[00:22] <callmepk> morning
[05:30] <seb128> goood morning desktopers!
[05:30] <callmepk> morning seb128 
[05:33] <seb128> hey callmepk , how are you?
[05:34] <luna_> morning
[05:34] <seb128> hey luna_,!
[05:35] <callmepk> seb128 I want to say I am fine but I am just way too tired recently :|
[05:36] <seb128> it's 6:30am and I'm at the component, I share that sentiment even if it's not for the same reasons ;-)
[05:37] <luna_> yeah early here too, but had nightmares and woke up early
[05:41] <callmepk> I am going to have fun again with my 3 assignments and 3 school projects later tonight :)
[05:47] <seb128> that's a making for long days right?
[05:53] <duflu> Morning seb128 and luna_. Afternoon callmepk 
[05:54] <callmepk> seb128 yeah....
[05:54] <callmepk> afternoon duflu 
[05:54] <callmepk> Also morning luna_ 
[05:56] <luna_> morning
[06:11] <jibel> morning all
[06:13] <duflu> Morning jibel 
[06:13] <jibel> hi duflu 
[06:29] <seb128> hey duflu, how are you?
[06:29] <seb128> salut jibel, en forme ?
[06:30] <duflu> seb128, a bit tired and not great. But I was super fresh all Monday so it balances out. You?
[06:31] <seb128> tired as well, I was up at 5:30am which isn't a proper time to get up for me! also on the edge of a start of a cold, let's see how today is going
[06:33] <seb128> jamesh, hey, weekly meeting reminder, you seem to be the only one missing for this round ;-)
[06:34] <oSoMoN> good morning desktoppers
[06:35] <jamesh> seb128: okay
[06:35] <jamesh> thanks
[06:37] <luna_> won't attend the meeting today as its my birthday
[06:37] <seb128> jamesh, thanks :)
[06:38] <seb128> lut oSoMoN, comment ça va ?
[06:38] <seb128> luna_, oh, happy birtday! I hope you enjoy the day!
[06:38] <oSoMoN> seb128, bien, et toi?
[06:38] <oSoMoN> happy birthday luna_ !
[06:38] <luna_> thx
[06:38] <seb128> oSoMoN, levé trop tôt et un signe de début de rhume mais sinon ça va
[06:39] <oSoMoN> ah zut, soigne-toi bien!
[06:39] <luna_> oSoMoN: https://archive.mozilla.org/pub/firefox/releases/84.0b1/
[06:39] <duflu> Morning oSoMoN 
[06:40] <oSoMoN> good afternoon duflu 
[06:40] <oSoMoN> luna_, I know, but we usually don't start packaging betas before b3
[06:40] <luna_> not sure if Ubuntu packages the betas or not however 
[06:40] <luna_> oSoMoN: ah alright
[06:40] <luna_> so friday then
[06:41] <luna_> i did package it up for another distro 
[06:46] <oSoMoN> which one?
[06:47] <luna_> Arch
[06:52] <jibel> Salut seb128, en forme :)
[06:58] <oSoMoN> salut jibel 
[07:16] <jibel> Salut oSoMoN 
[07:19] <didrocks999> good morning
[07:32] <duflu> Hi didrocks999 
[07:40] <didrocks999> hey duflu 
[07:41] <didrocks999> already so many 9 early? Let’s fix it
[08:48] <marcustomlinson> morning desktoppers
[08:52] <didrocks> hey marcustomlinson 
[08:53] <duflu> Morning marcustomlinson 
[09:04] <Laney> moin
[09:08] <seb128> hey marcustomlinson, Laney, didrocks , how are you today?
[09:18] <jibel> Bonjour marcustomlinson Laney 
[09:19] <marcustomlinson> seb128: doing alright thanks you?
[09:25] <duflu> Morning Laney 
[09:29] <Laney> hey seb128 & jibel & duflu 
[09:29] <Laney> doin GOOOOOD
[09:32] <seb128> marcustomlinson, I'm okish, was woken up a 5:30 and I've a slight start of a cold, let's see
[09:32] <seb128> hey Laney 
[09:32] <seb128> also having fun with livecd-rootfs now
[09:33] <seb128> trying to figure out why the canary overlay.minimal is lacking /sys/kernel/security and if that's a bug and if the snap seeding is supposed to happen in that layer
[09:33] <seb128> I did some steps forwards on the ISO build but still to learn!
