Yes, it is the version that I updated to that broke my second profile.  I removed that version and installed the previous version and my second profile works as expected.

The version that works correctly is

I do not know.  On my system, it listed a total of about 7-8 devices.  All but two of them had the hardware names of the device so it made it easy to know which two were the keyboard and mouse.

The folder names under /sys/bus/usb/devices/ is correct and normal, but I don't have a clue what is causing that small white square under the command.  Is grep installed on that machine?  If not, I would think that it would throw an error.

As root, run this command to see if grep is installed.

The command that did not work, did it produce any output at all?  I used these commands as user, but I don't think it would harm anything to run this as root.  I ran the command on my system as root and it produced the same desired output. 

I have used this on several systems - desktop, laptops, etc. - and it always listed all usb devices so that you could use the exact information to add to the rc.local file.  For example, the information that you added to the rc.local file does not look right.  For example, my usb devices are not listed as usb1, usb2, and so on, but rather, it lists them as 1-5, 1-6, etc. And from this I can choose the keyboard and mouse devices from the others that will not be used to wake the system up.

Granted, this is mentioned to wake from suspend and hibernate, but I have also used it to fix a situation just like yours - waking the monitor up.

No, it is not necessary on a vps.  I have never installed a desktop (kde, xfce, mate, etc.) on a vps.  This is up to you on how you want to manage it - either from a command line or a panel, like webmin.

I still would like to know what is causing the problem, but right now I can not think of anything else to try.  I have done this migration many times and it always worked. 

The only thing that I can think of is that maybe you missed a step in the migration path, or there may be some kind of apt pinning.

This will check for any left over apt-pinning.  What is the output of the command:

You might want to try to reinstall dbus.  The following command will simulate a reinstall without doing anything.  It will only report what it will do if you run the command without the -s flag.  It will show you which package will get reinstalled, either the devuan or debian version.  To actually reinstall dbus, run the command without the -s flag.

You quoted the following:

The following packages have unmet dependencies:
 dbus : Depends: libdbus-1-3 (= 1.12.24-0+deb11u1devuan1) but 1.12.28-0+deb11u1 is to be installed
E: Unable to correct problems, you have held broken packages.

This is not correct for chimaera.  This is what my chimaera versions show:

root@Host:/home/nixer# apt policy libdbus-1-3
libdbus-1-3:
  Installed: 1.12.24-0+deb11u1devuan1
  Candidate: 1.12.24-0+deb11u1devuan1
  Version table:
 *** 1.12.24-0+deb11u1devuan1 500
        500 http://deb.devuan.org/merged chimaera/main amd64 Packages
        100 /var/lib/dpkg/status

If we are unable to make this migration work, you might want to inquire about another method of getting devuan on you vps.
- reinstalling debian on the vps and try the migration again
- re-install the vps using a custom.iso of devuan.  This could be a netinstall disk, or a custom build.  You will need to check with the vps provider and see if this is possible. 
- keep what you have got and rsync a local build onto the vps.

Edit:  changed apt reinstall command from dbus to libdbus-1-3

Try this,

and then try

Edit:  You might need to use this command to install devuan-keyring.  This was mentioned in the migration article but you may have overlooked it.

Once the keyring is installed, you might want to read again the migration article to make sure that all steps have been completed in the proper order.

tyrlak,

What is the output of

What is the output of

-and-

What is the output of

What is the output of

This is enough to get started.  I am wondering if the dbus packages mentioned in the error are installed?  And like golinux said, please use

code

-or

tags around the terminal informations.  It does make your information easier to see and read.

Edit:  After posting this, I ran the "apt policy" commands above and found that these two missing packages default-dbus-session-bus and dbus-session-bus are not found in the chimaera repos, but they are in my ceres repo.  I am interested in what the contents of your /etc/apt/sources.list will reveal.

Also, before doing something drastic, rename the /home/USER/.config/xfce4  folder to something like xfce4.bak.

Then try to restart xfce again and see if it recreates that folder and its contents.

I found some debian packages in daedalus-backports.  But warning... by default it does try to pull in systemd-standalone-sysusers, and I would strongly urge you to pin this out so that it will not get installed.  There is a non-systemd package called opensysusers which will install in its place if the systemd* packages are pinned from getting installed.

For example, this is my apt pin in /etc/apt/preferences.d/00nosystemd:

Make sure the daedalus-backports repo is active in sources.list.

# apt policy incus

# apt install -t daedalus-backports incus

I have not installed, nor do I know if these packages will work.  I see that systemd is a "suggested" package, but not a dependency.  Let us know how it goes.  Thanks.

