| gnu_srs1 | No, the bug was closed #1056980 with wontfix. | 00:05 |
|---|---|---|
| rrq | netcat-traditional maintainer is Anibal Monsalve Salazar <anibal@debian.org> | 00:06 |
| bgstack15 | Thanks to whatever chuckleheads run Debian, I've just chosen to use netcat-openbsd instead. Hopefully it works for basic "nc -v remotehostname" | 00:13 |
| bgstack15 | Man, Debian was the last sane base, but now it's been too redhatized too | 00:13 |
| fsmithred | I just noticed another difference between my two ceres with usrmerge. The one that works (upgrade from daedalus) is amd64 and the broken one (from an unreleased devuan-live ceres iso) is i386. | 01:28 |
| HimeHaieto | i386? shuuun!! | 01:29 |
| HimeHaieto | <- still has her primary laptop running i386 | 01:29 |
| fsmithred | I tried reinstalling the kernel, but update-initramfs fails for 6.6.9. Works fine on 6.5 | 01:30 |
| fsmithred | Shit. The one that works has no dbus. I just realized I need to "fix" that to see if it breaks it. | 01:37 |
| fsmithred | and task-xfce-desktop | 01:39 |
| rrq | fsmithred: debootstrap with "--variant=minbase --merged-usr ceres" works fine and result in a "usrmerge" system... without -merged-usr also works fine and results in a normal system | 06:31 |
| rrq | (today) | 06:31 |
| rrq | likewise for debootstrapping excalibur | 06:33 |
| rrq | you may also handle "/usr" pathname nonsense by creating the directory beforehand and add a link "usr->." to it | 06:35 |
| rrq | with that you bcan have o not have a "/usr" prefix on any pathname | 06:36 |
| bb|hcb | That would be rootmerge? ;) | 07:26 |
| rrq | yes, let's name it! .. bound to attract followers :) | 07:37 |
| bb|hcb | I hope that no one will misunderstand the joke and start promoting that :P | 07:50 |
| fsmithred_ | rrq, which arch did you debootstrap? | 10:43 |
| rrq | amd64 | 12:03 |
| rrq | ah, right.. you had i386 | 12:04 |
| rrq | (trialling i386 with multi-arch) | 12:06 |
| rrq | I'm running this on a daedalus amd64 with debootstrap=1.0.128+nmu2devuan2 | 12:10 |
| rrq | excalibur and ceres also fine for --arch=i368 | 12:10 |
| rrq | all using "--variant=minbase" seems to be the crunch; without that it fails with that cron-daemon-common issue | 12:17 |
| rrq | (minbase doesn't include apt ?) | 12:54 |
| fsmithred | apparently not. | 13:03 |
| fsmithred | I tried to do a minbase on top of the failed one, but when I chrooted, I couldn't use apt. I tried dpkg to install all the debs in /var/cache/apt/archive, but I just canceled it. | 13:04 |
| fsmithred | I'm going to try without the minbase and exclude cron-daemon-common | 13:04 |
| sakrecoer | First Devuan meet of the year, today January 4th, 2024 @ 20:30 UTC \o/ | 13:05 |
| sakrecoer | Pad is here: https://pad.dyne.org/code/#/2/code/edit/oAv2YR1Ax+7wjdB8Yg3upjiq/ | 13:05 |
| sakrecoer | Meet is here: https://open.meet.garr.it/devuan | 13:05 |
| rrq | fsmithred: "--exclude=logrotate,cron,cron-daemon-common" gives a successful debootstrap without --variant=minbase | 13:32 |
| rrq | and having that plus "--include=apt" gives a goodish --varint=minbase filesystem | 13:33 |
| rrq | .. and its rsyslog=8.2312.0-1devuan1 that recommends logrotate that depends cron depends cron-daemon-common .. causing the gried | 13:37 |
| rrq | f | 13:37 |
| rrq | Maintainer: Devuan Developers <devuan-dev@lists.dyne.org> | 13:39 |
| rrq | i.e. changing rsyslog to merely suggest logrotate would bypass the issue. | 13:41 |
| rrq | /names | 13:50 |
| fsmithred | rrq, does that mean we can't have logrotate? | 13:55 |
| rrq | can't have cron without sorting the issue | 13:55 |
| rrq | and logrotate needs cron for its function | 13:56 |
| rrq | though doesn;t rsyslog rotate logs without using logrotate? | 13:56 |
| rrq | of course "cron" is quite important in itself | 13:59 |
| rrq | the debootstrap problem arises because cron pre-depends on cron-common-daemon, which depends on systemd-standalone-sysusers, but that is set to be installed *after* cron during debootstrap | 14:03 |
| rrq | i.e. it's down to a weakness (bug?) in debootstrap's dependency handling | 14:04 |
| onefang | Did you try the same thing with mmdebstrap? | 14:07 |
| onefang | From the description "In contrast to debootstrap it uses apt" so might be better at sorting out dependencies. | 14:09 |
| rrq | otherwise with debootstrap, it needs a) include systemd-standalone-sysusers, then as it breaks, b) manual dpkg -i for systemd-standalone-sysusers and libcap2, folllowed by c) manual dpkg --configure cron-dameon-common, and d) manual dpkg -i -E -R var/cache/apt/archives .. to complete the install | 14:19 |
| rrq | and if you haven't had fun enough, you should start that debootstrap session without --merged-usr (or with --no-merged-usr), which also leads to a happy ending, although without any usrmerged applied... | 14:23 |
| * rrq bedtime | 14:30 | |
| fsmithred | onefang, thanks for the tip. mmdebstrap works. I forgot to add --merged-usr and it looks like I successfully installed usrmerge in the chroot. | 15:46 |
| fsmithred | I still have a few things to do to see if cryptsetup will be happy. I guess I haven't had enough, because I'm installing into an encrypted partition. | 15:47 |
| Nietz | regrets will be unable to make meet. may i add the primer for an idea i have and we can catch up later? | 20:48 |
| golinux | Nietz: Of course | 21:29 |
| plasma41 | Meeting time | 21:30 |
| golinux | Anybody coming??? | 21:41 |
| Nietz | was in appointment. still open? | 21:52 |
| golinux | Yes. Seems you couldn't connect . . . | 22:05 |
| fsmithred | I had a look at DevuanPup daedalus iso. It uses devuan repos, but it only has main enabled. -updates and -security are commented. | 22:50 |
| fsmithred | In addition to that, it only has a few of our forked packages. | 22:50 |
| fsmithred | ii fdisk 2.38.1-5devuan1+b1 amd64 collection of partitioning utilities | 22:51 |
| fsmithred | ii gir1.2-udisks-2.0:amd64 2.9.4-4devuan1 amd64 GObject based library to access udisks2 - introspection data | 22:51 |
| fsmithred | ii libudisks2-0:amd64 2.9.4-4devuan1 amd64 GObject based library to access udisks2 | 22:51 |
| fsmithred | ii udisks2 2.9.4-4devuan1 amd64 D-Bus service to access and manipulate storage devices | 22:51 |
| fsmithred | so I don't really know what it is. | 22:51 |
| fsmithred | And I didn't download it locally, so I can't boot it easily. | 22:51 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!