15:03 #startmeeting 15:03 Meeting started Thu Mar 15 15:03:15 2018 UTC. The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 15:03 15:03 Available commands: action commands idea info link nick 15:03 [TOPIC] Lightning Round 15:03 $ echo $(shuf -e slangasek bdmurray xnox tdaitx doko sil2100 rbalint infinity cyphermox mwhudson rcj philroche Odd_Bloke tribaal fginther juliank) 15:03 xnox Odd_Bloke tdaitx bdmurray cyphermox sil2100 rbalint mwhudson fginther slangasek philroche infinity rcj tribaal juliank doko 15:03 xnox: HELLO 15:03 meeting moved up one hour in europe I think 15:04 IIRC 15:04 maybe we're all confused :D 15:04 sure, per the calendar :) 15:04 I'm not confused about the meeting time, lots of other things though. 15:06 Odd_Bloke is out today 15:06 tdaitx said he might miss the meeting 15:06 bdmurray: 15:06 * OpenJDK 9 as default JDK 15:06 - ongoing fixes and rebuilds 15:06 - syncing & waiting updates from debian 15:06 (done) 15:06 discussion with webops regarding retracer nagios alerts 15:06 uploaded software-properties fix for LP: #1693032 15:06 uploaded Artful SRUs for (LP: #1690541, LP: #1689668) 15:06 developed test case, uploaded fix for LP: #1693038 15:06 Launchpad bug 1693032 in software-properties (Ubuntu) "missing dependency on gnome-session-bin" [Medium,Fix released] https://launchpad.net/bugs/1693032 15:06 Launchpad bug 1690541 in aptdaemon (Ubuntu Artful) "Scrolled windows in update-manager are too small to read" [High,Fix released] https://launchpad.net/bugs/1690541 15:06 Launchpad bug 1689668 in aptdaemon (Ubuntu Artful) "dialog to resolve config file conflicts can be only one line high" [High,Fix released] https://launchpad.net/bugs/1689668 15:06 Launchpad bug 1693038 in software-properties (Ubuntu) "needs to support restart on Lubuntu and Xubuntu" [Medium,Fix released] https://launchpad.net/bugs/1693038 15:06 RU verification of aptdaemon (LP: #1690541, LP: #1689668) 15:07 investigation into retracer hangs (LP and ET) 15:07 discovered cause of retrace hangs (sandbox & glibc issue) 15:07 removed a bunch of core files from swift 15:07 updated retracer process clean up job for gdb call 15:07 investigation into rsyslog log rotation bug (LP: #940030) 15:07 Launchpad bug 940030 in rsyslog (Ubuntu Bionic) "rsyslog stops working after logrotate until restarted" [Critical,Fix released] https://launchpad.net/bugs/940030 15:07 research into how plymouth --ping works (LP: #1705345) 15:07 found a way to make plymouth --ping hang 15:07 wrote a couple of tools to clean up unnecessary core files 15:07 Launchpad bug 1705345 in plymouth (Ubuntu) "Installing packages hangs on plymouth --ping" [Critical,Triaged] https://launchpad.net/bugs/1705345 15:07 review of https://code.launchpad.net/~xnox/ubuntu-release-upgrader/lp1749199/+merge/341342 15:07 commented on LP: #1725359 again 15:07 Launchpad bug 1725359 in ubuntu-release-upgrader (Ubuntu) "release detection relies on lsb-release" [Low,In progress] https://launchpad.net/bugs/1725359 15:07 manual retracing / review of some Launchpad retracer failures 15:07 done 15:07 email discussion regarding snap errors at errors.u.c 15:08 uh, out of order done marker confused me 15:08 - MIR reviews 15:08 - doc/website/blog writing for netplan 15:08 - grub2 review for initramfs-less fallback 15:08 - [HIGHLIGHT] netplan.io 0.34 release 15:08 - preparing netplan.io 0.34.1: small bugfixes, build easier on other distros 15:08 - [HIGHLIGHT] made ubuntu-archive-assistant its own project 15:08 - investigating lubuntu ubiquity issues 15:08 - investigating casper netboot issues 15:08 (done) 15:08 Argh, timezone madness 15:09 o/ 15:09 sil2100: good timing, you're up :) 15:09 sil2100: your turn 15:09 I'm... I'm not ready! 15:09 At least you're here :) 15:09 (please skip me for now, I'll go last) 15:10 rbalint: 15:11 ok; and no fginther? 15:11 hmm, ok, me then 15:11 um no 15:11 fginther is on site today 15:11 surprise meeting ... 