[15:00] <bdmurray> o/
[15:00] <dbungert> o/
[15:00] <schopin> o/
[15:00] <juliank> o/
[15:00] <waveform> o/
[15:00] <slyon> o/
[15:00] <ginggs> o/
[15:00] <bdmurray> #startmeeting Weekly Ubuntu Foundations team
[15:00] <meetingology> Meeting started at 15:00:47 UTC.  The chair is bdmurray.  Information about MeetBot at https://wiki.ubuntu.com/meetingology
[15:00] <meetingology> Available commands: action, commands, idea, info, link, nick
[15:00] <bdmurray> #topic Lightning Round
[15:01] <bdmurray> echo $(shuf -e vorlon bdmurray doko sil2100 mwhudson juliank waveform slyon jawn-smith dbungert xypron ginggs schopin alexghiti)
[15:01] <bdmurray> waveform juliank schopin mwhudson doko bdmurray xypron dbungert slyon sil2100 vorlon ginggs alexghiti jawn-smith
[15:01] <waveform> * Verified fix for impish USB boot failures on various models (LP: #1943097)
[15:01] <waveform> * Finished ISO testing generator pi (and/or other) test cases (https://code.launchpad.net/~waveform/ubuntu-manual-tests/+git/ubuntu-manual-tests/+merge/408427)
[15:01] <waveform> * Still poking about various sense hat packages
[15:01] <waveform> * Investigated Core18 ALSA issue reported by cert team (probable non-issue; discovered newer Pi400 boards fail ethernet in the process, but Pi400 not supported on Core 18 anyway)
[15:01] <waveform> * Investigating plymouth failure on Pi server images (definitely all installed in initrd, definitely running ... but not working)
[15:01] <waveform> * Verifying fix for USB storage issues (LP: #1930629) with a variety of SATA-USB interfaces on several Pis
[15:01] <waveform> * Pi meetings
[15:01] <waveform> (done)
[15:02] <bdmurray> juliank:
[15:02] <juliank> * sponsored openssl 1.1.1l-1ubuntu1
[15:02] <juliank> * fixed the grub2-signed versioned deps, and uploaded grub2-unsigned to match recent grub2 upload
[15:02] <juliank> * uploaded the riscv64 changes for sbsigntool, gnu-efi, cpu-checker to Debian unstable
[15:02] <juliank> * investigation into, and fixing of, autopkgtest systemd upstream CI
[15:02] <juliank> * grub2/bionic wanted initrdfail fixes, but does not even support initrdless, backporting feature too
[15:02] <juliank> (done)
[15:02] <schopin> * openssl:
[15:02] <schopin>   + look some more into transition to v3, looking for tracking issues in the upstream projects that FTBFS
[15:02] <schopin>   + autopkgtest regressions for 1.1.1l-1ubuntu1:
[15:02] <schopin>     - mosquitto: cannot reproduce locally, but retrigger didn't work?
[15:02] <schopin>     - puma, asterisk, kopanocore: retriggered, all OK
[15:02] <schopin>     - sysdig (arm64): seems to be a build problem with kernel headers, retrying with linux-meta-aws as additional trigger worked
[15:02] <schopin> * netplan:
[15:02] <schopin>   + Housekeeping:
[15:02] <schopin>     - close an old PR as superseded by a later contribution
[15:02] <schopin>     - close PR #8 regarding partial application of invalid configuration as WONTFIX
[15:02] <schopin>   + Looked into LP: #1942930 and formulated a plan...
[15:02] <schopin>   + ... that necessitates new C APIs, prompting me to work on the whole libnetplan API clarification
[15:02] <schopin>     PR which can be found there: https://github.com/canonical/netplan/pull/227
[15:02] <schopin>   + Looked into using linker scripts to be able to evolve our APIs without breaking the current ABI too much
[15:02] <schopin> (done)
[15:03] <bdmurray> doko:
[15:03] <doko> - OpenJDK 17 final release, SRUs to hirsute, focal and bionic
[15:03] <doko> - c-t-b build mess ... Still cannot reproduces the issue on the buildds locally
[15:03] <doko> - fixing some ftbfs seen with no-change uploads
[15:03] <doko> - libffi update. It's NOT a transition, and doesn't have new features, just getting rid off the insane snapshot introduced during my vacations last year.
