw/+bug/173334115:34
ubot5Launchpad bug 1733341 in postgresql-mysql-fdw (Ubuntu) "fails to work with postgresql 10.1" [Undecided,New]15:34
dokoI'll remove it15:34
apwdoko, it seems they are only being tested because we had packages on those arches in -release at the time of the test15:39
apwdoko, we can ignore those two now, and it should be fine next time round15:39
dokothat should be sufficient for the current transitions, yes15:40
apwand hinted ... lets see if that does the do15:41
dokoand now waiting for the just uploaded lo build :-/15:42
Laneyis this the big migration?15:42
dokoyes15:43
dokohttps://bugs.launchpad.net/ubuntu/+bug/173274215:43
ubot5Launchpad bug 1732742 in Ubuntu "entangled transitions for bionic" [Undecided,New]15:43
Laneynice work15:43
dokowell, then at least we'll see what we are missing ...15:44
dokoI fear there is more qt stuff15:45
apwdoko, that is a beast, feels like half the archive15:45
-queuebot:#ubuntu-release- Unapproved: accepted util-linux [source] (artful-proposed) [2.30.1-0ubuntu4.1]15:46
LaneyThe following packages have unmet dependencies. libsaml9 : Depends: libxerces-c3.1 but it is not installable libxmltooling7 : Depends: libxerces-c3.1 but it is not installable15:49
LaneyE: Unable to correct problems, you have held broken packages.15:49
* Laney tries rebuilding those15:49
-queuebot:#ubuntu-release- New: accepted hibagent [source] (bionic-proposed) [1.0.1-0ubuntu1]15:50
apwis that "all" of the remaining brokeness ?15:50
LaneyI'm not sure15:50
LaneyI just looked at some random stuff on update_output_notest15:50
Laneymost of it is liberoffice15:50
* xnox has one more package to publish for postgresql-common - the pgagent fix, refreshing silo ppa to publish to then copy those into the archive15:51
xnoxand publishing15:51
-queuebot:#ubuntu-release- New binary: hibagent [amd64] (bionic-proposed/none) [1.0.1-0ubuntu1] (no packageset)15:53
-queuebot:#ubuntu-release- New: accepted hibagent [amd64] (bionic-proposed) [1.0.1-0ubuntu1]15:56
-queuebot:#ubuntu-release- New source: hibagent (trusty-proposed/primary) [1.0.1-0ubuntu1~14.04.1]16:56
-queuebot:#ubuntu-release- New source: hibagent (xenial-proposed/primary) [1.0.1-0ubuntu1~16.04.1]16:57
-queuebot:#ubuntu-release- New: accepted hibagent [source] (xenial-proposed) [1.0.1-0ubuntu1~16.04.1]17:07
-queuebot:#ubuntu-release- New binary: hibagent [amd64] (xenial-proposed/none) [1.0.1-0ubuntu1~16.04.1] (no packageset)17:10
xnoxLaney, apw - even thought postgresql-mysql-fdw got kicked out of bionic-release, it appears to still be londing up postgresql-common, can you badtest postgresql-mysql-fdw? it will need source fixes to get it working with postgresql-10 again.17:18
xnoxautopkgtest for postgresql-mysql-fdw/2.3.0-1: amd64: Regression ♻ , armhf: Regression ♻ , i386: Regression ♻ , ppc64el: Regression ♻ , s390x: Regression ♻17:18
xnoxor maybe remove it from bionic-proposed too.17:18
xnox..17:18
xnoxretried pgagent with postgresql-common as trigger, should pass with the new pgagent.17:19
apwso we just want to ignore its tests for the moment, and block it in proposed so it doesn't migrate; then once teh rest does retun its tests and unblock it ?17:20
xnoxapw, correct.17:20
xnoxhttps://bugs.launchpad.net/ubuntu/+source/postgresql-mysql-fdw/+bug/173334117:20
ubot5Launchpad bug 1733341 in postgresql-mysql-fdw (Ubuntu) "fails to work with postgresql 10.1" [Undecided,Confirmed]17:21
apwxnox, except nothing is depending on it ..17:32
xnoxapw, i see postgresql-common triggering tests of postgresql-mysql-fdw, still, even after it was removed from -release. not sure if update_excuses is out of date, or buggy.17:38
xnoxoh it's gone17:39
apwits not there now, i suspect since it has been removed in -release finally caught up17:39
xnoxall is good =) i must have been looking at stale update_excuses.17:39
xnoxpgagent is now green with postgresql-common; so maybe the next run will show us what's left for migration of doom.17:40
xnoxboo, my retries did not make into this run either. hopefully next one then.17:54
apwxnox, there still seems to be pyqt junk on there17:59
-queuebot:#ubuntu-release- Unapproved: django-tastypie (xenial-proposed/universe) [0.12.0-1build1 => 0.13.3-1~16.04.1] (no packageset)18:12
dokoLaney: is update-output-helper currently working for you?18:21
-queuebot:#ubuntu-release- Unapproved: ipxe (xenial-proposed/main) [1.0.0+git-20150424.a25a16d-1ubuntu1 => 1.0.0+git-20150424.a25a16d-1ubuntu1.2] (ubuntu-desktop, ubuntu-server)18:27
-queuebot:#ubuntu-release- Unapproved: nova (xenial-proposed/main) [2:13.1.4-0ubuntu4.1 => 2:13.1.4-0ubuntu4.2] (openstack, ubuntu-server)18:28
