15:01 <vorlon> #startmeeting 15:01 <meetingology> Meeting started Thu Apr 4 15:01:42 2019 UTC. The chair is vorlon. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 15:01 <meetingology> 15:01 <meetingology> Available commands: action commands idea info link nick 15:01 <vorlon> [TOPIC] Lightning round 15:01 <vorlon> $ echo $(shuf -e vorlon bdmurray xnox tdaitx doko sil2100 rbalint infinity cyphermox mwhudson juliank waveform) 15:02 <vorlon> rbalint bdmurray infinity tdaitx vorlon xnox waveform cyphermox doko mwhudson sil2100 juliank 15:02 <xnox> rbalint, did you nominate vorlon last week? 15:02 <juliank> may I move up? 15:02 <bdmurray> juliank: I'll trade you 15:03 <juliank> bdmurray: thanks 15:03 <vorlon> rbalint: 15:04 <rbalint> a moment 15:04 <vorlon> juliank: you can go first then ;) 15:04 <juliank> * opened merge request for ua apt integration, fixed up comments, and got it merged 15:04 <juliank> * poked RT to allow aptdaemon in :) 15:04 <juliank> * ran my version regression script 15:04 <juliank> * uploaded update-notifier motd stuff to disco 15:04 <juliank> * committed (regression) test case for apt/trusty regression to apt 1.8.y branch 15:04 <juliank> * answering APT questions on IRC and email 15:04 <juliank> * filed another gnome-shell bug and forwarded the last one to upstream 15:04 <juliank> * fixed up the update-notifier upload to add missing build-dependency 15:04 <juliank> * SRUed update-notifier ESM stuff to trusty 15:04 <juliank> * Reviewing and merging apt merge requests 15:04 <juliank> (done) 15:05 <rbalint> (short week) 15:05 <rbalint> * checked openjdk11 sru transition for changed conffiles and if u-u breaks 15:05 <rbalint> * made lintian autpkgtes parallel 15:05 <rbalint> * gardened branches in lp:~ubuntu-core-dev/ubuntu/+source/systemd , please use it for preparing new uploads 15:05 <rbalint> * joined a few discussions 15:05 <rbalint> * uploaded fix for LP: #1823125 15:05 <ubottu> Launchpad bug 1823125 in tomcat9 (Ubuntu Bionic) "tomcat9 fails to install in 18.04" [Undecided,In progress] https://launchpad.net/bugs/1823125 15:05 <rbalint> * added ubuntu-wsl to ubuntu-meta, seed change reviews are pending 15:06 <infinity> rbalint: Can that lintian change be forwarded to Debian. I'd hate to carry a delta just for that. 15:06 <bdmurray> juliank: Is your version regression script anywhere other people can use it? 15:06 <rbalint> infinity, it is done and merged 15:06 <infinity> rbalint: Ahh, excellent, it wasn't when I looked. 15:06 <juliank> bdmurray: https://paste.ubuntu.com/p/W2bd6pJ334/ 15:06 <bdmurray> juliank: Does update-notifier need reviewing for trusty? 15:06 <juliank> bdmurray: sil2100 approved it 15:06 <bdmurray> okay 15:07 * sil2100 was fast in the morning 15:07 <sil2100> But now I'm slow again 15:07 <infinity> rbalint: Also, were you (done)? 15:07 * juliank heads out, started too early today and still recovering from hearing loss on Monday, so going for a walk 15:08 <rbalint> (done) 15:08 <infinity> * last-minute merges (dpkg, lsb, etc) 15:08 <infinity> * help investigating Qt/KDE s390x issues 15:08 <infinity> * prepping (hopfully) final glibc upload for disco 15:08 <infinity> * SO MANY QUEUE REVIEWS 15:08 <infinity> (done) 15:08 <infinity> PS: Won't be here for squad standup this afternoon; I have a doctor's appt. 15:08 <vorlon> tdaitx: 15:08 <tdaitx> * openjdk-11 bionic security transition 15:08 <tdaitx> - reviewing update excuses adt failures 15:08 <tdaitx> - investigating conffiles modification based on rbalint's u-u bb to bb-proposed work 15:08 <tdaitx> - looked over libreoffice issues/fixes 15:08 <tdaitx> * looking at pending merges 15:08 <tdaitx> (done) 15:08 <rbalint> infinity, the -j flag was dropped in lintian instead of my change 15:08 <vorlon> * short week, was off all last week 15:08 <vorlon> * end of cycle reviews (NEW, unapproved, etc) 15:08 <vorlon> * SRU reviews (openssl into bionic-proposed) 