15:01 <bdmurray> #startmeeting Weekly Ubuntu Foundations team 15:01 <meetingology> Meeting started Thu Oct 31 15:01:45 2019 UTC. The chair is bdmurray. 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 <bdmurray> #topic Lightning rounds 15:02 <bdmurray> bdmurray sil2100 cyphermox mwhudson tdaitx doko juliank xnox vorlon infinity rbalint waveform 15:02 <bdmurray> updated daisy retracer code for more python-cassandra changes 15:02 <bdmurray> updated daisy retracer code to support retracing of focal 15:02 <bdmurray> discovered issue with Conents.gz being 0 bytes for eoan release pocket 15:02 <bdmurray> updated lp retracer config to support retracing of focal 15:02 <bdmurray> created apport test crashes for focal 15:02 <bdmurray> updated retracer cronjob to count users of 20.04 15:02 <bdmurray> investigation into when 'apt list --upgradable' is displayed 15:02 <bdmurray> reviewed MP from sil2100 re ubuntu-release-upgrader 15:03 <xnox> bdmurray: contents will be fixed, right? 15:03 <bdmurray> ✔ done 15:03 <cyphermox> - netplan NetworkManager plugin 15:03 <cyphermox> - coordinating net-tools deprecation 15:03 <cyphermox> - debugging grub2 "unknown tpm" issue (patch ready, to SRU today) 15:03 <bdmurray> xnox: Yes, contents is fixed already and permanent fix is landing 15:03 <cyphermox> - reiwa support patched 15:03 <cyphermox> (done) 15:03 <cyphermox> *patches 15:03 <bdmurray> tdaitx: 15:04 <bdmurray> I think he said something about lunch 15:04 <bdmurray> doko: ? 15:04 <doko> - working on libreoffice autopkg test failure, without 15:04 <doko> support from the Debian/Ubuntu package maintainters. 15:04 <doko> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92267 15:04 <doko> Can't wait for the comments that I broke the archive ... 15:04 <doko> - GCC 9 update 15:04 <bdmurray> how about we skip to xnox! 15:04 <doko> - bug reports for Python3 packages 15:04 <ubottu> gcc.gnu.org bug 92267 in libstdc++ "[9 Regression] crash with a cppunit test case (built by GCC 9) and cpptest (built with GCC 8)" [Normal,Assigned] 15:04 <doko> (done) 15:05 <xnox> Work with mwhudson on klibc bug #1850184 15:05 <ubottu> bug 1850184 in linux (Ubuntu Focal) "losetup -f broken in 2.0.6-1ubuntu2" [Undecided,Incomplete] https://launchpad.net/bugs/1850184 15:05 <xnox> Attended OIN office hours in lyon (private summary to follow) 15:05 <xnox> Created merge proposal for :i386 resolution in germinate, it now produces matching output as the static EOAN list, but fails testsuite in other ways 15:05 <xnox> Customer meetings prep 15:05 <xnox> Roadmap sprint prep 15:05 <xnox> done 15:05 <bdmurray> vorlon: 15:05 <vorlon> * archive opening 15:05 <infinity> * Very Long weekend and short week. 15:05 <vorlon> * autopkgtest shepherding 15:05 <vorlon> * python2 removals (NBS, process-removals, etc) 15:05 <vorlon> * working around python3.8 build failures in packages that need update to drop python2 15:05 <infinity> * Some help with transitions. 15:05 <infinity> * Getting d-i hobbling along while we decide how to kill it. 15:05 <infinity> * Work with LP on Contents issue. 15:05 <infinity> * Misc AA/SRU stuff. 