15:01 <slangasek> #startmeeting
15:01 <meetingology> Meeting started Thu Oct  5 15:01:26 2017 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
15:01 <slangasek> [TOPIC] Lightning round
15:01 <slangasek> $ echo $(shuf -e slangasek bdmurray xnox tdaitx doko sil2100 rbalint infinity cyphermox mwhudson rcj philroche Odd_Bloke tribaal fginther)
15:01 <slangasek> fginther bdmurray rcj xnox tribaal rbalint tdaitx philroche doko infinity cyphermox Odd_Bloke slangasek mwhudson sil2100
15:01 <rbalint> o/
15:01 <sil2100> o/
15:02 <sil2100> Wow, dead-last
15:02 <fginther> goodness, one moment
15:03 <fginther> * Continued library development to facilitate automated publication
15:03 <fginther> * provided an automation script to IS to improve deployment of cloud-image jenkaas slaves
15:04 <fginther> * Enabled launchpad <-> trello automation
15:04 <fginther> (done)
15:04 <bdmurray> Research into LP: #1515513 re: dkms leftovers in /boot
15:04 <bdmurray> Uploaded X and Z SRUs for LP: #1515513
15:04 <bdmurray> Tested LP: #1720331 re whoopsie restarting constantly
15:04 <bdmurray> Investigation into gdebi python3.6 issue LP: #1708947
15:04 <ubottu> Launchpad bug 1515513 in initramfs-tools (Ubuntu) "/boot/initrd.img-*.old-dkms files left behind" [Undecided,Confirmed] https://launchpad.net/bugs/1515513
15:04 <ubottu> Launchpad bug 1720331 in whoopsie (Ubuntu) "Whoopsie continually relaunching" [Medium,Triaged] https://launchpad.net/bugs/1720331
15:04 <ubottu> Launchpad bug 1708947 in gdebi (Ubuntu) "gdebi-kde crashed with ModuleNotFoundError in /usr/lib/python3/dist-packages/PyKDE4/__init__.py: No module named 'DLFCN'" [Medium,Triaged] https://launchpad.net/bugs/1708947
15:04 <bdmurray> HIGHLIGHT(?): Wrote blog post re free-ing space in /boot - http://www.murraytwins.com/blog/?p=156
15:04 <bdmurray> reported LP: #1721364 re apt autoremove and Never-MarkAuto-Sections
15:04 <ubottu> Launchpad bug 1721364 in apt (Ubuntu) "Never-MarkAuto-Sections not working" [High,Opinion] https://launchpad.net/bugs/1721364
15:04 <bdmurray> tested LP: #1715490 again, looked for similar bugs
15:05 <ubottu> Launchpad bug 1715490 in plymouth (Ubuntu) "pretty version of plymouth won't accept passphrase" [High,Incomplete] https://launchpad.net/bugs/1715490
15:05 <Tribaal> SCHWEET
15:05 <bdmurray> ✔ done
15:05 <bdmurray> SCHWEET?
15:05 <Tribaal> Swiss-German for "Sweet"
15:06 <Tribaal> JIT meeting :)
15:06 <slangasek> skip rcj for the moment
15:06 <rcj> * sprinted
15:06 <slangasek> heh
15:06 <rcj> * got sick after sprint
15:06 <Odd_Bloke> * failed to read lists of people
15:06 <rcj> * automated cloud-image build work
15:06 <rcj> done
15:07 <xnox> artful ntp vs systemd: systemd is in -proposed, ntp is the unapproved queue - please review
15:07 <xnox> zesty: systemd failed verification, fixup and more fixes are in the unapproved queue - please review
15:07 <xnox> xenial: systemd passed verification, there are adt regressions. Filed bugs for linux-* and open-iscsi, these appear to be broken tests. nplan seems to fail for a testcase in NetworkManager, and nplan/autostart test case appears to have regressed under lxc on xenial. However, creating autopkgtest lxc container does not work for me on artful
15:07 <bdmurray> xnox:
15:07 <xnox> openssl-ibmca libica et.al. landed thus we have full accelerated z14 encryption landed
15:07 <xnox> mdadm/systemd shutdown - still in progress
15:07 <xnox> fixed s390-dasd for final beta (failing to format DASD drives)
15:07 <xnox> zfcpdump-kernel - failing to upgrade for GA, to be escalated to IBM
15:07 <xnox> cloud-init leases merge in progress (started, and handed over to cloud-init team)
15:07 <xnox> ...
15:07 <slangasek> Tribaal:
15:07 <Tribaal> * Some of the memory issues investigated previously were due to transparent huge pages being set to "enabled". That led to memory fragmentation.
15:07 <Tribaal> * Vaguard duties for the build system this week.
15:07 <Tribaal> * Helping partners kick the tires on the final 17.10 beta to make sure everything is ok for the final release.
