== Meeting information == * #ubuntu-meeting Meeting, 10 Jan at 16:02 — 16:57 UTC * Full logs at [[http://ubottu.com/meetingology/logs/ubuntu-meeting/2019/ubuntu-meeting.2019-01-10-16.02.log.html]] == Meeting summary == === Lightning round === The discussion about "Lightning round" started at 16:02. === Bugs === The discussion about "Bugs" started at 16:18. === 18.04.2 === The discussion about "18.04.2" started at 16:35. * ''LINK:'' https://bugs.launchpad.net/ubuntu/bionic/ actually shows no bugs targeted to 18.04.2 at all === AOB === The discussion about "AOB" started at 16:49. == Vote results == == Done items == * (none) == People present (lines said) == * vorlon (78) * cyphermox (37) * bdmurray (33) * juliank (32) * sil2100 (28) * doko (21) * ubottu (16) * tdaitx (14) * rbalint (13) * gaughen (8) * waveform (7) * meetingology (3) == Full Log == 16:02 #startmeeting 16:02 Meeting started Thu Jan 10 16:02:16 2019 UTC. The chair is vorlon. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 16:02 16:02 Available commands: action commands idea info link nick 16:02 [TOPIC] Lightning round 16:02 $ echo $(shuf -e vorlon bdmurray xnox tdaitx doko sil2100 rbalint infinity cyphermox mwhudson juliank waveform) 16:02 rbalint mwhudson xnox infinity juliank cyphermox waveform doko tdaitx vorlon sil2100 bdmurray 16:02 Not sure if rbalint is still around 16:02 o/ 16:02 Oh! 16:03 * update-excuses: 16:03 - fixed hidden ABI breakage in imagemagick in disco 16:03 - triaged and added hint for twisted 16:03 * verified unattended-upgrades srus 16:03 * partner work and other discussions 16:03 * delayed NMU-d imagemagick in Debian 16:03 * uploaded forked-daapd 26.4-1, wireshark 2.6.6-1, libnfs 3.0.0-1 16:03 * performed the libnfs transition in Debian and Ubuntu 16:03 * helped with resolving the imagemagick ABI breakage in Debian and asked for rebuilds in Debian and did them in Ubuntu 16:03 * updated and sru-d gce-compute-image-packages 16:03 * reported LP: #1811210 16:03 Launchpad bug 1811210 in initramfs-tools (Ubuntu Bionic) "intramfs-tools 0.130ubuntu3.3 in bionic-security breaks netplan.io in -security or in release" [High,New] https://launchpad.net/bugs/1811210 16:03 (done) 16:04 no mwhudson or xnox 16:04 don't see infinity so juliank is up 16:04 * Some update_excuses triaging to get stuff migrated, especially gnutls28; also uploads for that (in below) 16:04 * merged wpa, wget, lz4, amd64-microcode (still needs SRUs I think), packagekit, gnupg2, dnsmasq 16:04 * fixed bugs in hddemux, gnupg2, mariadb-10.1 autopkgtests 16:04 * prepared fix for mysql-5.7 autopkgtests, rbasak uploaded to sid and synced it 16:04 * apt stuff: 16:04 - improved the apt-mark minimize-manual merge request 16:04 - sent an email to deity@l.d.o to move discussion of repo enable/disablement forward (-32768 pinning) 16:04 - sent an email to deity@l.d.o to discuss new versioning scheme / pkg-config compatibility 16:04 - (also look forward to a ~5% performance improvement on cache open on 2011+ x86 [sse4.2 crc32c cache hashing] I achieved for christmas :D) 16:04 * stuff forwarded to Debian: 16:04 - debian bug 918666: wine hardcodes /run/user/$UID directory (which I fixed in gnupg2 for the time being, not sure if /run/user/$UID is supposed to exist everywhere) 16:04 - debian bug 918608: hddemux: testsuite: Avoid failure when https_proxy is set 16:04 Debian bug 918666 in src:wine "wine hardcodes /run/user/$UID directory" [Normal,Open] http://bugs.debian.org/918666 16:04 Debian bug 918608 in hddemux "hddemux: testsuite: Avoid failure when https_proxy is set" [Normal,Open] http://bugs.debian.org/918608 16:04 - reason for debian bug 918722 - caused recent upstream debootstrap to fail to bootstrap ubuntu (that upload was then reverted) 16:04 Debian bug 918722 in debootstrap "debootstrap: says InRelease file expired" [Serious,Fixed] http://bugs.debian.org/918722 16:04 (done) 16:05 cyphermox: 16:05 [HIGHLIGHT] netplan.io is now in Debian; synced to Disco for now. 16:05 paperwork: 16:05 - shim-review: CPSD, todo today: ROSA, CentOS, SUSE, OS ALT. send help^H^H^H^H^H^H 16:05 - small fixes to DKMS doc 16:05 - merging Debian netplan changes into upstream git tree 16:05 juliank: how are these implemented, ifunc? 