[15:03] <doko> - work around two ftbfs on ppc64el
[15:03] <doko> - some haskell updates to get packages installable on riscv64 again
[15:03] <doko> (done)
[15:04] <bdmurray> testing of Error Tracker in PS5
[15:04] <bdmurray> 360s
[15:04] <bdmurray> sponsored m4 (II), scilab (BB, FF) for schopin
[15:04] <bdmurray> sponsored plymouth fixing LP: #1942987 for duflu for Impish
[15:04] <bdmurray> special SRU review of libreoffice for HH
[15:04] <bdmurray> review of 21.10 bugs in JIRA
[15:04] <bdmurray> research into command-not-found data for riscv64 and Focal
[15:04] <bdmurray> wrote some launchpadlib code to check package sizes in manifest files
[15:04] <bdmurray> spent some time looking into why core files might be truncated on the Live CD
[15:04] <bdmurray> 18.04.6 point release work
[15:04] <bdmurray> 21.10 Beta work
[15:04] <bdmurray> ✔ done
[15:04] <bdmurray> xypron is out
[15:04] <dbungert> * subiquity - add CLA check
[15:04] <bdmurray> dbungert:
[15:04] <dbungert> * subiquity/curtin -  apt components handling - PRs for python-apt usage and
[15:04] <dbungert> component adjustment pending review
[15:04] <dbungert> * subiquity/u-d-i - Disable autoinstall.  Add ability to Subiquity on command
[15:04] <dbungert> line of server to explicitly set autoinstall to an empty string, and in this
[15:04] <dbungert> scenario it should not load autoinstall data from the explicit file or
[15:04] <dbungert> cloud-init.  Add matching command line arg to u-d-i.
[15:04] <dbungert> * subiquity/u-d-i - for a time, the curtin version that subiquity wanted was
[15:04] <dbungert> different from what u-d-i was actually building into the snap.  Add a CI test
[15:04] <dbungert> to enforce.  Also send a PR to help automate that process of updating
[15:04] <dbungert> subiquity and updating the snapcraft file to match.
[15:04] <dbungert> * subiquity/u-d-i - Design updates for the Storage API, and begin the V2
[15:04] <dbungert> version that doesn't have to converse in curtin actions.
[15:04] <dbungert> * squashfs-tools vs things - squashfs-tools had a security upload, which
[15:04] <dbungert> retriggered some tests.  Casper failed on a known flaky test LP: #1938325,
[15:04] <dbungert> which was cleared for now by test retries.  On LP: #1943008 it was pointed
[15:04] <dbungert> out that this depend on libgcc-s1 should be made explicit, which I agree with
[15:04] <dbungert> and have uploaded a debdiff which is pending sponsorship.  To get this
[15:04] <dbungert> cluster to migrate requires a snapd with cgroupsv2 (fixed in git)
[15:04] <dbungert> and systemd/248.3-1ubuntu7.
[15:04] <dbungert> (done)
[15:05] <slyon> - joining MIR team & some discussions (wiki updates, rustc process, ..)