15:08 <vorlon> * proposed-migration: shepherding the various stuck things 15:08 <vorlon> * code reviews for automating nvidia driver support in GKE 15:08 <vorlon> * code review of tooling to manage all the different signed kernels now https://code.launchpad.net/~apw/ubuntu-archive-tools/copy-proposed-kernels-routing-updates/+merge/365420 15:08 <infinity> rbalint: Yeah, it should auto-j without it, as I read the code. Works for me. 15:08 <vorlon> * discussions around final plan for OpenJDK11 SRU release 15:08 <vorlon> (done) 15:09 <vorlon> oh, also discussing removing the rails packages 15:09 <vorlon> (double-done) 15:09 <vorlon> xnox: 15:09 <xnox> * subiquity, reviews/merges/releases/testing/ci-debug 15:09 <xnox> * openssl-sru revisions & fixes / debugging wget-ssl 15:09 <xnox> - openssl 1.1.1 is in bionic-proposed now \o/ 15:09 <xnox> * done 15:09 <rbalint> vorlon, (well done) :-) 15:09 <waveform> * tested armhf+arm64 disco images on variety of pi platforms (3B, 3B+, 3A+, CM3, CM3+), need to test 2B images this week 15:09 <waveform> * updated linux-firmware-raspi2 package to latest firmware release (in PPA; will need SRU for CM3+ boot on bionic) 15:09 <waveform> * SRU verification for netplan.io, #1819014 15:09 <waveform> * worked on python3.7 fix on armhf, #1821869 15:09 <waveform> * resolved reviews of pi "amazing" spec 15:09 <waveform> * finished pi first boot spec, currently in review 15:09 <waveform> * investigating zfs on pi issue, #1822797 15:09 <waveform> * verifying pi-bluetooth configuration on classic 15:09 <waveform> * learning console-conf ... for my sins 15:09 <waveform> (done) 15:10 <cyphermox> - netplan SRU verifications 15:10 <cyphermox> - grub2 SRU verifications 15:10 <cyphermox> - fixed missing grub-check-sig run in trusty's grub2 SRU (grub2-signed/1.34.20) 15:10 <cyphermox> - investigating how I can setup an openstack lab for more better netplan testing 15:10 <cyphermox> - revised https://code.launchpad.net/~cyphermox/debian-cd/nomodeset/+merge/364344 15:10 <cyphermox> - ubiquity slideshow update 15:10 <cyphermox> - looking at ubiquity bugs 15:10 <cyphermox> - sponsored livecd-rootfs/2.575 for r.c.j; fix duplicate snap listings in manifest 15:10 <cyphermox> - reran cross-toolchain-base-mipsen tests a few times, still tracking down i386 failures 15:10 <cyphermox> (medium rare done?) 15:10 <doko> tdaitx: lo is now addressed 15:10 <doko> - currently at Linaro Connect 15:10 <doko> - GCC updates for bionic and cosmic 15:10 <doko> - openjdk 11 11.0.3+5 15:10 <doko> - pestered^W helped waveform to setup arm* chroots 15:10 <doko> - python3.x uploads to address bus error (fixed by waveform) 15:10 <doko> (done) 15:11 <waveform> :) 15:11 <sil2100> - Sick most last week and on Monday-ish + holidays last week, doctor appointments 15:11 <sil2100> - SRU reviews and releases 15:11 <cyphermox> waveform: thanks for the sru verif btw :) 15:11 <sil2100> - Releasing of many kernels, new kernel cycle starting 15:11 <sil2100> - core18: 15:11 <sil2100> * Preparing the PR for date-based version numbers 15:11 <sil2100> * Debugging the date-based version change, poking with m_vo + finally getting help from Sergio - now done! 15:11 <sil2100> * Lots of PR reviews, corrections and merges 15:11 <sil2100> * Experimenting with snap prepare-image --classic 15:11 <sil2100> * Some progress on the core18 promoter scripting 15:11 <sil2100> - Review and merge of the more-strict-parser PR to ubuntu-image 15:11 <sil2100> - Reviewing the Raspi Pi specs (times two!) 15:11 <sil2100> - Some pi discussions here and there 15:11 <sil2100> - Fixing the core18 auto-snap-builder to properly notice packages in the additional PPAs 15:11 <sil2100> - Helping out with the systemd SRU autopkgtest regressions to make the package releasable today 15:11 <sil2100> - FFe and UIFe reviews 15:11 <sil2100> (good done) 15:12 <tdaitx> doko: yeah, when I was looking at the issue I noticed