16:05 disco: 16:05 - grub2: added breadcrumbs for grub-mkconfig (parsing which files when generating conf) 16:05 - grub2: fixing up grub-check-signatures to compare against known keys in firmware 16:05 doko: function multiversioning 16:05 bionic, cosmic: 16:05 - attempting to fix syslinux -- patch for gnu-efi 3.0.8 apparently needs update 16:05 but new patch FTBFS... annoying and hard to validate this SRU 16:05 xenial: 16:05 - verified shim-signed update to finish upgrade to shim 15 16:05 - dkms + shim-signed to finish Secure Boot signing changes 16:06 trusty: 16:06 - updated grub2 patchset for UEFI Secure Boot 16:06 - apply patches for chainloading issues 16:06 - dkms + shim-signed to finish Secure Boot signing changes 16:06 (✓ done) 16:06 doko: one is defined with __attribute__((target("sse4.2")), one with __attribute__((target("default")) 16:06 see https://salsa.debian.org/apt-team/apt/commit/37ae749b20485b6c8237d5b5a08cfdd58a2364e1 16:06 and https://salsa.debian.org/apt-team/apt/commit/fb1c5167f1a69fc47ca8e741b86fab1ae7c08309 with a fix 16:07 waveform: hi, did anyone warn you about this? any status you want to share? 16:08 I think I forgot to mention to waveform to prepare notes beforehand! 16:08 besides drinking from the firehose 16:08 :) 16:08 ahhh, that would've been ... erm ... useful :) 16:08 no worries 16:08 doko: 16:08 - finished the disco / disco-gcc9 test rebuilds 16:08 - start submitting compiler issues for GCC 9 16:08 - GCC 8 & 9 updates 16:08 - jpeg-turbo update to fix ftbfs 16:08 (done) 16:08 anyway, irqbalance issue found in aarch64 rpi3 image, grabbed fix from upstream 16:08 waveform: sorry! Well, I was writing those by hand in real time in the past as well when I first joined the team 16:08 \m/ 16:09 tdaitx: 16:09 that means I should go update my rpi3 now I guess 16:09 A happy new year to you all! And warm welcome to waveform =) 16:09 * openjdk security update 16:09 - openjdk 8 and 11 from oracle, openjdk 7 from icedtea update 16:09 - patch was made available during the holidays 16:09 * openjdk-lts autopkgtest disk space issues 16:09 - cosmic have autopkgtest enabled but it is not cleaning up tests 16:09 - checking a fix for disco (which also enabled autopkgtests) 16:09 cyphermox: not yet! ;p 16:09 Other: 16:09 - classes by EOD today and again in two weeks (22-24th January) 16:09 and built u-boot 2018.07-rc3 (working on 3b and 3b+) 16:09 - apartment under renovation pushing me to work on odd hours sometimes, also need to run some errands every now and then, will notify on irc when something big is ahead 16:09 (done) 16:09 sil2100: ;.; 16:09 cyphermox: ...soon 16:10 anyway, that's it from me - working on building images otherwise 16:10 vorlon: 16:10 * happy new year! 16:10 * deep dive on fstrim support in scalingstack to debug a complaint from IS about autopkgtest instance disk usage 16:11 * ubuntu-drivers,cloud-init design review 16:11 * a little bit of transition wrangling 16:11 * first Debian MU of pam in a while 16:11 * prep for Cape Town Sprint 16:11 * upcoming: 16:11 * out tomorrow for travel to cape town; back Jan 28 16:11 (done) 16:11 sil2100: 16:11 - Some kernel SRU reviews and releases 16:11 - SRU reviews and