[01:09] <dbungert> so I've got a partial fix for LP: #1821412 that I think is worth an upload - how do I changelog a partial fix?  That bug also mentions wayland-specific issues that my fix doesn't address.  Or I could track the wayland stuff in a new bug maybe?
[01:13] <mwhudson> dbungert: my take is that the wayland stuff is a different issue so yeah
[01:14] <mwhudson> put that bug in your changelog and make sure the other stuff is covered in a different bug
[01:14] <mwhudson> dbungert: also want me to target that bug to some series? just focal or also h and i?
[01:26] <dbungert> mwhudson: I think this should be a pretty easy SRU actually, let's try B/F/H/I.  Does this need higher launchpad permissions?  I attempted to target the bug to a series but didn't see anything relevant.
[01:26] <mwhudson> yes, it requires some permission bit or other, i forget exactly which
[01:27] <mwhudson> dbungert: ok updated
[01:45] <mwhudson> how do i get apt-get update to output progress info when stdout is not a terminal?
[01:46] <Eickmeyer[m]> mwhudson: If you figure it out, I'd like to know. I feel like there's a dbus backend, but I've always wanted progress in a gui.
[01:47] <mwhudson> Eickmeyer[m]: well you can do it with python-apt if you want
[01:47] <mwhudson> but i don't really want to do that
[01:47] <Eickmeyer[m]> Yeah, that's fair.
[01:48] <sarnold> mwhudson: pv -d pid maybe?
[01:55] <jchittum> took a bit, but i've confirmed that lxc containers for jammy are available now. lxc launch ubuntu-daily:jammy will get you something made today
[01:56] <bryceh> jchittum, ooh awesome, thanks
[01:56] <jchittum> cloud-images.ubuntu.com have dailies up as well, and streams are gtg. had some bumps getting streams together :/
[02:40] <mwhudson> Eickmeyer[m]: seems it's -q0
[02:41] <Eickmeyer[m]> mwhudson: Oh! Thanks!
[18:27] <mwhudson> oh boy the numbers on https://autopkgtest.ubuntu.com/running are not good
[18:41] <sarnold> oh my
[18:54] <tumbleweed> test all the things!
[21:27] <bdmurray> dbungert: Given that KillMode=process isn't recommended can you explain a bit more why this needed? Bug 1821412 for reference
[21:29] <dbungert> bdmurray: you may remember that I have branches that work and involve code changes.  Those successful ones all have in common that the main process sticks around until the report process is allowed to finish.  I mistakenly diagnosed problems with the report process to be a glib issue, but when I put together a simpler test case using glib but not systemd I saw no problem.
[21:30] <dbungert> the system crash dialog is being run by a chain of processes that ultimately is started by that systemd user service.  With the current code, the dialog process finishes first and leaves a child report process around.  systemd is cleaning up that child process and we don't want it to.
[21:31] <dbungert> we need the original process to go away or the gtk dialog stays around until the report process is done.
[21:45] <bdmurray> dbungert: okay, I'll upload it for jammy but the bug report could likely use more information for SRU'ing it.
[21:47] <dbungert> bdmurray: I agree, I not consider it SRU ready yet.  Was going to sort that out while the Jammy version was aging :)
[21:48] <dbungert> *I do not