[09:35] <didrocks> hey seb128, going well, and you? sounds like you are enjoying building ISOs :)
[09:36] <didrocks> seb128: IIRC, we decided to put the seeding of snaps in the lower layer possible because snap was a requirement for the next installer
[09:36] <didrocks> and was part of the "base"
[09:36] <didrocks> but I think this can be revisited ofc
[09:36] <seb128> didrocks, ah, makes sense
[09:36] <didrocks> (and only added in the first layer you are adding a snap for instance)
[09:36] <seb128> it's failing on snap-store
[09:36] <seb128> but would probably fail the same way on subiquity
[09:36] <didrocks> I think yeah
[09:37] <seb128> error: cannot preseed without the following mountpoints:
[09:37] <seb128>  - /build/overlay.minimal/dev
[09:37] <seb128>  - /build/overlay.minimal/sys/kernel/security
[09:37] <seb128> I need to figure out if those a created in another layer and need to be moved down
[09:37] <didrocks> maybe just a bindmount missing, but hoping it doesn’t sysctl your system
[09:39] <seb128> didrocks, right, it's a bit weird it should work with those recents enough commits
[09:39] <seb128> https://git.launchpad.net/livecd-rootfs/commit/live-build/auto/build?id=84397b5098967
[09:40] <seb128> and https://git.launchpad.net/livecd-rootfs/commit/?id=109e6c6
[09:40] <seb128> there is a mount --bind /sys/kernel/security chroot/sys/kernel/security there
[10:54]  * seb128 learnt about ssh-import-id lp:launchpadid , quite useful!
[11:00] <Laney> 😎
[11:28] <cpaelzer> tkamppeter: I have disk left of a system that is broken, and I have a new system (both 20.04). I remember that the printer config was horribly complex (some odd thermal printer) - is there a file/directory I could copy over to get the old printer config to the new system?
[11:48] <didrocks> Laney: do you know if it’s doable to have some kind of systemd user timer that can’t be disabled/mask by the user?
[11:53] <Laney> didrocks: no it's not, the search paths are always gone through in order and ones in the home directory take precedence over system ones
[11:53] <didrocks> people will still be able to mask or edit the unit locally if we set/symlink it to /etc/systemd/user/ in the local path, this is what I was afraid of
[11:53] <didrocks> I wonder how we can do this kind of "mandatory" units while still being initiated by the user to get its env variable
[11:54] <Laney> what are you trying to protect from?
[11:54] <didrocks> GPO refreshing being diabled
[11:54] <didrocks> disabled*
[11:54] <didrocks> while the GPO list and access are controlled by the krb5 ticket created by sssd
[11:55] <Laney> if it's running from the user context then the user is going to be able to hax0r it one way or another
[11:55] <didrocks> (ticket owned only by the user and referenced by KRB5CCNAME)
[11:55] <Laney> ah yes, I remember this variable :p
[11:55] <didrocks> indeed, this was our conclusion, and why we are trying to get this env variable for all logged in user
[11:56] <didrocks> then, the daemon, running as root, can reuse the ticket
[11:56] <Laney> you can have systemd system units with User= too
[11:56] <didrocks> BUT unsure how we can safely get it, as systemd --user doesn’t have this env variable either (and show-environement can be spawn anyway)
[11:56] <Laney> that might be helpful
[11:56] <Laney> right, --user doesn't have it because of that email thread I started the other day
[11:56] <didrocks> yeah, but as systemd --user won’t have it, I wonder if the service would have it
[11:57] <didrocks> and it needs to be time-based
[11:57] <didrocks> exactly
[11:57] <Laney> sssd needs to have a 'session' pam thing to actually set that
[11:57] <didrocks> and so, I wonder how we can do that elegantly without relying and process environment seeking
[11:59] <Laney> it needs to run 'in' the session or can it happen to the session from outside?
