15:01 <slangasek> #startmeeting 15:01 <meetingology> Meeting started Thu Apr 28 15:01:56 2016 UTC. The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 15:01 <meetingology> 15:01 <meetingology> Available commands: action commands idea info link nick 15:03 <slangasek> [TOPIC] Lightning round 15:03 <slangasek> $ echo $(shuf -e barry doko bdmurray slangasek caribou infinity sil2100 robru cyphermox pitti tdaitx xnox chiluk) 15:03 <slangasek> sil2100 xnox chiluk slangasek doko cyphermox pitti robru infinity caribou bdmurray tdaitx barry 15:03 <slangasek> no sil2100 at the moment, so xnox ? 15:04 <slangasek> hah, no, xnox is off today 15:04 <slangasek> chiluk: 15:04 <chiluk> Sorry I worked on no "user" issues this week resulted in valid public launchpad case work this week. 15:04 <chiluk> -done- 15:04 <chiluk> it was a depressing week 15:05 <slangasek> chiluk: do we need to send you more users that file valid bugs? 15:05 <pitti> I think we have two left 15:05 <cyphermox> slangasek: I think chiluk is looking for stuff to do, he could do the debian-installer merge? 15:05 <pitti> (per minute or so) 15:05 <chiluk> lol.. 15:05 <slangasek> heh 15:05 <cyphermox> :) 15:06 <doko> flash-kernel is another good one 15:06 <slangasek> * release week, congrats on 16.04! 15:06 <slangasek> * tried to SRU a pre-dependency into sysv-rc for xenial to remove some conffile prompting, got LP: #1573240 for my trouble 15:06 <ubottu> Launchpad bug 1573240 in sysvinit (Ubuntu) "package initscripts 2.88dsf-59.2ubuntu2.1 failed to install/upgrade: pre-dependency problem - not installing initscripts" [Critical,Confirmed] https://launchpad.net/bugs/1573240 15:06 <slangasek> * triaged some incoming reports of upgrade failures against shim-signed, so far none of them appear to be (fixable) bugs in the SecureBoot stack 15:06 <cyphermox> doko: flash-kernel is next on my list. 15:06 <slangasek> * merges for yakkety 15:06 <slangasek> * taking a few minutes every day to try to absorb the name 'yakkety' 15:06 <slangasek> * sprint prep 15:06 <slangasek> I'm TIL on flash-kernel, how is it on any of your lists? :) 15:06 <slangasek> (done) 15:07 <cyphermox> slangasek: if you want to do it I won't complain :) 15:07 <doko> - archive opening 15:07 <doko> - icu, libpng, boost transitions 15:07 <doko> - more transitions, ftbfs fixes, merges, ... 15:07 <doko> - GCC 6.1 rc2 and GCC 6.1 release 15:07 <doko> - openjdk-8 updates 15:07 <doko> (done) 15:07 <cyphermox> - console-setup merge 15:07 <cyphermox> - trivial watchdog service fix (bug LP: #1559468) 15:07 <cyphermox> - fix ltrace on ppc64el (bug LP: #1398143) 15:07 <cyphermox> - debugging NM wifi issues 15:07 <cyphermox> - fixing klibc ipconfig DHCPREQUEST calls (bug LP: #1327412) 15:07 <cyphermox> - all the SRUs for Secure Boot 15:07 <ubottu> Launchpad bug 1559468 in watchdog (Ubuntu Wily) "Missing single-quote in watchdog.service prevents service from working correctly" [Low,In progress] https://launchpad.net/bugs/1559468 15:07 <cyphermox> - porting mokutil and efivar to precise & trusty 15:07 <cyphermox> (done) 15:07 <ubottu> Launchpad bug 1398143 in ltrace (Ubuntu Trusty) "Ltrace is broken on ppc64el" [Low,New] https://launchpad.net/bugs/1398143 15:07 <ubottu> Launchpad bug 1327412 in klibc (Debian) "Delay during PXE Boot, IP-Config gives up" [Unknown,New] https://launchpad.net/bugs/1327412 15:08 <pitti> autopkgtest: 15:08 <pitti> - Add option to set more appropriate parallel= build option in LXC (#1569750) 15:08 <pitti> - Add workaround for broken PTMU in Canonical's