you had uploaded the fix for it 15:12 <bdmurray> searched for duplicates of LP: #1811694 - found a couple 15:12 <bdmurray> investigation into apport crash LP: #1820132 - missing executable 15:12 <bdmurray> wrote an apport test case for add_gdb_info about a missing executable 15:12 <bdmurray> uploaded apport to disco with fixes for LP: #1809247, LP: #1396160 and the above bug 15:12 <ubottu> Launchpad bug 1811694 in aptdaemon (Ubuntu Disco) "/usr/sbin/aptd:RuntimeError:_inline_callbacks:get_uid_from_dbus_name:return_value:_inline_callbacks:_check_simulated:_inline_callbacks" [High,Fix released] https://launchpad.net/bugs/1811694 15:12 <ubottu> Launchpad bug 1820132 in apport (Ubuntu Disco) "/usr/share/apport/whoopsie-upload-all:AssertionError:/usr/share/apport/whoopsie-upload-all@162:collect_info:process_report:add_gdb_info:gdb_command" [High,In progress] https://launchpad.net/bugs/1820132 15:12 <ubottu> Launchpad bug 1809247 in apport (Ubuntu Disco) "SetReportCrashes to true dbus call failed" [High,In progress] https://launchpad.net/bugs/1809247 15:12 <ubottu> Launchpad bug 1396160 in apport (Ubuntu Disco) "/usr/share/apport/recoverable_problem:PermissionError:/usr/share/apport/recoverable_problem@75:main:add_proc_info" [Medium,In progress] https://launchpad.net/bugs/1396160 15:12 <bdmurray> also modified apport's relationship with policykit in part b/c of LP: #1821415 15:12 <ubottu> Launchpad bug 1821415 in policykit-1 (Ubuntu) "pkexec fails in a non-graphical environment" [Undecided,New] https://launchpad.net/bugs/1821415 15:12 <bdmurray> research into LP: #1822886, testing of possible upgrade failure scenarios 15:12 <ubottu> Launchpad bug 1822886 in ubuntu-release-upgrader (Ubuntu) "universe missing after bionic->cosmic do-release-upgrade" [Critical,New] https://launchpad.net/bugs/1822886 15:12 <bdmurray> worked on a better method for u-r-u to check for releases and archive has 15:12 <bdmurray> approved netplan.io SRU exception 15:12 <bdmurray> review of juliank's update-notifier esm changes 15:12 <bdmurray> well done 15:14 <bdmurray> vorlon: I think that's everyone 15:14 <vorlon> great 15:14 <vorlon> any questions? 15:15 <vorlon> #topic Release incoming bugs (disco) 15:15 <vorlon> #link http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-dd-incoming-bug-tasks.html#foundations-bugs 15:15 <vorlon> bdmurray: ? 15:15 <bdmurray> bug 1821677 was tagged by doko and but it was on 4/1 15:15 <ubottu> bug 1821677 in glibc (Ubuntu) "dl_open segment fault in ubuntu18.10 glibc2.28" [High,New] https://launchpad.net/bugs/1821677 15:15 <vorlon> I see there's a new bug reporting a segfault in a glic we're not shipping 15:16 <infinity> Hrm? 15:16 <doko> I got a ping from upstream about that, didn't look at that in detail 15:16 <vorlon> from upstream of what? 15:16 <doko> glibc 15:16 <doko> glibc/intel 15:16 <infinity> "not shipping"? 15:16 <xnox> desktop team? 15:16 <vorlon> infinity: the bug report says 2.28 15:17 <infinity> Which is in cosmic. 15:17 <vorlon> but it's been marked dd-incoming 15:17 <vorlon> is it also reproducible in disco? 15:17 <infinity> That would be doko. 15:18 <infinity> I'll have a look at that today/tomorrow (doctor/etc premitting today) before I do the disco upload. 15:18 <vorlon> infinity: so you think we should commit to fixing it in disco? 15:18 <vorlon> (I can't reproduce it here on either cosmic or disco) 15:19 <infinity> Can't reproduce with their test case? 15:19 <infinity> Well, that might make it harder. :P 15:19 <infinity> I was committing to look, not to fix if I can't repro. 