15:07 <Tribaal> * OoO tomorrow and beginning of next week
15:07 <Tribaal> done.
15:07 <rbalint> * GCE new upstream LP: #1718922
15:07 <ubottu> Launchpad bug 1718922 in gce-compute-image-packages (Ubuntu Zesty) "Update google compute-image-packages to 20170921 " [Undecided,New] https://launchpad.net/bugs/1718922
15:07 <rbalint> * UBuntu Rally
15:07 <rbalint> * image minimization
15:07 <rbalint> * updating runc to 1.0.0~rc4 and creating docker-runc with the version needed by docker.io in Debian
15:07 <rbalint> * unattended-upgrades 0.98 for Debian
15:07 <rbalint> (done)
15:08 <tdaitx> * Trying to figure out OpenJDK 8 build failure in i386 (error: "unsupported size for integer register")
15:08 <tdaitx> * +1 maint
15:08 <tdaitx> * Checked and (re)setup jenkaas slaves for jck after the lgw01 scalingstack redeploy (zeroed slaves)
15:08 <tdaitx> (done)
15:08 <philroche> - Cloud image build system vanguard
15:08 <philroche> - Ubuntu rally (via hangouts)
15:08 <philroche> - Cloud image build system development
15:09 <philroche> - Cloud partner calls
15:09 <philroche> (done)
15:09 <philroche> doko:
15:10 <bdmurray> no infinity either
15:11 <bdmurray> cyphermox:
15:11 <cyphermox> yay
15:11 <cyphermox> shim:
15:11 <cyphermox> - preparing shim for signing
15:11 <cyphermox> - Microsoft and shim-review submissions
15:11 <cyphermox> artful:
15:11 <cyphermox> - unbroke keyboard selection/translations in ubiquity due to console-setup
15:11 <cyphermox> - reword partman-efi non_efi_system template again
15:11 <cyphermox> - testing subiquity image
15:11 <cyphermox> SRUs:
15:11 <cyphermox> - [xenial] preparing ubiquity SRU for WPA2 Enterprise support (LP: #1107935)
15:11 <cyphermox> - [xenial] investigating search domains issue in klibc.
15:11 <ubottu> Launchpad bug 1107935 in ubiquity (Ubuntu Xenial) "Support for WPA Enterprise wireless networks" [Undecided,In progress] https://launchpad.net/bugs/1107935
15:11 <cyphermox> - [zesty] isc-dhcp DNS search domains handling
15:11 <cyphermox> other:
15:11 <cyphermox> - catching up on cloud image minimization work
15:11 <cyphermox> - netplan 0.29 backport to 16.04 is waiting in the unapproved queue
15:11 <cyphermox> (done)
15:11 <Odd_Bloke> * Ubuntu Rally last week
15:11 <Odd_Bloke> * Finalising new GCE publication flow in preparation for bb opening
15:11 <Odd_Bloke> * Preparing for product sprint next week
15:12 <Odd_Bloke> (done)
15:12 <Odd_Bloke> slangasek:
15:12 <slangasek> * at the Ubuntu Rally last week in New York, lots of great discussions and development work
15:12 <slangasek> * first in-person meetings for a while with quorum of the archive, SRU, and release teams
15:12 <slangasek> * worked with the foundations team to fix papercuts around netplan, systemd-network migration
15:12 <slangasek> * worked with the kernel team to design solution for extra initramfs generation on package removal
15:13 <slangasek> * HIGHLIGHT (if sil2100 doesn't do it) Artful Final Beta released last week
15:13 <slangasek> * working on livecd-rootfs changes for artful and xenial:
15:13 <slangasek> * landing support for preinstallation of snaps in images
15:13 <slangasek> * adding support for minimizing cloud images
15:13 <slangasek> * next week, at the Product Roadmap Sprint, again in NYC
15:13 <slangasek> (done)
15:13 <slangasek> sil2100:
15:13 <sil2100> - Sprinting, then swap day, then partially away due to sickness (not really productive yet)
15:13 <sil2100> - Setting up e-mail for release-bits
15:13 <sil2100> - SRU reviews, cleanup and releases
15:13 <sil2100> - ubuntu-image work on coverage, refactoring the builder classes
15:13 <sil2100> - Reading up on preparations for final release
15:13 <sil2100> - Last week: Final Beta :O
15:13 <sil2100> Ah, indeed, that was a good highlight!
15:13 <sil2100> (done)
15:14 <slangasek> any questions on status?