releases 16:11 - Giving Dave some work to do! 16:11 - core18 PR reviews for additional testing and making travis CI happy 16:11 - Verification of the ubuntu-image 1.6 SRU, bumping hints 16:11 - raspi3 'upstreaming': 16:11 * Pushed flash-kernel changes to disco 16:11 vorlon: oh, so I get to chair the meeting next week? 16:11 * Filled in SRU bugs for flash-kernel and uploaded SRUs to cosmic and bionic 16:11 * Fetched ubuntu-core's raspi3-firmware version and SRUed it to bionic 16:11 * Checked Dave's u-boot packages based on the ubuntu-core's version 16:11 * Enabled disco raspi3 builds on cdimage, added missing livecd-rootfs changes 16:11 * Started working on enabling multiverse during raspi3 gadget tree builds 16:12 - Looked into the Bileto outage after code update (settings got wiped) 16:12 * Bileto still not completely 'fixed', some issues with redirects still 16:12 - Reviewed Brian's quick-fix for sru-report 16:12 - Discussion regarding core18 beta->candidate test feedback fetching 16:12 (done) 16:12 Why are we fixing bileto? 16:12 huh, speaking of rpi3, is this still only for the rpi3 b+ or something? some newer rev of the hardware? 16:12 This includes some stuff I did before the break 16:12 bdmurray: because people are using it 16:12 cyphermox: if you wish, and the week after too :) 16:13 vorlon: zug zug 16:13 sil2100: I thought we wanted to get rid of it 16:13 tested and converted three daisy scripts from pycassa to python-cassandra 16:13 submitted MP regarding mojo spec update for daisy code revno 16:13 submitted RT regarding update of daisy code in production and testing cronjobs 16:13 worked with webops regarding issues with daisy code update 16:13 uploaded B,C fixes for software-properties-gtk crash LP: #1807373 16:13 Launchpad bug 1807373 in software-properties (Ubuntu Cosmic) "/usr/bin/software-properties-gtk:AttributeError:on_driver_selection_changed" [High,Fix released] https://launchpad.net/bugs/1807373 16:13 SRU verifications of LP: #1807373, LP: #1807043, LP: #1773637 16:13 Launchpad bug 1807043 in ubuntu-release-upgrader (Ubuntu Bionic) "when rewriting sources.list ubuntu-release-upgrader doesn't check to see if new release is available" [Medium,Triaged] https://launchpad.net/bugs/1807043 16:13 Launchpad bug 1773637 in ubuntu-release-upgrader (Ubuntu Cosmic) "Distribution upgrade to 18.04: used nvidia display driver is commented out" [Undecided,Fix committed] https://launchpad.net/bugs/1773637 16:13 investigated and uploaded fix for nvidia driver upgrade issue LP: #1773637 16:13 reported ubuntu-drivers-common bug LP: #1808874 16:13 worked on, uploaded fixes for issues with apport autopkgtests in disco 16:13 worked on migration for part of oopses.py from oopsrepository to python-cassandra 16:13 modified daisy to use renamed oopsrepository module 16:13 submitted RT re update of daisy and oopsrepository in production 16:13 fixed another apport test issue due to usrmerge 16:13 ✔ done 16:14 Launchpad bug 1808874 in ubuntu-drivers-common (Ubuntu) "nvidia-detector doesn't seem to detect nvidia" [Low,New] https://launchpad.net/bugs/1808874 16:14 bdmurray: there's no decision regarding that yet 16:14 cyphermox, for the 3b and 3b+ (and allegedly 2b but I haven't tested that) 16:15 but none of the single-core models 16:15 bdmurray: we're doing bare maintenance for now, so only fixing stuff when it breaks, but even then with not super high priority 16:15 waveform: thanks; I'll have to check again which rev I have... if it's not the right one I'll repurpose for legacy gaming anyway 16:15 waveform: I'll go add you to my meeting script now, too, I guess 16:15 tdaitx: is the 11 update already released? can't find a tag 16:15 sil2100, bdmurray we weren't doing any new work on bileto, but were keeping the lights on. 16:15 doko: no, not yet released, comes from partner 16:15 but we had discussed that if at some point keeping the lights on got really hard, we would need to look at ending it 16:15 gaughen: so the lights went off and we turned them back on? 16:16 bdmurray, we jiggled the switch 16:16 other questions over status? 