[00:40] <Wimpress> Happy to help o/
=== vicamo_ is now known as vicamo
[06:08] <didrocks> good morning
[06:11] <pieq> Morning desktoppers!
[06:11] <didrocks> hey pieq
[06:11] <pieq> jibel, so I raised this issue: https://bugs.launchpad.net/ubuntu-manual-tests/+bug/1845582 but I cannot assign it to you for some reason.
[06:11] <ubot5> Launchpad bug 1845582 in Ubuntu Manual Tests "Installation tests – Indicate when using UEFI or Legacy BIOS mode" [Undecided,New]
[06:12] <pieq> and I was wondering if there was a plan to make the UEFI boot screen as nice as the Legacy BIOS version (if that's possible). At least to be able to select the language/locale for the live session
[06:12] <pieq> salut didrocks :)
[06:13] <jibel> pieq, that's a good point. I don't know how to best differenciate uefi and legacy test cases. We don't want to duplicate the tests. Maybe just add a couple of uefi tests (+sb) and everything else is legacy
[06:14] <jibel> otherwise we'll never know if the test has been run in legacy or uefi mode
[06:14] <pieq> jibel, indeed
[06:14] <pieq> jibel, do we have numbers in terms of what people use nowadays? I haven't used legacy BIOS in ages
[06:14] <pieq> but I might not be the typical user :D
[06:16] <jibel> pieq, yes we do.
[06:17] <jibel> pieq, it's 50/50
[06:18] <pieq> interesting
[06:28] <pieq> I don't know if it's easy to reproduce what we have for Legacy BIOS in UEFI mode... IIRC, for Legacy BIOS we use a Grub with gfxmode
[06:30] <RAOF> We should be able to be even fancier, because we can let the UEFI firmware leave the framebuffer in place.
[06:32] <RAOF> cf: Window 10's spinner.
[06:33] <pieq> RAOF, that would be great! Cause I feel the user experienced has regressed when moving from Legacy to UEFI. That sad black and white Grub menu looking at me...
[06:34] <pieq> and it's even worse on 4k screens, like the one on the XPS 13... the entries are so small!
[06:36] <pieq> RAOF, when you mean "even fancier", what does in include? I'm looking at https://wiki.archlinux.org/index.php/GRUB/Tips_and_tricks#Visual_configuration but I don't really understand if there are things that are UEFI-specific
[06:38] <RAOF> pieq: The reverse of that; rather than having GRUB set a mode, you can use the mode that the UEFI firmware has set (and keep the contents) and draw on it.
[06:39] <RAOF> So the limit is how much rendering you want to do; seamless transitions from POST to GRUB to GDM should be possible.
[06:39] <pieq> RAOF, is it the technique used for "flicker free" boot sequence as shown by https://archive.fosdem.org/2019/schedule/event/flickerfree/ ?
[06:40] <RAOF> Yeah, that'd be it.
[06:41] <pieq> RAOF, "seamless transition" you mean no black screen between each step? But could the options currently in Legacy BIOS Grub screen (select language, select to install "free software only" mode, etc.) be used for UEFI mode?
[06:42] <RAOF> Yes. You can do arbitrary drawing to the efi framebuffer if you want.
[06:42] <RAOF> All you need is some rasteriser!
[06:43] <pieq> errrr.... ok? :D
[06:44] <pieq> I'm always concerned when a sentence startts with "all you need is" or "simply do...."  :)
[06:44] <RAOF> That is indeed the trick!
[06:48] <pieq> RAOF, https://fedoraproject.org/wiki/Changes/FlickerFreeBoot
[06:49] <pieq> there are some interesting info there
[06:51] <oSoMoN> good morning desktoppers, and happy Friday!
[06:55] <didrocks> happy Friday oSoMoN
[06:58] <pieq> jibel, do you know the package name I have to use to file a bug against the GNOME settings?
[06:59] <oSoMoN> salut didrocks, pieq, jibel
[06:59] <oSoMoN> pieq, gnome-control-center
[07:00] <RAOF> pieq: We actually don't need most of that work in order to make the GRUB menu better. We just have to *actually draw* the better grub menu :)
[07:01] <pieq> merci oSoMoN !
[07:01] <oSoMoN> de rien
=== pstolowski|afk is now known as pstolowski
[07:06] <pieq> what about the welcome screen that is shown the first time the user logs into a freshly installed Ubuntu 19.10? The screen that suggests to connect to online accounts, setup location, etc.