15:05 <infinity> (done) 15:05 <vorlon> * 20.04 roadmap prep 15:05 <vorlon> (done) 15:06 <rbalint> * poked systemd and other packages to migrate 15:06 <rbalint> * SRUs: systemd (eoan) 15:06 <rbalint> * SRU verifications: unattended-upgrades 15:06 <rbalint> * reported LP:#1850667 and set up ppa:rbalint/systemd-unified-cgh for testing the new upstream systemd default which is reverted in Ubuntu to avoid regressions 15:06 <rbalint> * merged systemd again from Debian, poking it to migrate 15:06 <rbalint> (done) 15:06 <bdmurray> waveform: 15:06 <waveform> * Added flash-kernel entry for Pi 3A+; LP: #1841955 15:06 <waveform> * Testing kernel fixes for Pi 4 USB issue across various pis - all looking good except for flash-kernel issue (next); LP: #1848790 15:06 <waveform> * Investigating Bionic SRU of Pi 4 support 15:06 <waveform> * Noticed flash-kernel only updates dtb for platform it is being run on. Unfortunately, there is a baked-in assumption that it will only ever deal with one dtb (while Pi requires it to handle all dtbs, and preferably overlays too). Working on a fix; LP: #1850678 15:06 <waveform> (done) 15:06 <ubottu> Launchpad bug 1841955 in flash-kernel (Ubuntu) "Can't upgrade kernel on raspberry pi3a+" [Undecided,In progress] https://launchpad.net/bugs/1841955 15:06 <ubottu> Launchpad bug 1848790 in linux-raspi2 (Ubuntu) "USB not working under arm64 on Pi4" [High,Confirmed] https://launchpad.net/bugs/1848790 15:06 <ubottu> Launchpad bug 1850678 in flash-kernel (Ubuntu) "flash-kernel only updates the dtb of the Pi it's on" [Undecided,New] https://launchpad.net/bugs/1850678 15:06 <xnox> rbalint: one can just use kernel boot option no? or is that too hard for users? 15:07 <xnox> rbalint: i.e. i don't think we will be able to move off hybrid in 20.04 15:08 <rbalint> xnox, it is fine if we don't move off, the bug is there to track readiness 15:08 <xnox> rbalint: ack 15:08 <bdmurray> #topic Release incoming bugs 15:08 <bdmurray> http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-ee-incoming-bug-tasks.html#foundations-bugs 15:09 <bdmurray> #link http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-ee-incoming-bug-tasks.html#foundations-bugs 15:09 <bdmurray> actually lets start with ff 15:09 <bdmurray> http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-ff-incoming-bug-tasks.html 15:09 <bdmurray> There's one new here bug 1849477 15:09 <ubottu> bug 1849477 in numpy (Ubuntu Focal) "numpy fails on s390x" [Undecided,Fix committed] https://launchpad.net/bugs/1849477 15:10 <bdmurray> somebody tagged it rls-ff-incoming and somebody said it was fixed but the tag was left 15:10 <xnox> bdmurray: but also it is fixed in -proposed already by mwhudson 15:10 <bdmurray> I've cleaned it up but lets try not to leave that for me 15:10 <bdmurray> xnox: bug 1847852 15:10 <ubottu> bug 1847852 in livecd-rootfs (Ubuntu) "lxd group is added to too many products" [Undecided,New] https://launchpad.net/bugs/1847852 15:11 <bdmurray> Okay, now ee 15:11 <bdmurray> http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-ee-incoming-bug-tasks.html#foundations-bugs 15:12 <bdmurray> I still need to review those but roadmap work is taking priority 15:12 <bdmurray> sorting by date created bug 1849156 is recent 15:12 <ubottu> bug 1849156 in systemd (Ubuntu) "systemd-timesyncd.service broken on upgrade to 19.10 if ntp was installed" [Undecided,Confirmed] https://launchpad.net/bugs/1849156 15:13 <xnox> https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1849156/+activity 15:13 <ubottu> Launchpad bug 1849156 in systemd (Ubuntu) "systemd-timesyncd.service broken on upgrade to 19.10 if ntp was installed" [Undecided,Confirmed] 15:14 <bdmurray> xnox: it looks like rblalint asked a question about splitting the package 15:14 <rbalint> taking it 15:15 <bdmurray> Will splitting the package fix it for Eoan? 15:15 <rbalint> bdmurray, sruing the change will 15:15 <xnox> bdmurray: yes, but upgrade testing will be fun. 15:18 <bdmurray> Well we'll do something for Focal, so I'll card this. The SRU fix sounds a bit more complicated. 