[11:59] <Laney> not sure what kind of actions you might need to take here
[12:00] <Laney> it does feel like a rootish thing to enforce lock down policies on user sessions doesn't it
[12:00] <didrocks> it needs to communicate with the daemon, the only thing we need 'in the session' for is to get that KRB5CCNAME env variable
[12:01] <didrocks> the enforcement is done by a daemon as root
[12:02] <didrocks> basically, what we want is: "What is your user KRB5CCNAME content" pings at regular time interval, from the user, without a way for the user to disable this ping
[12:02] <Laney> regular because it can change?
[12:03] <didrocks> regular because the remote policy can change and needs to be refreshed
[12:03] <didrocks> (the issue is not a logging time, but for regular refresh)
[12:03] <Laney> ah the env var doesn't change, but the remote stuff it refers to can
[12:03] <didrocks> right
[12:04] <didrocks> or
[12:04] <didrocks> maybe it can change, if the ticket expires
[12:04] <didrocks> and sssd renewed it without logging out and back
[12:04] <didrocks> (I think sssd would reuse the same ticket name, but I couldn’t check that yet)
[12:04] <Laney> hmm then it won't be communicated to running processes
[12:05] <didrocks> yep
[12:05] <didrocks> unsure about that, but not the core of the question if this is technically possible anyway
[12:05] <didrocks> it seems systemd don’t have the notion of "mandatory user units"
[12:06] <Laney> I think do the pam/sssd fix, and then you can rely on the environment variable being set for systemd --user
[12:06] <Laney> and somehow™ the daemon can get that
[12:06] <didrocks> the question is somehow™
[12:06] <Laney> like watch logind for new session signals to know when to go look
[12:06] <didrocks> because things like systemctl --user show-environment needs the user bug
[12:06] <didrocks> bus*
[12:07] <Laney> to be honest with you, making sssd write that out to /run wouldn't be a terrible idea
[12:07] <didrocks> and we are back to the same questionning of getting it from the root daemon
[12:08] <didrocks> yeah, or having the daemon storing it there, and refreshing it on login
[12:08] <didrocks> (like a symlink by uid)
[12:08] <Laney> $XDG_RUNTIME_DIR/KRB5whateveritis
[12:08] <didrocks> yep
[12:08] <Laney> assuming it's one per user, not one per session
[12:09] <Laney> i.e. further logins should use the same ticket
[12:09] <didrocks> and then, handling the refresh by a system unit timer
[12:09] <didrocks> which will get them back
[12:09] <didrocks> (for all logged in user, hopefully)
[12:09] <didrocks> sounds like the best bet for now
[12:10] <Laney> so then the session pam thing would be: read that file, set the env var -> for compatibility with existing software
[12:10] <didrocks> I would even add: ping the daemon to setup policy with the env var
[12:10] <didrocks> then, the daemon will set this symlink
[12:10] <didrocks> and reload
[12:10] <didrocks> for refresh
[12:11] <didrocks> sounds like the best strategy for now at least
[12:11] <didrocks> and it doesn’t really on messing with env variables of running processes, which is what we wanted to avoid anyway
[12:13] <Laney> definitely, it is just needed for compat, and probably if there's a library which programs actually use (not sure if there is or not) it would be changed to prefer the file and then eventually the var can just go away
[12:15] <didrocks> yeah, let’s see how it goes, thanks for confirming on the systemd thingy :)
[12:21] <Laney> no worriesssSSSS
[12:59] <seb128> Saviq, thanks for commenting on that bug, do you have a case where the verbose option from multipass transfer gives some output, just curious :)
[12:59] <seb128> it has several -v level but even in max mode it prints nothing on normal copy, file overwriten, invalid path
[13:00] <seb128> sounds like vvvvvv isn't really verbose still :p
[14:03] <seb128> oSoMoN, do you think there is any nagging we could do to get https://phabricator.services.mozilla.com/D95171 reviewed? or is distro patching an option?