-queuebot:#ubuntu-release- Unapproved: ipxe (zesty-proposed/main) [1.0.0+git-20150424.a25a16d-1ubuntu2 => 1.0.0+git-20150424.a25a16d-1ubuntu2.1] (ubuntu-desktop, ubuntu-server)18:28
=== kmously is now known as kmously-afk1hr
tsimonq2infinity: Any chance you're around?19:18
infinitytsimonq2: There's a chance.19:25
-queuebot:#ubuntu-release- New: accepted hibagent [amd64] (xenial-proposed) [1.0.1-0ubuntu1~16.04.1]19:26
tsimonq2infinity: Have you seen my MP to lp:ubuntu-cdimage?19:27
tsimonq2infinity: Lubuntu made a QA mistake and I'm wondering if it's at all possible to get new images for 17.1019:28
dokolooks like I hadn't finished the xerces-c transition ...19:28
tsimonq2infinity: We missed LVM packages and UEFI packages because the share seed wasn't included.19:30
LocutusOfBorgdoko, what is missing?19:31
tsimonq2infinity: (I'd like to know how to fix it)19:31
infinitytsimonq2: Yeah, I see that.  I'm not really a big fan of respinning a release a month after a release.19:31
tsimonq2infinity: Right, and neither am I. But I'm also getting annoyed at people poking me about it. Ofc they had to wait until after the release, but that's besides the point.19:34
tsimonq2The fix is simple but tedious.19:34
-queuebot:#ubuntu-release- New: accepted hibagent [source] (trusty-proposed) [1.0.1-0ubuntu1~14.04.1]19:35
infinitytsimonq2: Well, one thing at a time, I'll merge this and you can see if bionic looks how you think it should.19:36
tsimonq2infinity: Sure. Is there a delay between ubuntu-cdimage revisions being merged and actually being deployed?19:37
-queuebot:#ubuntu-release- New binary: hibagent [i386] (trusty-proposed/none) [1.0.1-0ubuntu1~14.04.1] (no packageset)19:37
slangasekonly if the person merging doesn't follow up with deploying19:38
Laneydoko: yes, but I might have some local modifications19:38
dokoLocutusOfBorg: uploads done.19:38
Laneywhat's wrong with it for you?19:38
dokothat seems to be open: http://autopkgtest.ubuntu.com/packages/r/ruby-em-http-request/bionic/amd6419:38
dokoLaney: let me check19:38
dokois this a proxy problem?19:39
tsimonq2slangasek: ack19:39
infinitytsimonq2: What Steve says.  The delay is anywhere between seconds and "until the end of time", as it relies on a manual pull on nusakan.  But most of us remember that step. :P19:39
dokoLaney: http://paste.ubuntu.com/26007342/19:40
LocutusOfBorgthanks doko, do you need some more work? I fail to see what is missing for the big one19:41
infinitytsimonq2: Done and done.19:41
tsimonq2infinity: And rebuilt. Thanks.19:47
dokoLocutusOfBorg: see above, the ruby autopkg test failure19:49
Laneydoko: here's my version http://paste.debian.net/996753/19:52
Laneypoor connection atm, sorry19:53
dokoLaney: yes that one works :)19:54
dokooSoMoN: LO build failure on amd64 :-/19:57
dokosegfault in a unit test19:58
dokooSoMoN: I pasted the tail of the log, and gave it back. http://paste.ubuntu.com/26007473/20:05
SchykenHeyo20:22
slangasekxnox: yes, please stop using nested KVM with the next systemd upload, given its not-support status.20:26
=== kmously-afk1hr is now known as kmously
oSoMoNdoko, mmm, that's weird, thanks for retrying, let's see how it goes this time20:56
=== santa is now known as Guest73570
=== Guest73570 is now known as santa__
-queuebot:#ubuntu-release- New: accepted hibagent [i386] (trusty-proposed) [1.0.1-0ubuntu1~14.04.1]22:50
-queuebot:#ubuntu-release- Unapproved: fwupdate-signed (artful-proposed/main) [1.14ubuntu0.2 => 1.14.1] (core)23:10
-queuebot:#ubuntu-release- Unapproved: fwupdate-signed (zesty-proposed/main) [1.13ubuntu0.1 => 1.13.1] (core)23:10
tsimonq2infinity: Testers have confirmed that the tooling change fixed the problem.23:42
tsimonq2(in Bionic)23:42
tsimonq2infinity: Would you be able to spin up some 17.10 ISOs for us to test the changes?23:43
xnoxslangasek, ok, i will force not using nested kvm, but the defaults should be changed too, nonetheless. Currently it checks of the presence of /dev/kvm but makes use of it, if found, but it doesn't itself load kvm modules and/or specifies kvm.nesting=1 at all.23:43
slangasekxnox: ah, of course; makes sense, and yes I agree the default should be changed, I think cpaelzer is planning to follow up on that23:44
xnoxslangasek, i will add an etra check to systemd-virt-detect --vm and thus ignore /dev/kvm presense (* unless z/vm, cause althought that's nested virtualisation, it's not linux-kvm-nesting)23:44
xnoxcpaelzer, ^^^23:44
tsimonq2infinity: Also, irt naming, would they be called 17.10.1 or would they overwrite the existing ISO?23:45

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