[07:14] <oSoMoN> pieq, gnome-initial-setup
[07:15] <pieq> oSoMoN, merci ! Sorry for the stupid questions :S
[07:15] <pieq> (and for the bugs I'm filing)
[07:16] <oSoMoN> de rien! thanks for the bugs, and those are not stupid questions, there's no straightforward way to infer the name of a package from a running app
[07:17] <pieq> oSoMoN, is it possible to re-run this initial screen after I closed it?
[07:19] <oSoMoN> yes, definitely, let me dig in the code to find how
[07:25] <oSoMoN> pieq, try deleting ~/.config/gnome-initial-setup-done
[07:27] <pieq> oSoMoN, thanks
[07:27] <oSoMoN> yw
[07:44] <paride> Hi
[07:44] <paride> Trevinho, so https://gitlab.gnome.org/GNOME/gnome-shell/merge_requests/1 didn't make it in Eoan, right?
[07:44] <gitbot> GNOME issue (Merge request) 1 in gnome-shell "Better mode extension support: status sync and disabling." [1. Feature, 5. Extensions, Merged]
[07:45] <paride> asking as I saw your comment there about cherry-picking it for Bionic, back then :)
[07:59] <jibel> Laney, another review of ubiquity https://code.launchpad.net/~jibel/ubiquity/+git/ubiquity/+merge/373302
[08:00] <willcooke> yawn.  Moring all
[08:00] <jibel> mornign willcooke
[08:01] <didrocks> hey willcooke
[08:01] <oSoMoN> good yawning willcooke
[08:01] <Laney> helloooooooooooooooooo
[08:02] <didrocks> hey Laney
[08:02] <oSoMoN> hello Laney
[08:02] <Laney> moin didrocks oSoMoN
[08:03] <Laney> happy (checks to be sure) friday!
[08:09] <Wimpress> o/
[08:09] <willcooke> So it seems a lot of my problems with ISO testing *were* because my USB stick died.  That was very confusing for a while.
[08:10] <didrocks> hey willcooke
[08:10] <didrocks> oupss
[08:10] <didrocks> Wimpress:
[08:10] <didrocks> :)
[08:10] <didrocks> toomany wi*
[08:10] <Wimpress> :-)
[08:10] <pieq> willcooke, always annoying when it happens!
[08:10] <pieq> (and good morning!)
[08:17] <jibel> willcooke, also this beta is far from being good
[08:17] <jibel> there's still a couple of weeks to make is awesome :)
[08:17] <jibel> it*
[08:17] <willcooke> jibel, yeah I found quite a few problems.  Most were in the installer I found.  What about you?
[08:23] <jibel> willcooke, all the things reported yesterday. grub, casper, ubiquity mainlly
[08:23] <willcooke> oki
[08:23] <jibel> apart from that nothing really important
[08:24] <willcooke> It looks like T_revinho's "lock up" fix didnt get in in time, is that right?
[08:24] <willcooke> (by which I mean, I managed to trigger it with the screen reader)
[08:24] <jibel> it's right, but Laney's fix to ubiquity is good enough
[08:26] <jibel> there is still bug 1840122 which is terribly annoying because it makes automated tests useless. willcooke did you get any traction from anyone?
[08:26] <ubot5> bug 1840122 in linux (Ubuntu Eoan) "System fails to reboot from live session or ubiquity-dm - squashfs_read_data failed to read block" [High,Confirmed] https://launchpad.net/bugs/1840122
[08:27] <willcooke> jibel, yeah that came up yesterday too, I think it was mwhudson who said he was looking at it
[08:28] <jibel> k
[08:30] <Laney> I got told it was xno_x who is assigned to that
[08:34] <willcooke> lemme find out for sure
[08:36] <willcooke> oh, something which came up yesterday, the OEM slideshow is using the old screenshots in Ubiquiuty sideshow. I looked at the source and I think they are symlinks, but they link to "../../../foo/bar/baz.png" - so they should just work.  Do I need to recreate them in the package?
[08:38] <Laney> not sure I knew there was a different slideshow there
[08:38] <willcooke> same :)
[08:38] <Laney> D:
[08:39] <Laney> the test slideshot from the package looks right to me
[08:39] <Laney> wouldn't the old images have been removed from the package and therefore the iso?