network (#1572026) 15:08 <pitti> - Make "git clone" more robust against temporary network glitches (#1571979) 15:08 <pitti> - Automate setup of armf LXD test controller and slaves in Scalingstack; now running, but failing an awful lot due to kernel bug (#1531768) 15:08 <slangasek> oh, forgot to list that I've also been working to clean up the 'unsubscribed' package list on http://reqorts.qa.ubuntu.com/reports/m-r-package-team-mapping.html#unsubscribed by finding people to blame for these 15:08 <pitti> - create watchdog for auto-hard-rebooting arm64 instances, as a bandaid for the above 15:08 <pitti> - various bug fixes in adt-run 15:08 <pitti> - Merge debci with upstream git 15:08 <pitti> - Set up configs, cloud images, and LXC containers for yakkety, some more intense maintenance during the yakkety opening flood 15:08 <pitti> FTBFS fixes: dict-moby-thesaurus, grep, libvigraimpex, mako, python-configglue, ros-robot-model, ubuntu-drivers-common 15:08 <pitti> apport: Fix invalid key name generation in hookutils (#1566975) 15:08 <pitti> apt: fix bad pre-depends handling, breaks upgrades a lot (#1560797) 15:08 <pitti> casper: Suppress auto-activation of swap partitons in live session (#1552539) 15:08 <pitti> ubiquity: Fix ubiquity-dm startup failure (#1570901) 15:08 <pitti> juju-core: Investigate test failure with mwhudson, propose fixes (#1571082) 15:08 <pitti> Debug installer swap space creation failure, come up with bandaid in casper (#1552539) 15:08 <pitti> lxc: Debug autopkgtest failure and recommend simple fix (#1572188) 15:08 <pitti> lxd: Investigate/fix looping startup failure (#1569698) 15:08 <pitti> systemd: Fix enabling of linked units (#1563590), fix lightdm flakiness in boot-smoke test (#1571673), fix wrong device naming with ibmveth (#1561096), fix USB network device names with randomized MACs (#1574483) 15:08 <pitti> Investigate ssh installation failure in QA desktop tests, fixed in UTAH now (#1554266) 15:08 <pitti> Review proposed openssl FIPS change, landed right before Xenial release (#1553309) 15:08 <pitti> merges: systemd, init-system-helpers, debhelper, gnome-color-manager 15:08 <pitti> Set up apport retracers and ddebs for yakkety 15:08 <pitti> tons of bug triage from incoming bug flood after xenial release 15:08 <pitti> ~ end ~ 15:09 <infinity> robru? 15:11 <pitti> I'd say timeout and infinity goes on? 15:11 <slangasek> oh yes, robru is out, sorry 15:11 <infinity> - Released xenial 15:11 <infinity> - Opened yakkety 15:11 <infinity> - Lots of work related to opening yakkety (LP, chroots, snakefruit, etc) 15:11 <infinity> - Start on yakkety merges 15:11 <infinity> - Upgrade buildds to xenial final (s390x) and trusty+lts-xenial (ppc/arm64) 15:11 <infinity> - Fix some FTBFS fallout from the -fPIE switch 15:12 <infinity> - Upgrade-related SRU work for trusty->xenial 15:12 <infinity> (done) 15:12 <slangasek> caribou: 15:12 <caribou> shortr week (one day off) : 15:12 <caribou> Bugfix : 15:12 <caribou> - multipath coredumps 15:12 <caribou> - Hash mismatch issues with debian archives when using mirror: 15:12 <caribou> - network kdump bugfix 15:12 <caribou> (done) 15:13 <chiluk> pitti: https://www.youtube.com/watch?v=PtTC3pGBjs4 15:14 <infinity> pitti: Did you ever figure out how you accidentally deleted ddebs and get switched over to the new key? 15:14 <tdaitx> bdmurray: is out as well... he sent his status to the ML 15:14 <tdaitx> = Security update for OpenJDK 7 15:14 <tdaitx> * Compared my broken hotspot patch to IcedTea's working one to understand what I