15:12 * juliank looked at unshuffled list... 15:12 * sprint last week; off Monday 15:12 * shepherding cloud kernels into bionic 15:12 * discussions around archive rebuild with retpoline; ultimately decided against based on most current information 15:12 * continued work on +1 maintenance (e.g. ruby), to clear the backlog for 18.04 release 15:12 * worked with tribaal to prepare an MP to grub2 to let systems do opportunistic initramfsless boot with fallback https://code.launchpad.net/~ubuntu-core-dev/grub/+git/ubuntu/+merge/341272 15:12 (done) 15:13 * Canonical sprint last week in Budapest 15:13 * Release review-gator snap https://dashboard.snapcraft.io/snaps/review-gator/ 15:13 * Monday swap day 15:13 * Cloud image build system maintenance and feature development 15:13 * Cloud image build system vanguard this week 15:13 * Worked on providing 18.04 preview to cloud partners. Users can grap a daily from http://cloud-images.ubuntu.com/bionic/current/ for testing. 15:13 (done) 15:13 * sil2100 is ready 15:13 sil2100: go ahead 15:14 - Last week: sprinting 15:14 - SRU reviews and releases 15:14 - Kernel reviews and releases 15:14 - FFe reviews 15:14 - Publishing the new google-cloud-sdk 15:14 - Working on and testing of proof-of-concepts for grub2 --auto-nvram 15:14 - Minor archive cleanups (stale/failed SRUs, removal of nplan) 15:14 - Trying to reproduce ubuntu-image issues with system-data + content: syntax 15:14 (done) 15:14 rcj: 15:15 * Engineering sprint in Budapest last week 15:15 * Discussed tweaks to improve to cloud-image CVE process 15:15 * Worked on image publication improvements to improve error handling 15:15 * Added squashfs image to final Precise MAAS serial to enable re-installs for ESM (amd64) 15:15 * Reviewing netplan.io site documentation 15:15 * Vacation day on Monday 15:15 (done) 15:17 tribaal: 15:17 tribaal is OoO 15:18 oh, tribaal is not here 15:18 - gnutls28 3.5.18 merge 15:18 - zstd emails 15:18 - got IS to enable https for changelogs.ubuntu.com (for meta-release files), changed update-manager and u-r-u to use that (bug 1744318), made apt(-get) changelog use it too ;) 15:18 - finding bugs on my newly setup bionic on my new laptop :) 15:18 (done) 15:18 oh, he is 15:18 well he's in the channel, how did I miss that 15:18 bug 1744318 in update-manager (Ubuntu Bionic) "changelogs.ubuntu.com should be using HTTPS" [High,Fix committed] https://launchpad.net/bugs/1744318 15:18 doko: 15:18 - GCC 7 & GCC 8 updates, including cross compilers 15:18 - enabled the sanitzer in GCC on s390x (nobody noticed before ...) 15:18 - cross-toolchain-base updates, built from glibc-2.27 15:18 - forwarded three GCC & binutils issues upstream 15:18 - python 3.7 beta2 15:18 - python 3.6 release candidate 1 15:18 - updated openjdk-10 package 15:18 - new openjdk-11 package \o/ 15:18 - made LLVM 6.0 the default LLVM 15:18 - processed around 10 MIRs 15:18 - fixing autopkg test failures ... ongoing saga 15:19 (done) 15:19 * xnox is very confused 15:19 xnox: do as the calendar says :) 15:19 xnox: daylight shreddings time 15:19 xnox: are you ready to give status? 15:21 • finished fixing keyring/invocation_id in containers both in systemd & kernel 15:21 • moving ruby transition along 15:21 • working on fixing network-manager <-> resolved integration 15:21 • fixing minimal image builds (in progress 15:21 • MIR/components missmatches zstd done 15:21 • done 15:21 * juliank is lucky gnome-online-account calendar stuff is broken so he looked at calendar a lot 15:21 xnox: "finished fixing" - oooh? 15:22 juliank: haha 15:22 any questions on status? 15:22 slangasek, yes; will resend kernel patch & will upload systemd patch. Basically i can make it work either way now. so to be fair kernel patch is redundant, but i know a lot more about keyrings now.... 15:23 xnox: but can you attach gif keychains to your keyrings 15:23 [TOPIC] Bugs 15:24 [LINK] http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-bb-incoming-bug-tasks.html#foundations-bugs 15:24 LP: #1753361 - xnox that looks like a straightforward request? 15:24 Launchpad bug 1753361 in systemd (Ubuntu) "Please set a maximum size for the sytemd journal" [High,New] https://launchpad.net/bugs/1753361 15:24 huh? 15:25 it does have one 15:25 I'll take #1751011 15:25 slangasek, it's already set 15:25 It's use up to 10%, and keep 15% free 15:26 there is a hard cap of 4GB as well 15:26 ok. should we take it for more investigation? 15:26 Well, 4.5G is about 4 15:26 doko: to do what with it? 15:26 doko: we discussed this bug previously and we had nacked changing bash to make qemu-user happy 15:27 in fact, I'm going to take that off the incoming list 15:27 slangasek, set that bug to invalid. 