16:17 cyphermox, did you push netplan.io to debian? 16:17 no questions here 16:17 negative, andrewsh did; he did put me up as uploader though 16:18 excellent 16:18 I had caught up with him at Plumbers 16:18 vorlon, that's all for me. 16:18 this is good. 16:18 it is good. 16:18 [TOPIC] Bugs 16:18 bdmurray: I wonder if we should talk valgrind 16:18 there are none. 16:18 vorlon: I was thinking we sould 16:18 should 16:18 Zarro boogs found. 16:19 LP: #1808508 16:19 Launchpad bug 1808508 in valgrind (Ubuntu) "Valgrind doesn't work in disco [Fatal error at startup: a function redirection which is mandatory for this platform-tool combination cannot be set up.]" [Critical,Triaged] https://launchpad.net/bugs/1808508 16:20 So that's keeping apport stuck in -proposed 16:20 and it's fundamentally not an apport or valgrind bug, it's a bug in the filesystem as a result of usrmerge 16:21 so I could probably see my way to a skiptest hint around apport currently, to unblock development 16:21 but this is important to fix for release 16:22 vorlon: you also mentioned this would affect gdb too 16:22 yes 16:23 I hate the gdb debug symbols lookup code ... 16:26 so one question is, if gdb and valgrind are the only things that need it, should we just extend their debug symbol lookup paths 16:27 libunwind was suggested, but: 16:27 $ strings /usr/lib/x86_64-linux-gnu/libunwind.so.8|grep usr/lib/debug 16:27 $ 16:27 so no hits there 16:27 ah, and gdb actually uses /usr/lib/debug/.build-id 16:27 so maybe only valgrind needs changing 16:28 vorlon: src/dwarf/Gfind_proc_info-lsb.c: static const char *debugdir = "/usr/lib/debug"; 16:29 but maybe it's disabled 16:29 build id is used by debhelper v9 and above 16:29 so we might as well fix valgrind and then we can have a philosophical discussion later about whether /usr/lib/debug should be usrmerged 16:29 I think I saw google-perftools broken once 16:29 juliank: libc6-dbg has .build-id 16:29 that was on my usrmerged debian system, though 16:29 so I think the common case is handled 16:29 alright, I think the quick path is to just fix valgrind 16:30 any volunteers? 16:30 one other option might be to have the debug package generate the symlinks as needed? 16:30 cyphermox: please no :) 16:30 oh, wait, no 16:30 yeah, no 16:30 ;) 16:30 I'm happy to fix valgrind 16:31 cyphermox: assigning, thanks 16:31 card? 16:31 I can make a card for you buddy 16:32 :) 16:32 tdaitx: Should there be a card for bug 1808383 16:32 bug 1808383 in ant (Ubuntu Disco) "Java 9+ multi-release jars are not supported in ant tasks" [Undecided,Confirmed] https://launchpad.net/bugs/1808383 16:34 bdmurray: sure, I will create one for it 16:34 tdaitx: cool 16:34 that's it from me regarding bugs then 16:35 [TOPIC] 18.04.2 16:35 specifically bugs for 18.04.2, so maybe a subtopic :) 16:35 do we have anything that we think should be fixed in point release but is not yet targeted? 