missed 15:14 <tdaitx> * Packaged IcedTea 2.6.6, updating OpenJDK 7 to 7u101-2.6.6; had to disable arm32 jit as hotspot security update broke it 15:14 <tdaitx> * Backported OpenJDK 7u101 from Wily to Trusty and Precise; provided the packages to the security team 15:14 <tdaitx> * Investigated arm32 jit to understand why it broke and - hopefully - how to fix it; the hotspot security update introduced a new bytecode into the Bytecode enum, but the arm32 jit uses hardcoded bytecode values and addresses all over the place 15:14 <tdaitx> = TCK 15:14 <tdaitx> * Back to testing on Xenial; testing scripts on a clean environment (newly deployed lxc agents and hosts) 15:14 <tdaitx> - Kerberos setup was broken, fixed so it works on both Xenial and Trusty 15:14 <tdaitx> - Pulseaudio was not starting up; fixed that; unable to load/change modules; fixed 15:14 <tdaitx> - investigating why a bunch tests are now failing 15:14 <tdaitx> * Once more going through TCK docs to ensure that both scripts and environment are in compliance with the requirements 15:14 <tdaitx> * Tidying up JIT/configuration template 15:14 <pitti> infinity: yes I did, and that's fixed now 15:14 <tdaitx> = Other 15:14 <tdaitx> * Tried my hand again at backporting OpenJDK 9 HiDPI fix to OpenJDK 8, still not clear how to backport some changesets (LP: #1497816); kind of a personal issue to me as I have a HiDPI display 15:14 <ubottu> Launchpad bug 1497816 in openjdk-7 (Ubuntu) "No support for HDPI/retina displays in Java 7/8" [High,Confirmed] https://launchpad.net/bugs/1497816 15:14 <tdaitx> * Upgraded to Xenial; planning move to ZFS 15:14 <tdaitx> * Getting used to loud noises after neighbors decided to renovate the whole apartment 15:14 <tdaitx> * Went through a few bugs 15:14 <infinity> pitti: \o/ 15:15 <barry> tdaitx: done? 15:16 <tdaitx> (done) 15:16 <tdaitx> sorry =) 15:16 <barry> shaving the yak 15:16 <barry> dirtbike 0.3-2; python-pip 8.1.1-2 (debian bug #821014 and debian bug #814292); debian bug #821223 (virtualenv _markerlib); debian bug #821442 (schroot broken); 15:16 <ubottu> Debian bug 821014 in src:python-pip "Recent versions of setuptools no longer has or needs _markerlib" [Normal,Fixed] http://bugs.debian.org/821014 15:16 <ubottu> Debian bug 814292 in python-pip "Missing python-setuptools dependency" [Normal,Fixed] http://bugs.debian.org/814292 15:16 <barry> various other archivy things for 16.04 15:16 <ubottu> Debian bug 821223 in virtualenv "Unable to create a virtualenv: invalid requirement: '_markerlib'" [Grave,Fixed] http://bugs.debian.org/821223 15:16 <ubottu> Debian bug 821442 in src:linux "overlay fs: chmod fails (Operation not permitted) on chowned files" [Important,Fixed] http://bugs.debian.org/821442 15:16 <barry> lots of playing (read: bug reports :) around w/juju 2 and charms; si-server charm 15:16 <barry> pycon 2016 language summit planning 15:16 <barry> LP: #1572416 verification-done 15:16 <ubottu> Launchpad bug 1572416 in ubuntu-release-upgrader (Ubuntu Xenial) "do-release-upgrade crashes in Greek locale" [High,In progress] https://launchpad.net/bugs/1572416 15:16 <barry> syncpackage bzr-fastimport back into ubuntu 15:16 <barry> debian bug #822750 (unreproducible) 15:16 <ubottu> Debian bug 822750 in python3-pip "python3-pip: Missing deps on packaging, distlib, ..." [Normal,Open] http://bugs.debian.org/822750 15:16 <barry> --done-- 15:18 <slangasek> ok, any questions? 15:20 <slangasek> sounds like there aren't :) 15:20 <slangasek> [TOPIC] AOB 15:20 <slangasek> anything else today? 