15:18 <xnox> i think the key is to have it done right for Bionic to Focal upgrades 15:18 <bdmurray> bug 1848892 15:18 <ubottu> bug 1848892 in grub2 (Ubuntu) ""error: Unknown TPM error." after upgrading to grub 2.04" [High,Confirmed] https://launchpad.net/bugs/1848892 15:19 <cyphermox> I'm on top of that 15:19 <cyphermox> will upload SRU with a patch that will mitigate this today; and I think this is a good candidate for fast-tracking 15:19 <cyphermox> there's a worrying number of reports 15:19 <bdmurray> cyphermox: Do you already have a card for it? 15:20 <cyphermox> (I submitted my patch already upstream) 15:20 <cyphermox> no, there isn't AFAIK 15:20 <cyphermox> I go create card 15:20 <bdmurray> cyphermox: okay, thanks! 15:22 <cyphermox> done 15:22 <cyphermox> anyway, agree on fast-tracking> 15:22 <cyphermox> ? 15:22 <cyphermox> vorlon: ^ 15:22 <bdmurray> cyphermox: does fast-tracking mean wait less than 7 days in -proposed? 15:22 <cyphermox> yes 15:23 <cyphermox> the patch is straightforward and the effects of the bug are quite jarring 15:23 <vorlon> I'm not committing to that just yet 15:23 <cyphermox> OTOH, it's grub, and omg 15:23 <vorlon> the bug currently has no SRU template, so that's a necessary first step 15:23 <bdmurray> As a fake SRU team member I'd have to see the patch before deciding 15:24 <bdmurray> But it does seem important enough to get the fix out there 15:24 <cyphermox> bdmurray: patch is linked in my last comment on the bug 15:24 <cyphermox> I'll prepare the package changes shortly and update the bug with a template 15:25 <bdmurray> cyphermox: it looks like you have bug 1848797 in hand too, but there isn't a code change correct? 15:25 <ubottu> bug 1848797 in grub2 (Ubuntu) "After upgrading to 19.10, boot screen shows: "Error: symbol 'grub_file_filters' not found."" [High,Confirmed] https://launchpad.net/bugs/1848797 15:30 <bdmurray> There is nothing new for rls-bb 15:30 <bdmurray> #topic Team proposed-migration report 15:30 <bdmurray> #link http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses_by_team.html#foundations-bugs 15:31 <xnox> doko: do we care about python3-defaults migrating with 3.8 as supported, rather than starting the transition to 3.8 as default? 15:31 <vorlon> I think despite the volume of packages here and the known entanglements, we are at a point where it's worth asking people to divide and conquer 15:32 <xnox> doko: imho it seems odd to ever support default as lower than a supported version. 15:32 <vorlon> because they're largely things that do need attention in order to get them through, not just things that are waiting for infra 15:32 <xnox> (because there are aloways things in the chain that embed default interpreters only) 15:32 <bdmurray> The other day you (vorlon) had mentioned tracking the conquering - is that right? 15:33 <bdmurray> or maybe its really the dividing that needs tracking ;-) the conquering is obvious 15:33 <vorlon> bdmurray: uh I'm not sure; the last thing I mentioned was about python2 removals 15:33 <vorlon> which is different from this 15:33 <vorlon> anyway, looking at python3-defaults alone, which is the oldest package on there (because archive opening), most of those are going to require attention 15:34 <doko> xnox: not my priority now. my priority is to work on the numpy stack, that's the most broken thing currently 15:34 <rbalint> let's go in order and pick 15:34 <vorlon> either by fixing source compatibility with python3.8, or (temporarily) dropping support in the package for python3-all and filing bugs in Debian 15:34 <xnox> doko: ack =/ 15:34 <vorlon> do we just want to round-robin assign a couple to everyone? 