16:36 https://bugs.launchpad.net/ubuntu/bionic/ actually shows no bugs targeted to 18.04.2 at all 16:38 there are some subiquity cards targeted, which is understood 16:38 maybe I should get this u-r-u SRU done before the point release 16:39 My goal is to get bug 1787460 fixed 16:39 bug 1787460 in linux-meta-hwe (Ubuntu) "Unattended upgrades removed linux-image-generic" [Critical,Triaged] https://launchpad.net/bugs/1787460 16:39 the apt part of it 16:40 feature bug 1811120 is also interesting, but could also wait 16:40 juliank: is that in a card? 16:40 bug 1811120 in apt (Ubuntu Cosmic) "Backport auth.conf.d" [Low,Triaged] https://launchpad.net/bugs/1811120 16:40 vorlon: it is 16:40 juliank: can you put the 18.04.2 tag on the card? 16:40 vorlon: done 16:40 the card for LP: #1793671 is targeted to 18.04.2 but is marked blocked 16:40 Launchpad bug 1793671 in ubiquity (Ubuntu Bionic) "Proprietary drivers are installed locally and not on the target system" [Undecided,New] https://launchpad.net/bugs/1793671 16:41 I think we need https://bugs.launchpad.net/ubuntu/+source/livecd-rootfs/+bug/1805668 on the milestone 16:41 Launchpad bug 1805668 in u-boot (Ubuntu Cosmic) "Building armhf and arm64 Raspberry Pi 3 images" [Undecided,Confirmed] 16:41 gets my vote :) 16:41 I'm untargeting "Make images not mark all task packages as manually installed" for 18.04.2 16:41 sil2100: agreed, milestone set 16:42 because #ubuntu-devel just had some potentially related regressions 16:43 juliank: doh, ok 16:43 (bug bug 1801629 fwiw) 16:43 bug 1801629 in xubuntu-meta (Ubuntu) "xubuntu-core needs to depend on cryptsetup and lvm2 or 'apt autoremove' will make a LUKS+LVM encrypted root partition non-bootable" [Undecided,Confirmed] https://launchpad.net/bugs/1801629 16:43 Should import as a card 16:44 was https://bugs.launchpad.net/cryptsetup/+bug/1767527 mentioned already? 16:44 Launchpad bug 1767527 in cryptsetup (Ubuntu) "[18.04] Installation boot failure. WARNING: invalid line in /etc/crypttab" [Critical,In progress] 16:44 we have an 18.04.2-targeted card for that 16:44 vorlon: thanks! 16:45 cyphermox: ^^ that cryptsetup bug has a card in progress that lies somewhere between you and xnox 16:46 is it in progress from your POV? 16:47 my question on the bug in October was never answered; marking it incomplete for now 16:47 ok so anything else for 18.04.2? 16:48 targeted bug list is now: https://bugs.launchpad.net/ubuntu/bionic/?field.milestone%3Alist=86800 16:48 going once... 16:49 twice... 16:49 [TOPIC] AOB 16:49 wait, before AOB ... 16:49 devel-proposed: https://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses_by_team.html#foundations-bugs 16:49 I'll be out tomorrow / swapping for some other holiday 16:50 right 16:50 apport already discussed 16:50 the only one we addressed is the valgrind one 16:50 anyone want to sort out autopkgtest/autodep8 regressing itself? 16:52 the other high profile one is ros-ros-comm 16:52 and systemd of course 16:52 doko: feel free to follow up out-of-band, but are the linux/binutils autopkgtest regressions ignorable? 16:53 ros-ros-comm looks like something that just needs a badtest hint rather than a regression we've caused 16:53 paramiko is also blocked by sahara test failures which might be more real 16:54 and foundations needs to file some MIRs: http://people.canonical.com/~ubuntu-archive/component-mismatches-proposed.svg 16:54 I don't think we should try to go through all of the rest right now, we clearly have a backlog due to the break 16:55 that's fine, but then we should reserve some time next week 16:55 doko: uchardet? 16:56 yes, kronosnet, intel-ipset-* 16:56 mailutils? 16:56 maybe just a suggests 16:56 how are those foundations? the mouseover says server 16:56 emacs is desktop 16:56 ;P 16:56 doko: I'm doing some autopkgtest stuff I could probably have a look at that 16:56 ohh, even better =) 16:57 anyway, someone want to volunteer please to open the MIR for uchardet? 16:57 ok, then I'll write uchardet myself 16:57 ok 16:57 and then we'll close the meeting 16:57 #endmeeting Generated by MeetBot 0.1.5 (http://wiki.ubuntu.com/meetingology)