Edit:  grammar and spelling.

Sorry for taking so long for reporting back, but this took a while. 

I did a physical install on a 13 year old dell server.  I made no changes at all, except for a user account name change and password change during installation.  I simply installed and tested.  My testing was centered on simply booting up, logging in, and rebooting into the next init. 

After installing all inits (boot options) worked fine, except s6-66.  My partition scheme is:

sda is a msdos master boot record, not gpt.
/  is /sda1
/home is /sda2
swap is /sda3
/mnt/data1 is sdb1

The first boot into s6-66, I got the same ModemManager error as reported above:

The system was not in a usable state so I was not able to view the log files, either locally or with ssh.  The second boot attempt I did not see that error and I got to the lightdm login screen.  The regular user was not able to login.  A login attempt with the correct user/pass just refreshed the page.  However, the root user was able to login.  Further checking:

ls -a /home

 
showed the entire /home folder as empty.  The /home partition did not get mounted at boot.  A manual mount of /home did work and then I was able to login.  I searched the log files and I found no mention about a mount error.  My fstab is:

I did notice that the s6-66 reversed the partitions sda -> sdb, sdb -> sda.  This was as far as I got with s6-66 as the third and 4th boot attempts produced the same modemmanager error. 

So, to recap, the s6-66 method did not mount the /home partition correctly at boot on the couple of times that it did work correctly.

If you can tell me what to check I will be glad to test further.  But without an error notation in a log file, I don't know where to start and I am very unfamiliar with inits.  I am not a very good tester, but I did have some time to test.

Testing was done from within virtualbox after installing, not a physical install on a hard drive, using a devuan host system - ceres (unstable)

At boot:

grub default selection - It worked fine the first boot, and maybe the second, and then it didn't.  When it did not successfully boot -

I logged in as root and  ran

"service lightdm start" will start the X session (lightdm) from the physical console of the virtual machine.  I successfully started Xfce by logging in as the user and entering "startxfce4".  I made some text bold to make it stand out, as it raised some questions with me.

While it was not successfully booted into X, I ssh(ed) into the machine and got this information:
/home/user/.xsession-errors

/var/log/lightdm/seat0-greeter.log
# cat /var/log/lightdm/seat0-greeter.log

/var/log/syslog
noting mentioned, warnings or errors regarding xorg or lightdm

Just to document what package if any had systemd mentioned -
root@init-diversity:/home/craig# dpkg -l | grep systemd
ii  dhcpcd   1:9.4.1-24~deb12u3     all  DHCPv4 and DHCPv6 dual-stack client (init.d script and systemd unit)

So, sometimes it works, sometimes it does not and I can't report what is wrong.  I thought the default selection was sysvinit, but the sysvinit selection works fine.  I have not checked the grub entries for any difference between the two selections.

grub sysvinit selection - it worked fine.
grub s6-rc selection - it worked fine.

grub s6-66 selection - fails to boot.
boot hangs with the message on screen that:

A virtualbox driver issue perhaps?  Repeated attempts with this method does the same thing with the same message.

grub runit selection - it worked fine.
This was my first time using runit and was very impressed with its speed and smoothness.

grub openrc selection - fails to boot.
I get the same or similar error messages as using sysvinit.
/home/user/.xsession-errors

cat /var/log/lightdm/seat0-greeter.log
(entire log, nothing excluded)

On the physical VM screen, it hangs at:

This or whatever it is trying to load next, but is not displayed yet, may be the problem.

My changes to the default installed system:
Removed vim and firmware-libertas firmware-myricom firmware-netxen firmware-qlogic ti-connectivity firmware-atheros firmware-bnx2x firmware-brcm80211
Installed gtkhash thunar-gtkhash libb2-1 geany autofs

I will be glad to assist in troubleshooting.  Please let me know.  This is an easy to use way to test and experiment with multiple init systems within the same install.  This is a very interesting concept.

Edit:  corrected typing mistakes.

This page shows what the sources.list should look like in daedalus, and this is only showing the main repo. 
https://www.devuan.org/os/documentation … o-daedalus

You may want to add the additional options like adding "contrib non-free non-free-firmware" to each line.

But at any rate it should still show up as it is there with "main".   

Have you run "apt update" before looking for it? Or, are you using synaptic?  A better question is, how are you trying to look for it and install it, with a terminal using "apt" or with the gui program synaptic?  If using apt you must might first want to run "apt update", and if using synaptic, click on "Reload" first, before searching.

Hope this helps.