[15:05] <slyon> - handling adsys MIR review, LP: #1936907
[15:05] <slyon> - systemd cgroupsv2, snapd landed the support upstream \o/, lxc: bug #1943704
[15:05] <slyon> - integrated/staged proper fix for systemd-oomd (LP: #1942113)
[15:05] <slyon> - investigating potentiall systemd-resolved regression (LP: #1934221)
[15:05] <slyon> [15:05] <slyon> - some roadmap grooming
[15:05] <slyon> (done)
[15:05] <bdmurray> sil_2100 is away at the moment
[15:05] <bdmurray> vorlon:
[15:06] <vorlon>  * uploaded pam for LP: #1860826; focal to follow after it migrates
[15:06] <vorlon>    * merge of pam 1.4.0 deferred until JJ
[15:06] <vorlon>  * looked at internal kpis, found that my data source had disappeared, working with IS
[15:06] <vorlon>  * uploaded gnome-settings-daemon for LP: #1943768, per discussion on ubuntu-devel
[15:06] <vorlon>  * reviewing the state of libffi in -proposed, as it was noted to now be blocking a number of packages
[15:06] <vorlon>  * sorting of various issues for the 18.04.6 point release
[15:06] <vorlon>    * remembering seeds need updated for the kernel udeb ABI bump
[15:06] <vorlon> (done)
[15:06] <bdmurray> ginggs:
[15:06] <ginggs> * looked at the autopkgtest failures in boost 1.74.0-8ubuntu3 now that it is testing the 1.74 packages instead of the ones from 1.71
[15:06] <ginggs> * we've decided to wait until mdadm 4.2 release for focal, but i've been looking at the packaging changes for a backport
[15:06] <ginggs> * worked on fuse3 for s390-tools LP: #1935666, updated the PR
[15:06] <ginggs> * investigating the impact of merging or removing tasksel LP: #1929807
[15:07] <ginggs> (done)
[15:07] <alexghiti> Riscv64 packages:
[15:07] <alexghiti> https://bugs.launchpad.net/ubuntu/+source/pocl/+bug/1942895 => actual problem fixed and PR pending to solve the described problem (https://github.com/pocl/pocl/pull/988)
[15:07] <alexghiti> https://bugs.launchpad.net/ubuntu/+source/sasmodels/+bug/1943152 and https://bugs.launchpad.net/ubuntu/+source/silx/+bug/1943149 should be unblocked with pocl
[15:07] <alexghiti> https://bugs.launchpad.net/ubuntu/+source/nx-libs/+bug/1943129 working patch but wait for PR to validate it (https://github.com/ArcticaProject/nx-libs/pull/1023)
[15:07] <alexghiti> https://bugs.launchpad.net/ubuntu/+source/nvidia-settings/+bug/1943030 working patch but discussion ongoing about its “requirement”
[15:07] <alexghiti> https://bugs.launchpad.net/ubuntu/+source/rust-nodrop-union/+bug/1943148 fails on all arches and deprecated -> quid?
[15:07] <alexghiti> https://launchpad.net/ubuntu/+source/tachyon/0.99~b6+dsx-9ubuntu1/+build/22072848 tachyon packages now available on riscv64, bug report to debian ongoing.
[15:07] <alexghiti> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=994390 : fails to install because depends on libreoffice-core, patch pending to add build dependency instead.
[15:07] <alexghiti> Ubuntu-meta (ubuntu-desktop[-minimal]): PR ongoing
[15:07] <alexghiti> Sifive Unmatched board
[15:07] <alexghiti> Patch linux-riscv to have functional reset: working, needs cleanup before taking this and submit upstream
[15:07] <alexghiti> Patch u-boot to have functional reset: more or less working, debug ongoing.
[15:07] <alexghiti> (done)
[15:07] <bdmurray> jawn-smith:
[15:08] <jawn-smith> * Openssh failure digging
[15:08] <jawn-smith>     * Narrowed down to an issue with host running kernel <= 5.8 while container runs impish (5.13)
[15:08] <jawn-smith>     * Still determining if it's a test environment issue or actual bug in openssh
[15:08] <jawn-smith> * Re-created LP: #1943097 and did a little troubleshooting before handing off to kernel team
[15:08] <jawn-smith> * Go 1.17 baseline retesting
[15:08] <jawn-smith>     * 19 packages fail with Go 1.17 but not Go 1.16
[15:08] <jawn-smith>     * Created FFE request. LP: #1943502
[15:08] <jawn-smith> * +1 Maintenance
[15:08] <jawn-smith>     * gitlab-workhorse: uploaded patch to fix FTBFS with Go 1.17, but it failed to build on 1.16. oops.
[15:08] <jawn-smith>     * golang-github-biogo-hts: patch uploaded for Go 1.17
[15:08] <jawn-smith>     * golang-github-gin-gonic-gin: patch uploaded for Go 1.17
[15:08] <jawn-smith>     * golang-github-sevylar-go-daemon: patch uploaded for Go 1.17
[15:08] <jawn-smith>     * textql: patch uploaded for Go 1.17
[15:08] <jawn-smith>     * golang-github-marten-seemann-qtls-go1-15: appears to be a leaf that provides no binaries
[15:08] <jawn-smith>         * Should be removed, but I need to determine if it is also a leaf in Debian first
[15:08] <jawn-smith> [done]
[15:09] <bdmurray> any questions on status?