15:14 <bdmurray> Nope
15:15 <slangasek> [TOPIC] Bugs
15:15 <slangasek> probably late in the cycle to be committing to anything non-critical for the 17.10 release :)
15:15 <slangasek> [LINK] http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-aa-incoming-bug-tasks.html#foundations-bugs
15:15 <slangasek> there are a couple of 'high' bugs here
15:15 <bdmurray> I mentioned bug 1721364 but would like to discuss it
15:15 <ubottu> bug 1721364 in apt (Ubuntu) "Never-MarkAuto-Sections not working" [High,Opinion] https://launchpad.net/bugs/1721364
15:16 <slangasek> bdmurray: fwiw I have it in my inbox to review the commit jak pointed to
15:16 <slangasek> bdmurray: but I guess we should discuss what the path forward is if that doesn't get reverted?
15:17 <bdmurray> slangasek: Yeah, autoremove could have surprising results.
15:17 <slangasek> frustrating
15:17 <slangasek> bdmurray: so your position is that it's too high-risk?
15:18 <bdmurray> Well, I'm guessing people remove metapackages like ubuntu-desktop but could be wrong.
15:18 <slangasek> I don't know; they might because they wanted to remove one of the packages it pulled in
15:19 <bdmurray> I mean if they've used autoremove in recent history then they would have hit an issue.
15:19 <bdmurray> s/hit an issue/ran into this/
15:20 <slangasek> bdmurray: yeah.  I think we should revisit with upstream the decision to change how markauto is handled, but that won't change the fact that systems already installed will have their package database in a state that makes autoremove dangerous
15:21 <slangasek> bdmurray: which I guess leaves only the remove-newly-autoremovable path, which rbalint has mentioned?
15:22 <bdmurray> slangasek: Yeah, I think so
15:22 <slangasek> ok
15:22 <slangasek> annoying, that's going to take more work to implement
15:23 <slangasek> but +1 for this on the path forward
15:23 <bdmurray> Well the newly-autoremovable only works if you are install the kernel update with u-u
15:23 <slangasek> couldn't we make it work with update-manager?
15:24 <slangasek> by implementing the same logic as in u-u
15:24 <bdmurray> Oh, I see
15:24 <slangasek> this isn't something we need to magically solve for users running apt from the commandline
15:24 <slangasek> they will run apt autoremove or not
15:25 <slangasek> bdmurray: agreed?
15:25 <bdmurray> slangasek: Yes
15:25 <slangasek> ok
15:25 <slangasek> anything else on bugs?
15:25 <slangasek> we have two 'high' in incoming
15:25 <bdmurray> bug 1720331 - I'm not sure if its whoopsie or the control panel
15:25 <ubottu> bug 1720331 in whoopsie (Ubuntu) "Whoopsie continually relaunching" [Medium,Triaged] https://launchpad.net/bugs/1720331
15:26 <rbalint> slangasek, bdmurray: the remove-newly-autoremovable path is already released in u-u which is why i'm hesitant to revert it, but special handling for kernel packages seem to be a safer bet
15:27 <bdmurray> rbalint: We weren't talking about reverting it but using the same code in update-manager
15:29 <bdmurray> cyphermox: Is not being able to use 'videoinfo' in grub relevant with bug 1715490?
15:29 <ubottu> bug 1715490 in plymouth (Ubuntu) "pretty version of plymouth won't accept passphrase" [High,Incomplete] https://launchpad.net/bugs/1715490
15:29 <slangasek> bdmurray: AIUI videoinfo is specific to VESA, and if you're using the qemu vga driver you don't even have VESA
15:29 <cyphermox> bdmurray: I don't think so
15:30 <slangasek> so you don't have videoinfo because it's literally not configurable
15:30 <rbalint> bdmurray: yes, i mentioned that because https://launchpad.net/linux-purge is a valid safer alternative imo
15:30 <bdmurray> Okay, well I couldn't find any more bugs like 1715490 so lets punt it then
15:30 <rbalint> bdmurray: and i think it is worth a separate discussion
15:30 <cyphermox> sounds like it's "just" broken video in VM, "as usual"
15:31 <sil2100> heh
15:31 <slangasek> ok; anything else on bugs?
15:31 <cyphermox> oh, also
15:32 <cyphermox> bdmurray: you on artful and running in wayland?
15:32 <bdmurray> cyphermox: the host or the guest?
15:32 <cyphermox> because that breaks arrow keys, so it wouldn't be overly surprising if it broke video too somehow
15:32 <rbalint> bdmurray: but adding remove-newly-autoremovable option to update-manager is also valuable for the sake of consistency IMO, it may or may not be the default
15:32 <cyphermox> on the host
15:34 <slangasek> rbalint: this is only interesting to do if we make it the default
15:34 <bdmurray> cyphermox: not wayland
15:34 <cyphermox> aok
15:34 <bdmurray> I did get the arrow key issue though
15:34 <cyphermox> ok
15:35 <rbalint> slangasek: ok
15:37 <slangasek> [TOPIC] AOB
15:37 <slangasek> anything else?
15:38 <bdmurray> nope
15:38 <slangasek> #endmeeting