15:19 <vorlon> infinity: I'm not using their so, which I don't see attached 15:19 <vorlon> maybe it's in this tarball 15:19 <vorlon> anyway, I guess we can tentatively commit and uncommit later 15:19 * infinity nods. 15:20 <vorlon> ok, targeting 15:20 <vorlon> bdmurray: next? 15:20 <infinity> make: *** [Makefile:10: runtest] Segmentation fault (core dumped) 15:20 <infinity> Seems to repro here. 15:20 <vorlon> ok 15:20 <bdmurray> bug 1820676 15:20 <ubottu> bug 1820676 in ubiquity (Ubuntu) "Something has gone seriously wrong: import_mok_state0 failed" [Undecided,Triaged] https://launchpad.net/bugs/1820676 15:21 <vorlon> infinity: (I was trying to repro from the bug description, not the attached tarall) 15:21 <vorlon> b 15:21 <infinity> Might have been a mismerge of that patch, or just needs more ifdef arm. 15:22 <doko> I can confirm this one, renaming the exe helped 15:22 <vorlon> cyphermox: ^^ is that not fixed in the disco dailies? 15:22 <vorlon> related to missing mmx64 15:22 <cyphermox> it was fixed not long ago 15:22 <vorlon> was it fixed after 3/18? :) 15:22 <cyphermox> checking 15:23 <infinity> Would have been right around then. 15:23 <vorlon> bdmurray: let us follow up on that, it's probably fixed 15:23 <cyphermox> 3/20 15:23 <infinity> Cause the images weren't getting the new d-i bits due to a hung build on nusakan. 15:25 <cyphermox> spot checking on cdimage.u.c shows that ubuntu-mate and ubuntu dailies have the file 15:25 <infinity> Yeah, they all should now. 15:25 <bdmurray> that's all I noticed in dd-incoming 15:25 <cyphermox> yup 15:25 <vorlon> yeah I've followed up on the bug 15:26 <vorlon> I'm not bothering to target to disco because if there is a bug, it's probably not in a package in the archive 15:26 <vorlon> anyway I will follow through 15:26 <vorlon> bdmurray: what aout LP: #1815599 and LP: #1812688 that are on the report? 15:26 <ubottu> Launchpad bug 1815599 in Ubuntu on IBM z Systems "multipath shows '#:#:#:#' for iscsi device after error injection" [Medium,Triaged] https://launchpad.net/bugs/1815599 15:26 <ubottu> Launchpad bug 1812688 in automake-1.16 (Ubuntu) "self-check-report.sh fails on armhf/disco, was set to skip as an ubuntu delta without explanation" [High,Confirmed] https://launchpad.net/bugs/1812688 15:27 <bdmurray> the multipath one is incomplete, has it been updated? 15:27 <vorlon> xnox nominated the first 15:27 <vorlon> no 15:27 <vorlon> so I guess we can ignore 15:27 <vorlon> the other one, I tried to close and seb128 called me on it because I didn't read bdmurray's previous comment right 15:27 <xnox> vorlon, the multipath one was handindled by cpaelzer a bit 15:28 <vorlon> bdmurray: I think we should decline LP: #1812688, if we were willing to override the test to ignore the failure, it must not be critical enough to block a release 15:28 <ubottu> Launchpad bug 1812688 in automake-1.16 (Ubuntu) "self-check-report.sh fails on armhf/disco, was set to skip as an ubuntu delta without explanation" [High,Confirmed] https://launchpad.net/bugs/1812688 15:29 <bdmurray> vorlon: okay 15:29 <vorlon> done 15:29 <vorlon> #topic Team proposed-migration report 15:29 <vorlon> #link http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses_by_team.html#foundations-bugs 15:30 <bdmurray> I'm looking at apport 15:30 <vorlon> xnox: fuse-zip status? 15:30 <doko> g inggs pinged several times about scipy 15:30 <vorlon> I see you merged a new upstream version of fuse-zip, so that it can now fail its tests consistently on all archs 15:31 <infinity> vorlon, doko: Should cross-toolchain-base-whatever be addded to huge packages in autopkgtest, or should the massively massive (with hugeness) rebuild test maybe be pared down? 15:31 <vorlon> doko: I know he pinged me while I was on vacation and not in a position to do anything with it. If someone has a clear rationale for ignoring that test failure, MPs welcome 15:31 <xnox> vorlon, didn't get to it, but also not critical to migrate as far as i remember. yes which is an "improvement" =) my next trick was going to be to cherrypick fuse patch from upstream, or comment out / ignore the test that doesn't do the right clean up. 15:31 <xnox> vorlon, skipping one test, makes it all pass locally. 15:31 <cyphermox> infinity: +1, I think? 