15:28 xnox: ok 15:29 slangasek: hmm 15:29 JFTR: I installed my T480s yesterday evening, and now at 4pm, journal is already using 4GB 15:29 slangasek: that's not recorded in the bug report ... 15:29 but might have been broken chrome :) 15:29 doko: I thought I commented on the bug 15:30 doko: but you also shouldn't be grabbing things from the rls-bb-incoming queue and committing to them individually, as that's an untriaged incoming queue that anyone can throw bugs into 15:30 or I could merge bash ... 15:30 juliank, good! please read your journal to figure out if it is gnome-shell spamming it! 15:31 to me it pukes stacktraces into journal all the time.... 15:31 I had a broken .cache/google-chrome and chrome was logging that multiple times a second :) 15:31 charming 15:31 xnox: is LP: #1662137 still present? you claimed it a year ago, now someone has tagged it for 18.04 15:32 Launchpad bug 1662137 in systemd (Ubuntu) "16.04 recovery shell works only for two minutes" [High,Confirmed] https://launchpad.net/bugs/1662137 15:32 (it's a symlink to /run/user/1000/google-chrome which I create with systemd-tmpfiles) 15:32 slangasek, we outht to fix that yes.... 15:32 xnox: ok, claiming/tagging 15:32 slangasek, i believe this is friendly-recovery vs systemd issue, e.g. friedly-recovery should conflict / pause the emmergency shell; or for example, use the emergency shell things 15:34 LP: #1750403 - I think this is already in progress and probably not actually a casper issue 15:34 Launchpad bug 1750403 in casper (Ubuntu) "Live Session - Increased memory usage with preinstalled snap (fails to start with - gnome-session timeout)" [High,Confirmed] https://launchpad.net/bugs/1750403 15:35 (will punt that for now, so we keep tabs on it) 15:35 LP: #1592405 definitely needs fixing; claiming 15:35 Launchpad bug 1592405 in plymouth (Ubuntu) "plymouth hook in initramfs needs font but doesn't Depend on it" [High,Triaged] https://launchpad.net/bugs/1592405 15:36 LP: #1601997 ? 15:36 Launchpad bug 1601997 in e2fsprogs (Ubuntu) "Ubuntu 16.10+ installer uses ext4 feature 'metadata_csum' which is incompatible with older (LTS) e2fsprogs" [High,Triaged] https://launchpad.net/bugs/1601997 15:36 slangasek, my opinion is that there are no actions to take. 15:36 +1 15:36 I agree 15:37 +1 15:38 the only place where this is an issue is dual-boot 15:39 someone did bring up the idea of an iscsi disk hosted on 16.04 serving out as a target to 18.04 15:39 I don't see how dual boot is an issue. It's -new- filesystems, not old ones suddenly changing :D 15:39 the 18.04 would add a filesystem flag that the 16.04 host wouldn't be able to handle 15:39 dunno if that's a realistic case or not, but it's something i hadn't considered before 15:40 nacc: so that falls on its face, and the user remakes the fs with -O -metadata_csum; done? 15:41 nacc, why would iscsi disk server; touch the exported disks? surely what filesystems client puts on iscsi disk, is none of the iscsi disk server problems...?! 15:41 slangasek: yes, that's probably true 15:41 xnox: fair response; I believe NAS-like devices do fs consistency checks periodically, but I might be wrong 15:41 it also wouldn't fail an fsck 15:41 xnox: it was simply a thought that was brought up by a responsive community member 15:41 nacc, no no no no =) that would be bad 15:42 sure. 15:42 it's not like new e2fsprogs can't understand filesystems made with old options 15:42 I think we all agree it's an "expert" use case, and those who do something affected should know to disable metadata_csum 15:42 I think you can even r/o mount it on an older _kernel_ release that does not understand metadata_csum, or am I wrong? 15:42 and 16.04 will get bionic's kernel 15:42 and that, yes 15:42 via hwe 15:42 discussed to death yet? 15:43 but there are people not using hwe :D 15:43 I've marked the bug wontfix 15:43 LP: #1682637 15:43 Launchpad bug 1682637 in friendly-recovery (Ubuntu) "during recovery mode, enable network failed due to /etc/resolv.conf not being present" [High,Triaged] https://launchpad.net/bugs/1682637 15:43 xnox probably wants this one too :) 15:43 sure 15:48 ok there are still a few more in the queue but I think that's enough time spent on these for now 15:48 [TOPIC] AOB 15:48 anything else? 15:49 Nope 15:49 Nope 15:49 #endmeeting