[08:40] <willcooke> i would have thought so
[08:40] <willcooke> I need to recreate it here
[08:41] <pieq> jibel, I haven't seen this issue yet. Does it happen randomly or did you find a way to reproduce it?
[08:41] <Laney> some stupid outage has broken my juju environments
[08:41]  * Laney weeps
[08:45] <willcooke> Bah, it is the right slideshow
[08:48] <willcooke> Wimpress, marking you OEM installer bug as invalid.  WFM.  Please let me know if I'm wrong.
=== vicamo_ is now known as vicamo
[10:30] <Laney> jibel: is your zfs UI signed off by design now?
[10:30]  * Laney is just reviewing that atm
[10:40] <jibel> Laney, it's slightly different from the design proposed by mpt. We cannot implement it this cycle because it also removed LVM from the guided partitioning page and moves it to a new dialog.
[10:41] <Laney> jibel: yeah sure, maybe as an ok compromise / achieveable implementation / at least not totally unacceptable :-)
[10:41] <Laney> reviewing shell scripts is fun
[10:41] <Laney> MUHAHAHA
[10:43] <jibel> i'll refresh the screenshot on the MP and ask for an explicit approval
[10:49] <jibel> mpt, could you review the scrreenshot attached to https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1843768/comments/3 and +1 if it's an acceptable compromise. We're very late in he cycle and adding a new dialog to the installer is not something we can do. We'll implement the design you proposed in 20.04.?
[10:49] <ubot5> Launchpad bug 1843768 in ubuntu-meta (Ubuntu) "[FFe] Ubiquity with zfs install option" [Undecided,New]
[10:49] <jibel> -?
[10:50] <mpt> looking
[11:10] <mpt> jibel, done
[12:10] <Laney> jibel: I tried your MP and it created a broken system, no home directory for my user
[12:12] <Laney> zpool list shows nothing, / is ext4
[12:12] <jibel> Laney, can you paste the installation log somewhere?
[12:12] <Laney> perhaps I did something wrong, I just apt installed the new ubiquity into a live session
[12:12] <jibel> /var/log/syslog and /var/log/installer/debug
[12:13] <Laney> yes sure, just trying to get it online :>
[12:18] <Laney> jibel: https://people.canonical.com/~laney/weird-things/installer.tar.xz that's a tarball of /var/log/installer
[12:18] <Laney> s/the new ubiquity/zfsutils-linux and &/
[12:19] <Laney> still will submit my code review that I wrote before testing it
[12:20] <jibel> Sep 27 11:50:22 ubuntu ubiquity: umount: /target: target is busy.
[12:20] <jibel> Sep 27 11:50:22 ubuntu ubiquity: E: Failed to unmount /target. Exiting!
[12:21] <jibel> the script aborted
[12:21] <jibel> now why target is busy remains a mystery
[12:23] <Laney> nod, good that this was detected, we just didn't recover very well
[12:24] <jibel> it seems that the creation of the swap file on /target went on its way.
[12:31] <Laney> review submitted, hope you like nitpicks /o\
[12:31]  * Laney goes for lunch
[12:32] <Trevinho> paride: it has, but not to bionic
[12:32] <Trevinho> morning folks
[12:32] <jibel> mpt, latest rev following your comments https://launchpadlibrarian.net/444444810/guided_partitioning_with_zfs_2.png
[12:33] <paride> Trevinho, yeah of course not, but looks like I can't find what I was expecting in Eoan too.
[12:33] <mpt> jibel, that’s much better
[12:33] <paride> but maybe I misinterpreted the scope of that patch
[12:33] <mpt> jibel, the vertical spacing is still off, though that was the least important issue
[12:34] <jibel> mpt, yes, I didn't find what's causing it yet.
[12:34] <Trevinho> paride: mhmh... What not? didrocks knows that way better than me though :-)
[12:34] <mpt> jibel, when the checkboxes are disabled, can you disable their captions too (i.e. grey them out)?
[12:35] <jibel> mpt, I can do that
[12:36] <jibel> mpt, however I cannot fix the entire UI and things not related to adding zfs :)
[12:36] <jibel> but i'll do my best
[12:39] <paride> Trevinho, I thought (possibly wrongly) that change allowed to enable/disable "built-in" extensions like the ubuntu dock, or at least made the Ubuntu Dock extension status in Gnome Tweaks consistent with the reality (= shows up as enabled if it
[12:39] <paride> if it's actually enabled).