15:32 <xnox> (the test that tries to setup invalid/broken state, and assert on it) 15:32 <cyphermox> cross-toolchain-base-mipsen? 15:32 <doko> infinity: it's building exactly *one* cross toolchain. you could reduce the load further by not building the multilibs 15:32 <xnox> not sure what i should do about fuse-zip. horum. 15:32 * xnox tries to trow things into bileto ppa, to make it stick. 15:32 <doko> but vorlon wanted to look why it takes so much longer in some builds 15:32 <vorlon> xnox: "not critical to migrate" - someone in this channel uplodaed fuse with the intent that it release, and you committed to resolve this 15:33 <doko> xnox: should we remove fuse & fuse-zip from -proposed? 15:33 <vorlon> xnox: so if you're following up, all good 15:33 <xnox> vorlon, yes, and when i questioned why fuse was synced post debian import freeze no reply followed. doko - do you recall why we need/want new fuse? 15:33 <xnox> vorlon, but kind of can kicking, removing fuse now from -proposed will just come back in a couple of weeks. 15:34 <doko> I forgot, it was something to fulfill b-d's probably 15:34 <xnox> and well needs to be folved now, or later. 15:34 <xnox> doko, ah, and now they are built, we don't need it anymore, right ?! =)))))))) har har 15:34 <vorlon> we wouldn't know, since we can't launch new rebuild tests 15:34 <xnox> =) 15:35 <gaughen> vorlon, lies. ;-) 15:35 <vorlon> heh 15:37 <gaughen> so what are we doing with fuse? 15:37 <gaughen> removing it? 15:37 <infinity> Removing it is pointless. 15:37 <gaughen> so what are we doing with fuse? 15:38 <vorlon> infinity, doko: cross-toolchain-base: the issue was the variance in build times, which doko pointed to as an indicator of a problem in the autopkgtest infra; but I don't think giving it a bigger testbed prevents us from still doing that analysis 15:38 <vorlon> gaughen: xnox has a list of next steps for getting it migratable 15:38 <gaughen> k 15:38 <vorlon> and sphinx is still there 15:38 <xnox> vorlon, que? 15:40 <xnox> is it only me that gets very choppy audio 15:40 <xnox> vorlon, i can't hear you at all... 15:41 <waveform> xnox, audio's fine for me 15:42 <vorlon> xnox: sphinx, my recollection is that you were going to follow up on #ubuntu-release to sort out what to do about sphinx vs webkit2gtk, and I don't think that happened 15:43 <xnox> vorlon, we did talk about it with the desktop team, but i can't remember if we conclude something or not. 15:49 <vorlon> xnox: I didn't see a conclusion and it didn't get fixed in the archive :) 15:49 <vorlon> xnox: shall we try again to take this to #ubuntu-release after this meeting? ;) 15:49 <xnox> hm. 15:52 <vorlon> [TOPIC] AOB 15:54 <rbalint> i'd like to have LP: #1821924 rc 15:54 <ubottu> Launchpad bug 1821924 in lxd (Ubuntu) "LXD Deb->snap transition fails in WSL due to snap command not working (yet)" [Low,Triaged] https://launchpad.net/bugs/1821924 15:54 <rbalint> because it breaks do-release-upgrade which is ours 15:55 <rbalint> lxd is not maintained by our team, but release team could weigh in 15:55 <bdmurray> is rc a verb there? 15:55 <rbalint> release critical 15:55 <infinity> release criticalified 15:55 <bdmurray> rc'ed then 15:56 <rbalint> ok :-) 15:56 <infinity> Inappropriate apostrophe alert. 15:57 <rbalint> any comment on the actual bug? 16:01 <vorlon> so, since snaps don't work under wsl, should the lxd deb be removed as part of the upgrade? 16:03 <bdmurray> don't we do that with gnome-calculator and other desktop packages? 16:03 <xnox> vorlon, i think that's valid, at least for now. at the moment the new .deb fails to configure, blocking upgrade. 16:03 <xnox> vorlon, and it may happen on non-wsl systems too. 16:04 <xnox> vorlon, imho the error code path on snapd-less systems when doing do-release-upgrade is currently very bad. 16:04 <vorlon> it sounds like this is something we need to take offline 16:04 <vorlon> #endmeeting