15:21 <slangasek> anyone have topics for UOS that they need help getting scheduled? 15:21 <pitti> I filed https://blueprints.launchpad.net/ubuntu/+spec/desktop-y-replace-upstart today 15:21 <slangasek> or topics for next week's sprint that they need help getting scheduled? :) 15:21 <pitti> apparently the first-ever blueprint for that uos 15:21 <slangasek> hah 15:21 <slangasek> pitti: you've invited appropriately desktoppy people to the session? 15:21 <pitti> slangasek: seb128 and xnox so far, I think I'll grab Laney too 15:22 <slangasek> sounds good 15:22 <slangasek> maybe invite willcooke also? 15:22 <pitti> sure, I'll mark him as non-essential 15:22 <pitti> (done) 15:23 <pitti> but that's just boring tech debt, nothing new an exciting really 15:23 <infinity> That's pretty exciting, IMO. 15:23 <infinity> Given that upstart is actively bitrotting. 15:24 * infinity glares at the build failures. 15:24 <pitti> another topic that I have on the shelf somewhere is to enable DNSSEC early in the cycle and get some experience with it 15:24 <pitti> (we'd disable it for the release as there's still some known-broken issues) 15:24 <slangasek> DNSSEC by default? 15:24 <pitti> infinity: yeah, that's my main concern 15:24 <slangasek> yes, we also have to get the phone off of upstart at some point 15:24 <pitti> yes; there was some heavy work in resolvd recently to do that, and get along with the common quirks of routers, etc. 15:25 <slangasek> doesn't seem like that's happening as part of the phone move to 16.04 15:25 <barry> doko: we probably *won't* do a 3.6 transition this cycle i'm thinking 15:25 <slangasek> (could be done with updates to their packages in the overlay ppa, but I don't think it's going to fit in scope) 15:25 <pitti> this wouldn't be a released feature for now, and I don't know much about it, but if there's interest we can discuss this 15:25 <cyphermox> pitti: DNSSEC> should be doable to play with it some more, we already have dnsmasq on the desktop... 15:25 <doko> barry, looks a bit early 15:26 <barry> doko: let's leave it for zesty zebra 15:26 <slangasek> pitti: interesting. DNSSEC by default is a big change, so I'd hope to see a lot of eyeballs on this 15:26 <slangasek> zippy zebra 15:26 <slangasek> next cycle is going to be focused on benchmarks 15:26 <slangasek> ;) 15:26 <pitti> ooh, bootspeed again? :) 15:26 <infinity> Ahh, bootspeed, the most useless benchmark ever. 15:27 <slangasek> bootspeed matters for client UX 15:27 <infinity> "My computer was up for 3 weeks, but thank god you saved me 5 seconds on reboot with the last kernel upgrade." 15:28 <slangasek> and though we haven't paid attention to it on the desktop, it has mattered on the phone 15:28 <infinity> slangasek: Oh, I understand all sorts of reasons why it's "important", but I still have opinions. ;) 15:28 <cyphermox> what about reviewing what logs, how verbosely, etc. ? 15:28 <slangasek> it's a "benchmark" in the same sense that mouse cursor responsiveness is a "benchmark" 15:28 <pitti> also, booting containers, VMs etc. is actually quite common 15:28 <slangasek> cyphermox: are you suggesting a UOS session? 15:29 <cyphermox> slangasek: that too 15:29 <cyphermox> also, crap, I think I just volunteered :) 15:29 <doko> speed matters? lets turn off pie again 15:30 <infinity> Pie also matters. Because I'm hungry. 15:30 <pitti> no, pie is good! 15:30 <pitti> turn off the broccoli instead 15:30 <doko> no pie is good? 15:30 <pitti> and commas matter too, FWIW :) 15:30 <slangasek> right then ;) 15:30 <slangasek> #endmeeting