[15:10] <bdmurray> #topic Release Incoming Bugs
[15:10] <bdmurray> #link http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-ii-incoming-bug-tasks.html#foundations-bugs
[15:11] <bdmurray> bug 1871538 is already carded I'll clean it up after
[15:11] <slyon> cleaned
[15:12] <bdmurray> I gather bug 1943288 isn't a bug per se
[15:12] <doko> the bug title is bogus. no soname change, no transition
[15:12] <bdmurray> bug 1943189 - juliank I think you'd looked at this once
[15:13] <vorlon> but behavior changes that are breaking GNOME revdeps, evidently
[15:13] <juliank> bdmurray: yeah, apparently we want to take it
[15:13] <vorlon> doko: you said you're working on it, do you undersand the GNOME regressions at this point?
[15:13] <doko> yes, it was one wrong config
[15:13] <vorlon> ok
[15:14] <doko> there are no regressions anymore. but am I now supposed to fix any accumulated glib regressions?
[15:14] <bdmurray> juliank: do you have time to sponsor it or should someone else do that?
[15:14] <juliank> bdmurray: I can do that
[15:15] <vorlon> doko: what glib regressions are you talking about?
[15:15] <bdmurray> juliank: thanks
[15:15] <doko> looking at https://autopkgtest.ubuntu.com/packages/oss4/impish/arm64
[15:15] <vorlon> doko: it looks like we should be able to unblock this; given the severity of LP: #1943362, it would be good to be sure that your fix solves that problem as well
[15:15] <doko> https://autopkgtest.ubuntu.com/packages/mutter/impish/armhf
[15:15] <doko> https://autopkgtest.ubuntu.com/packages/netplan.io/impish/armhf
[15:16] <vorlon> doko: well if you want this package to migrate, you'll need to do something?  But I'm unclear why no-change rebuilds were needed anyway
[15:16] <slyon> that netplan armhf/LXD test needs a systemd trigger, to use the version from -proposed, which contains a relevant systemd-networkd fix
[15:18] <slyon> (triggered)
[15:19] <bdmurray> I'm going to punt bug 1940011 to JJ
[15:19] <doko> ok, I'll add that one for netplan
[15:20] <bdmurray> bug 1939413
[15:20] <bdmurray> Is anybody familiar with this or should we look at it next week?
[15:21] <bdmurray> next week then
[15:21] <vorlon> skimming the bug it looks like it might be sorted
[15:22] <bdmurray> bug 1942603 already has a card so I've removed the tag
[15:22] <bdmurray> it looks like bug 1943109 is already fix committed
[15:23] <bdmurray> so let's drop the tag
[15:23] <bdmurray> #link https://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-hh-incoming-bug-tasks.html
[15:23] <bdmurray> that's empty
[15:23] <bdmurray> #link http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-ff-incoming-bug-tasks.html
[15:24] <bdmurray> this is really about linux-firmware not initramfs-tools correct?
[15:24] <schopin> looks like it, yes.
[15:24] <bdmurray> bug 1939986
[15:25] <bdmurray> okay let's mark the initramfs-tools as invalid then
[15:25] <vorlon> if someone's confident in that could they mark it invalid for initramfs-tools?
[15:25]  * vorlon nods
[15:26] <bdmurray> #topic Team proposed-migration report
[15:26] <bdmurray> https://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses_by_team.html#foundations-bugs
[15:26] <bdmurray> vorlon:
[15:27] <bdmurray> I think jawn-smith talked about openssh already
[15:27] <jawn-smith> I did
[15:28] <slyon> I'm still working on the systemd outfall. snapd landed support upstream recently and I'm talking to LXD team currently about a lxc fix
[15:28] <vorlon> and squashfs-tools has an open bug describing the issues there; dbungert are you driving this?
[15:28] <dbungert> vorlon: yes
[15:28] <vorlon> (possibly blocked on snapd team?)
[15:28] <vorlon> slyon: thanks
[15:28] <schopin> I'm working on openssl, mosquitto is the only remaining failure.