[14:03] <seb128> https://bugzilla.mozilla.org/show_bug.cgi?id=1674150
[14:04] <oSoMoN> seb128, nagging wouldn't hurt
[14:17] <seb128> bah
[14:17] <seb128> The following packages have unmet dependencies:
[14:17] <seb128>  libgssapi-krb5-2 : Depends: libkrb5-3 (= 1.17-10) but it is not going to be installed
[14:27] <hellsworth> good morning desktopers
[14:28] <kenvandine> good morning hellsworth 
[14:29] <Wimpress> Time appropriate greetings
[14:30] <hellsworth> hi yall :)
[14:30] <jibel> Good morning hellsworth 
[14:30] <oSoMoN> good morning hellsworth 
[14:30] <Wimpress> #startmeeting Desktop Team Weekly Meeting - 2020-11-17
[14:30] <oSoMoN> good afternoon Wimpress 
[14:30] <meetingology> Meeting started Tue Nov 17 14:30:29 2020 UTC.  The chair is Wimpress. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[14:30] <meetingology> Available commands: action commands idea info link nick
[14:30] <didrocks999> hey
[14:30] <Wimpress> Roll call:  didrocks, duflu, hellsworth, jamesh, jibel, kenvandine, Laney, marcustomlinson, oSoMoN, seb128 , tkamppeter, Trevinho, robert_ancell, callmepk
[14:30] <hellsworth> o/
[14:30] <marcustomlinson> \o
[14:30] <Trevinho> o/
[14:30] <seb128> _o/
[14:30] <kenvandine> o/
[14:30] <oSoMoN> o/
[14:31] <Wimpress> #topic rls-bb-bugs
[14:31] <Wimpress> http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-bb-incoming-bug-tasks.html
[14:31] <Wimpress> All clear.
[14:31] <Wimpress> http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-bb-tracking-bug-tasks.html
[14:32] <Wimpress> http://launchpad.net/bugs/1865226
[14:32] <seb128> that needs to be assigned to Trevinho 
[14:32] <Wimpress> Yep
[14:32] <Trevinho> I think I already am for few parts
[14:32] <Wimpress> You are.
[14:32] <seb128> right, bionic was missing, it's fixed now
[14:33] <Wimpress> Thanks.
[14:33] <Wimpress> http://launchpad.net/bugs/1899179
[14:33] <seb128> it doesn't sound like a rls issue to me, I would just wontfix
[14:34] <seb128> it's an autopkgtest issue on lts-1 and ppc64el
[14:34] <kenvandine> indeed
[14:34] <Wimpress> Ugh. 2FA day. Sec....
[14:35] <Wimpress> OK, done.
[14:36] <Wimpress> http://launchpad.net/bugs/1867996
[14:36] <Wimpress> I've completed the assignment to Trevinho :-)
[14:36] <seb128> that's fix commited
[14:36] <seb128> we don't review those
[14:37] <seb128> (since that means they are already uploaded and don't need more action)
[14:37] <Wimpress> So it is.
[14:37] <Wimpress> #topic rls-ff-bugs
[14:37] <Wimpress> http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-ff-incoming-bug-tasks.html
[14:38] <Wimpress> http://launchpad.net/bugs/1873052
[14:39] <seb128> those are assigned but need targetting and untagging
[14:40] <Wimpress> Is gnome-shell a valid "Affects"?
[14:42] <seb128> yes
[14:42] <seb128> Trevinho, ^ ?
[14:42] <Trevinho> seb128: mutter is the affected src
[14:43] <Wimpress> OK. Add focal target and assignments.
[14:43] <Wimpress> *Added
[14:43] <Wimpress> http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-ff-tracking-bug-tasks.html
[14:44] <Wimpress> http://launchpad.net/bugs/1903524
[14:44] <Wimpress> Looks like jibel has that one.
[14:45] <seb128> right
[14:45] <didrocks> (already uploaded in hirsute)
[14:45] <Wimpress> Excellent
[14:46] <Wimpress> http://launchpad.net/bugs/1865226
[14:46] <seb128> it's fixed now
[14:46] <Wimpress> That one now.
[14:46] <Wimpress> http://launchpad.net/bugs/1902148
[14:47] <hellsworth> ah so i flagg'd this because i am not really sure what to do
[14:47] <seb128> -1 for rls, it's the snap
[14:47] <hellsworth> on on e hand i can make the snap match the deb and remove the startcenter and math icons
[14:48] <hellsworth> but there's pushback from a TDF person on removing icons.. i see the argument and could use some advice
[14:48] <oSoMoN> that's a separate issue though?
