Stultumanto beat me to it. Someone in IRC mentioned the same problem, so I just did a uefi install in qemu, and sure enough, I screwed up the first time and didn't boot uefi. When I did it right, the install worked and it installed grub. I still had to go through the uefi shell in qemu to get to the boot file, but it did boot. On a real hardware boot, there's probably a key to press that will show you a boot device menu, listing all the hard drives and usb or cdroms that are present. There might be two entries for the usb - one for legacy and one for uefi.

As mentioned, check to see that /sys/firmware/efi exists when you boot the live. There are other ways, to tell.

Note the color of the grub boot menu background. If it's blue, you booted in legacy bios mode. And it's a syslinux/isolinux boot menu, which looks different from a grub boot menu. When you boot the desktop-live in uefi mode, you get a black background at the boot menu.

More on desktops:

lxqt pulls in pulseaudio (like lxde does)
cinnamon (I had forgotten about this one) pulls in pipewire.
That leaves gnome to be checked. Someone else can do that, please.

Steve,

Thanks for the report. It took me a little while to find "switch user". It's only in the action button on the right side of the panel. It doesn't appear in the Logout window you get from the Applications menu. I can confirm that switch user doesn't work in the default setup.  Also, the who command returns nothing.

I believe you found another bug in the Slim display manager. If you install lightdm, then "Switch user" becomes an additional choice in the logout screen, and it does work. Also, 'who' works then. This problem wasn't in the betas because slim was creating a black window over the desktop with autologin, so I used lightdm in the builds instead.

I was able to reproduce the grub failure in virtualbox, but in qemu it worked normally. I'll stop here on that point - I'm suppressing a rant about grub.

Point taken about the "copy files" button. If you do a uefi install, the button says "Install bootloader". That's an easy fix.

Edit: Forgot to mention ssh. The desktop-live is based on a default task-xfce-desktop install which does not include openssh-server.

List of rust packages in devuan: https://termbin.com/xakg

I can't post the whole list here.

413 Request Entity Too Large

TLDR:
70430 packages, including virtual packages
3561 packages, excluding virtual packages

I'm not seeing that behavior on fresh install of excalibur or upgrade from daedalus to excalibur. syslog and kern.log are there. /var/log/messages has been missing for years. Make sure rsyslog and cron are installed and running.

I did an upgrade from daedalus to excalibur yesterday and the date on /etc/rsyslog.conf is February of 2023. Nobody has edited it.

I got this on a kde install of excalibur I did yesterday.

root      1421  0.6  4.0 880628 88088 ?        Ssl  17:43   0:01 /usr/lib/xorg/Xorg -nolisten tcp -background none -seat seat0 vt2 -auth /run/sddm/xauth_yIsirg -noreset -displayfd 16

I have /etc/sddm/wayland-session and /etc/sddm/Xsession and neither one looks like a config file to me.

It says you're missing free firmware. Make sure firmware-linux-free is installed. If you need non-free firmware, you won't be able to use a libre kernel. Have you tried booting the daedalus kernel? It should still be there.

The alsa error has a fix somewhere else on this forum. Search for alsa_restore and I think you'll find it. It's an easy fix.

Also make sure the video drivers are installed. I'm not sure which you need but there's xserver-xorg-video-amdgpu xserver-xorg-video-ati xserver-xorg-video-radeon

Yes, follow the instructions on the linux-libre site. Install their key, add their sources file, update, install.
You'll still have the devuan sources active and can upgrade packages. The kernel will come from the freesh sources.

Note: I pasted the 'modernized' sources below. It's possible to use the modernized source for freesh along with the older sources.list for the devuan sources. I just tested it and 'apt update' didn't complain.

Here's an example iso of 32-bit devuan with the linux-libre kernel. Refracta is just devuan with some tweaks and different set of packages. This one is old. I need to make new isos and stop calling them Testing.
https://get.refracta.org/files/testing/ … 5_1010.iso

/etc/apt/sources.list.d/devuan.sources (generated by running 'modernize-sources')

# Modernized from /etc/apt/sources.list
Types: deb
URIs: http://deb.devuan.org/merged/
Suites: excalibur excalibur-security excalibur-updates
Components: main 
Signed-By: /usr/share/keyrings/devuan-archive-keyring.gpg

/etc/apt/sources.list.d/freesh.sources

X-Repolib-Name: GNU Linux-libre Freesh Repository
Enabled: yes
Types: deb
URIs: mirror://linux-libre.fsfla.org/pub/linux-libre/freesh/mirrors.txt
Suites: freesh
Components: main
Signed-By: /usr/share/keyrings/freesh-archive-keyring.gpg

That screenshot might be NSFW for some people.

If you're going from bookworm to excalibur, you're doing an upgrade at the same time as the migration, so it's probably a good idea to look over the upgrade instructions. One gotcha on upgrade is that you must install the usrmerge package before you upgrade.  I don't think it matters if your sources are pointed to bookworm or excalibur for that package. (I've done it both ways.)

Upgrade to Excalibur
https://www.devuan.org/os/documentation … -excalibur

Next paragraph in the release notes starts with:

Installer isos and live CDs for amd64 are available for download

And this section:

### Reduced support for i386

There is no i386 installer iso for Devuan 6 Excalibur, following Debian's
choice to remove their official i386 kernel and installer.

And if you look at the debian release notes it explains that you can run 32-bit in chroot or container.

I believe it's still possible to upgrade a 32-bit system. You won't get a new kernel, so if you have the kernel metapackage linux-image-686-pae you might need to remove it first. I'm not sure about that.

You can use a linux-libre kernel if you don't need proprietary (non-free) firmware.
see: https://dev1galaxy.org/viewtopic.php?id=7208

I gave up using fstab to mount nfs shares a few years ago. I think it was because of problems if the server or the client rebooted. I use autofs now. It auto-mounts the share when you access the share. Simple config desribed here: https://dev1galaxy.org/viewtopic.php?pid=58917#p58917

I use these options in /etc/export. It looks like I added 'sync' either in daedalus or excalibur. (I don't have a daedalus to check right now.)

(rw,sync,no_subtree_check,crossmnt)