[15:28] <vorlon> grub2 hmm why's that still stuck
[15:29] <dbungert> yes, I'm not sure that we can do much until new a snapd version is available
[15:29] <vorlon> juliank: grub2?
[15:29] <vorlon> wait do I need to refresh
[15:29] <vorlon> yes I need to refresh
[15:29] <vorlon> sorry
[15:29] <vorlon> not grub2
[15:29]  * juliank confused
[15:29] <vorlon> openssl
[15:29] <juliank> openssl is schopin's upload which I sponsored
[15:30] <vorlon> schopin: openssl is your upload, do you need any help sorting the regressions or are you on top of this?
[15:30] <schopin> mosquitto is the only holdout, still hoping to get through it :)
[15:30] <vorlon> thanks
[15:31] <vorlon> there's a sphinx autopkgtest regression blocking webkit2gtk, but it looks likely this is actually a webkit2gtk regression on armhf, so passing
[15:32] <vorlon> aptdaemon autopkgtest failure blocking pygobject, which is doko's upload; doko, are you going to drive sorting the autopkgtest regressions on these?
[15:32] <vorlon> (or should we back the uploads out of -proposed since they're not strictly needed)
[15:32] <vorlon> similar story on glib2.0
[15:32] <doko> I have been told by julian that aptdaemon just needs continuous retrying
[15:32] <vorlon> sure - are you driving this then?
[15:33] <vorlon> netplan.io/armhf, someone know the issue there?
[15:33] <doko> and it just passed
[15:33] <vorlon> ok
[15:34] <doko> we just talked about netplan, slyon retried
[15:34] <slyon> already re-triggerd the netplan.io/armhf test
[15:34] <vorlon> alrighty
[15:34] <slyon> (with systemd/248.3-1ubuntu7 trigger)
[15:34] <vorlon> then there are some packages just waiting for libffi to migrate, libffi should become a candidate soon as long as we're sure it's not breaking people's gdm and can close the blocking bug
[15:34] <vorlon> gcc-11-cross ftbfs; doko?
[15:35] <doko> yes, that's the c-t-b mess. on it
[15:35] <vorlon> ok
[15:35] <vorlon> then the rest are newer and either SEP (curl - security) or known issues or in progress
[15:35] <vorlon> bdmurray:
[15:35] <dbungert> I wonder if that ubiquity vs libffi is a real regression:  test_interface_translated (test_gtkui.TestFrontend) ... Segmentation fault (core dumped)
[15:36] <bdmurray> How could we check?
[15:36] <vorlon> dbungert: could be - the test happened with the older libffi
[15:36] <vorlon> retry with the newer?
[15:36] <doko> dbungert, it passed an hour ago
[15:37] <bdmurray> okay then
[15:37] <bdmurray> #topic AOB
[15:37] <jawn-smith> bdmurray, vorlon: I know we discussed it a bit yesterday, but is anything else needed for LP: #1943502 ?
[15:38] <vorlon> jawn-smith: I approved the FFe, so something needs to be uploaded?
[15:38]  * juliank is off next week: Mon & Fri, moving Sat
[15:38] <bdmurray> I can sponsor that
[15:39] <jawn-smith> We can see if mwhudson has opinions first
[15:39] <bdmurray> Just a friendly reminder that the Beta is next week
[15:41] <bdmurray> okay, thanks everybody
[15:41] <vorlon> thanks!
[15:41] <bdmurray> #endmeeting
[15:41] <meetingology> Meeting ended at 15:41:33 UTC.  Minutes at https://new.ubottu.com/meetingology/logs/ubuntu-meeting/2021/ubuntu-meeting.2021-09-16-15.00.moin.txt
[15:41] <slyon> o/
[15:41] <schopin> o/
[15:41] <ginggs> o/
[20:14] <mwhudson> jawn-smith, bdmurray: happy for someone else to jfdi
[20:15] <bdmurray> mwhudson: we decided to wait b/c of a riscv64 change which made it into 1.16 but not 1.17 yet
[20:18] <mwhudson> oh that
[20:18] <mwhudson> yes ok i should sponsor that too :)