[12:39]  * didrocks is out of context
[12:39] <didrocks> if you talk about the disable thingy, yes
[12:39] <didrocks> which is upstream
[12:40] <didrocks> our distro patch is to ensure session extensions are only loaded from the system and can't be upgraded out of it
[12:40] <paride> I see. Still it seems it can't be enabled/disabled from gnome tweaks
[12:40] <paride> which is what I thought that patch did
[12:41] <didrocks> the patch is gnome-shell only. Unsure if tweaks caught up
[12:41] <paride> I see. Well, thanks!
[12:41] <didrocks> gnome-shell-extension-prefs should show the correct state though
[12:43] <paride> now I'm in a GNOME (non-Ubuntu) session. I'll try later.
[13:09] <paride> didrocks, gnome-shell-extension-prefs works as expected. I thought gnome-tweaks was just an interface for the same thing. well, thanks!
[13:10] <didrocks> paride: unfortunately, no, they copy the logic :/
[13:10] <paride> TIL :)
[14:06] <clobrano> hey everyone 0/
[14:08] <clobrano> we discovered this problem with Yaru https://github.com/ubuntu/yaru/issues/1548
[14:08] <gitbot> ubuntu issue 1548 in yaru "Firefox in the mixed theme: active menu items are invisible" [Bug, Open]
[14:09] <clobrano> it's due to another fix to make unfocused tabs visible in no-titlebar mode (basically, you either have invisible tabs in no-titlebar-mode, or invisible menu in titlebar-mode)
[14:10] <clobrano> however, since by default Firefox starts in titlebar-mode, I wonder if a fix is necessary before eoan, if there's still time, with a freeze exception
[14:40] <clobrano> update ^, a fix for both titlebar-mode and no-titlebar mode is actually in review https://github.com/ubuntu/yaru/pull/1549
[14:40] <gitbot> ubuntu issue (Pull request) 1549 in yaru "apps: Fix invisible menuitems in Firefox under Yaru-ambiance" [Open]
[15:15] <Laney> clobrano: sounds reasonable to me
[15:34] <clobrano> Laney, thanks! Here's the UIFe https://bugs.launchpad.net/ubuntu/+source/yaru-theme/+bug/1845680
[15:34] <ubot5> Launchpad bug 1845680 in yaru-theme (Ubuntu) "[UIFe] Firefox active menu items are invisible in Yaru "Ambiance"" [Undecided,New]
[16:18] <xnox> Laney:  i bet i need to add the "noise" texture to caching paths to casper, as otherwise it fails to load the noise texture at shutdown causing failure to reboot or some shit like that.
[16:27] <xnox> Laney:  so caching "eject" and all the plymouth ubuntu-logo theme pngs makes it work for me.... i think
[16:27] <xnox> let me try again
[16:29] <xnox> and no
=== pstolowski is now known as pstolowski|afk
[16:39] <Laney> xnox: it was happening before that noise stuff
[16:40] <Laney> pretty sure anyway, check the date the bug was reported & it had been failing for some time then already
[16:51] <xnox> horum
[16:51] <xnox> i hate that we never got down to squashfs errors that started in disco.
[16:51] <xnox> yeah, not quite sure what's going on now.
[16:56] <xnox> Laney:  i am making changes to the system.... and it sometimes reboots just fine, what i don't know if my changes actually help at all, or not.
[16:58] <willcooke> night all, hope you all have good weekends
[16:58] <xnox> Laney:  i do suspect that we have fallout from https://bugs.launchpad.net/ubuntu/+source/casper/+bug/1436715
[16:58] <ubot5> Launchpad bug 1436715 in casper (Ubuntu Vivid) "Vivid DVD fails to reboot or shutdown w/o hard reset" [Critical,Fix released]
[16:58] <xnox> which newer systemd does not like
[18:30] <tomreyn> fwiw, i just ran into bug 1844026 while testing eaon beta1
[18:30] <ubot5> bug 1844026 in nautilus (Ubuntu) "nautilus crashed with SIGSEGV" [Medium,Confirmed] https://launchpad.net/bugs/1844026
[18:31] <tomreyn> that's after first reboot, fully updated unmodified minimal install.