[14:48] <seb128> I would say we can discuss that here after the meeting
[14:48] <seb128> it's not a rls issue
[14:48] <hellsworth> ok
[14:48] <seb128> no need to hold the meeting/Team on it
[14:48] <hellsworth> should have just brought it up in AOB instead of tagging
[14:48] <seb128> it's ok
[14:49] <seb128> or just discuss bugs on the channel any day/time
[14:49] <seb128> on the european afternoon you are likely to get people around
[14:49] <Wimpress> OK. We can circle back to that one.
[14:49] <Wimpress> #topic rls-gg-bugs
[14:49] <Wimpress> http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-gg-incoming-bug-tasks.html
[14:50] <Wimpress> All fine.
[14:50] <Wimpress> http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-gg-tracking-bug-tasks.html
[14:50] <Wimpress> http://launchpad.net/bugs/1901272
[14:51] <seb128> it's incomplete on the bluez side (which is ours)
[14:51] <seb128> we can skip
[14:51] <Wimpress> OK
[14:52] <Wimpress> I've been here with bluez on the Pi some years ago. I'll look into it tonight.
[14:52] <Wimpress> http://launchpad.net/bugs/1898005
[14:53] <seb128> that needs assigning to marco on the shell component
[14:53] <Wimpress> Added Daniel 
[14:53] <seb128> or that
[14:53] <seb128> I'm trying to convince Daniel to help with some of the dock and desktop issues, Marco is overworked
[14:53] <Wimpress> That sounds sensible.
[14:53] <Laney> +1
[14:54] <seb128> ah, I didn't see Daniel did pick it up and proposed a shell fix
[14:54] <Wimpress> #topic update_excuses_by_team.html#desktop-packages
[14:54] <Wimpress> https://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses_by_team.html#desktop-packages
[14:54] <seb128> I would still skip that one for now
[14:54] <seb128> start of cycle is cahos, the list isn't really reviewable in a meeting atm
[14:55] <Wimpress> OK
[14:55] <seb128> (and quite some things will get sorted out by themselve)
[14:55] <hellsworth> seb128: should i go ahead an get LO tests relaunched?
[14:55] <Wimpress> #topic AOB
[14:55] <Laney> there's some I noticed, like 
[14:55] <Laney> indicator-session/17.3.20+19.10.20190921-0ubuntu1: arm64 (log, history), armhf (log, history), ppc64el (log, history)
[14:55] <seb128> hellsworth, you can try, check for queued one first though because I expect other people do hammer the retry button (I did for some already)
[14:55] <Laney> but they will probably still be there next week
[14:55] <hellsworth> right i will check for queued ones first for sure
[14:55] <Laney> I would wait with Libreoffice, if you're talking about flaky tests
[14:55] <seb128> right, there are some build issues in there
[14:56] <Laney> for the queue to finish, that might make them more likely
[14:56] <Laney> to work, or fix the flakes *cough* :)
[14:56] <hellsworth> Laney: yeah i'm talking about flaky tests
[14:56] <hellsworth> yeah..
[14:56] <seb128> webkit2gtk is an ICE also, fun, if someone fancy working with d_oko on it ;)
[14:57] <Wimpress> Anything else before we finish and hellsworth can get to the LO discussion?
[14:58] <seb128> not from me
[14:58] <Laney> nah crack on
[14:58] <didrocks> nope
[14:58] <Wimpress> OK. Thanks everyone.
[14:58] <Wimpress> #endmeeting
[14:58] <meetingology> Meeting ended Tue Nov 17 14:58:45 2020 UTC.  
[14:58] <meetingology> Minutes:        http://ubottu.com/meetingology/logs/ubuntu-desktop/2020/ubuntu-desktop.2020-11-17-14.30.moin.txt
[14:58] <hellsworth> thanks
[14:58] <oSoMoN> thanks
[14:59] <marcustomlinson> thanks
[14:59] <kenvandine> thanks
[14:59] <hellsworth> so re #1902148, i figured the solution would be to remove the problematic icon because it aligns with removing hem from the deb. but there's mixed feelings about removing the startcenter and math icons from the deb: https://bugs.launchpad.net/ubuntu/+source/libreoffice/+bug/1696250
[14:59] <hellsworth> I see both sides and wanted the team's opinion.