15:34 <cyphermox> volunteering for python-apt, I guess 15:35 <vorlon> ok, python-apt - cyphermox 15:35 <vorlon> asciidoc - sil2100 15:35 <rbalint> probably we should all take a look at the breakages for python-defaults 15:35 <vorlon> cloudpickle - mwhudson (hi!) 15:35 <vorlon> dask - tdaitx 15:36 <vorlon> I'll skip doko because I know he's working on numpy 15:36 <rbalint> twisted -rbalint 15:36 <tdaitx> vorlon: ack 15:36 <cyphermox> notpandas notpandas notpandas notpandas... ;) 15:36 <waveform> I can take pyzmq 15:36 <tdaitx> let me know when it is ok to send my status 15:36 <xnox> jajajjajjajj 15:36 <rbalint> cron & systemd just needed a retrigger imo 15:36 <vorlon> django-auto-one-to-one - xnox (and fyi last I knew django base packages were incompatible with python3.8) 15:37 <vorlon> xnox: so you can have django-mailman3 also 15:37 <xnox> har, i can do django 15:37 <vorlon> gnudatalanguage - vorlon 15:37 <doko> there's a django transition ongoing in debian 15:37 <vorlon> gyoto - infinity 15:37 <vorlon> hyperkitty - rbalint 15:37 <vorlon> i3pystatus - waveform 15:38 <vorlon> are we ok to assign out a single round, and ask people to self-organize to take a second package from the list when they get it done? 15:38 <rbalint> +1 15:38 <doko> +2 15:38 <bdmurray> vorlon: self-organize and communicate where? 15:39 <vorlon> #ubuntu-release, please 15:39 <cyphermox> cool 15:39 <bdmurray> sounds reasonable 15:40 <bdmurray> Is that it regarding proposed-migration? 15:40 <vorlon> doko, xnox: right, there was python-django 2:2.2.4-1 that was synced in eoan and then reverted because the transition wasn't ready to go; so step one is probably to merge it 15:40 <vorlon> I think so for now 15:40 <bdmurray> #topic AOB 15:40 <rbalint> bdmurray, and use update-excuse tag which i forgot from my weeklog! 15:40 <bdmurray> but let's rewind so we can hear what tdaitx has been up to 15:40 <tdaitx> * whoopsie security update (LP: #1830865) 15:40 <tdaitx> - introduced a regression + DoS daisy server 15:40 <tdaitx> - fixed regression and coordinated a bit with IS 15:40 <tdaitx> * apport security fixes (LP: #1830862, LP: #1839795, LP: #1839413, LP: #1839415, LP: #1839420) 15:40 <tdaitx> * getting multipass serial console to work to test LP: #1831252 15:40 <tdaitx> Other: 15:40 <ubottu> Launchpad bug 1830865 in whoopsie (Ubuntu) "Integer overflow in bson_ensure_space (bson.c:613)" [High,Fix released] https://launchpad.net/bugs/1830865 15:40 <tdaitx> - need more coffee and sleep 15:40 <ubottu> Launchpad bug 1830862 in Apport "Apport reads arbitrary files if ~/.config/apport/settings is a symlink" [Undecided,New] https://launchpad.net/bugs/1830862 15:41 <tdaitx> (done) 15:41 <ubottu> Launchpad bug 1839795 in apport (Ubuntu) "PID recycling enables an unprivileged user to generate and read a crash report for a privileged process" [Undecided,Fix released] https://launchpad.net/bugs/1839795 15:41 <ubottu> Launchpad bug 1839413 in Apport "TOCTTOU ("time of check to time of use") "cwd" variable race condition" [High,New] https://launchpad.net/bugs/1839413 15:41 <ubottu> Launchpad bug 1839415 in Apport "Fully user controllable lock file due to lock file being located in world-writable directory" [Low,New] https://launchpad.net/bugs/1839415 15:41 <ubottu> Launchpad bug 1839420 in Apport "Per-process user controllable Apport socket file" [High,New] https://launchpad.net/bugs/1839420 15:41 <ubottu> Launchpad bug 1831252 in initramfs-tools (Ubuntu Disco) "panic=-1 is completely ignored by the initrd causing unexpected behaviour" [Medium,Fix committed] https://launchpad.net/bugs/1831252 15:41 <tdaitx> thanks bdmurray 15:41 <rbalint> aob: i'm out tomorrow 15:42 <bdmurray> and sil2100 is also out 15:44 <bdmurray> alright thanks everybody 15:44 <bdmurray> #endmeeting