[15:00] <hellsworth> seems like if upstream is making the icons, we should ship them? but there are quite a lot of LO icons, making it cluttersome in the applications
[15:00] <hellsworth> alternatively we could say well the deb doesn't have those icons and the snap does and i could fix the icon in the snap
[15:01] <oSoMoN> removing those icons in the deb is a less than satisfactory solution, so I wouldn't recommend doing that in the snap
[15:01] <seb128> the reason we remove it for the deb is?
[15:01] <oSoMoN> « Currently, LibreOffice takes up a whole row on my computer (6 icons) which seems excessive with GNOME Shell's current design. »
[15:01] <didrocks> something we discussed altogether IIRC a couple of years ago and agreed at the sprint
[15:01] <hellsworth> the reason it was removed from the deb was cosmetic, to reduce the clutter
[15:02] <didrocks> like too many items in the overview, confusing icon, and a third thingy I don’t remember
[15:02] <didrocks> (that was coming from an user survey)
[15:03] <oSoMoN> since the snap is not the default for Ubuntu users, IMO it can afford to be closer to upstream builds
[15:03] <didrocks> ack
[15:03] <hellsworth> ok that makes sense
[15:03] <hellsworth> so keep the icons in the snap
[15:03] <seb128> I would let the snap be the default upstream experience
[15:03] <seb128> that doesn't solve the matching bug though
[15:03] <kenvandine> +1
[15:04] <hellsworth> well the matching bug can be fixed theoretically. but when i found the icons were missing from the deb i thought a solution just fell into my lap
[15:04] <hellsworth> so ok thanks. i'll keep the icons and fix it
[15:05] <kenvandine> great
[15:07] <oSoMoN> that sounds like a plan
[15:07] <Laney> "Not everything can be put to some kind of vote first, or we'd have brexits all over the place" ♥
[15:08] <marcustomlinson> "It's easier to ask forgiveness than it is to get permission"
[15:11] <hellsworth> sure but i wanted to do the right thing.. and needed to talk about it to figure out what that was
[15:11] <kenvandine> yeah, loved the brexit comment :)
[15:11] <hellsworth> i mean i agree there are an absurd number of LO icons
[15:11] <kenvandine> hellsworth: that was from the bug report :)
[15:11] <hellsworth> i know i read it :)
[15:12] <hellsworth> but that kind of is how open source software works.. just about everything is consensus style
[15:56] <Laney> w@w
[15:56] <Laney> hirsute images passed smoke testing
[16:24] <hellsworth> is autopkgtest.ubuntu.com/running working for anyone?
[16:27] <hellsworth> marcustomlinson: would you please relaunch these autopkgtests? https://paste.ubuntu.com/p/7nWDdPWsJb/
[16:27] <marcustomlinson> sure
[16:27] <hellsworth> i did verify that none of these are running but then closed the autopkgtests tab.. went to reload it and can't get it back but i swear none of these were running :)
[16:27] <hellsworth> thanks
[16:29] <marcustomlinson> hellsworth: hmm, the pages aren't loading for me either
[16:30] <hellsworth> ok i can go alert IS in mattermost
[16:34] <Laney> works for me
[16:34] <hellsworth> IS redirected me to mr. Laney 
[16:34] <Laney> perhaps something went down temporarily
[16:34] <hellsworth> right that's probably what happened. my page eventually loaded but geez after more than 5 min
[16:34] <hellsworth> but great it works!
[16:34] <hellsworth> thanks for fixing it Laney :)
[16:35] <Laney> didn't do anything!
[16:35] <hellsworth> you looked at it!
[16:36] <marcustomlinson> nope still not loading for me
[16:37] <hellsworth> i just launched it in a new tab and it's thinking about it. maybe it will load eventually
[16:45] <marcustomlinson> "Internal Server Error"
[16:52] <Laney> soz was on a call, lemme look
[16:56] <Laney> I did the first two, did take a few seconds but they worked
[16:57] <marcustomlinson> ah ok loading now for me
[16:57] <marcustomlinson> I'll do 3 onward
[16:58] <Laney> maybe we got DoSed!
[16:59] <marcustomlinson> hellsworth: all triggered
[16:59] <marcustomlinson> still super slow