I think something you have changed on Miyo is meeting ceres (Debian sid) in a very bad time.
I am making this assumption because my one installation on metal is old miyo that evolved to ascii.
The newer pre-xtra miyo disliked ceres very much.  It just fell apart.  I think part of this is using
synaptic which used dist-upgrade on every upgrade.  I suspect the settings from Devuan affect
plain apt while synaptic is falling on Debian architecture plan.  I can't explain the vast difference
between apt upgrade and apt-get dist-upgrade on a straight devuan-ceres installation.  So I am
being very careful and selective what I allow to be upgraded.

Meanwhile Artix is "unbelievable" ... but did I already say this today?  Not in a competitve way
but just admiration of something a few weeks old to be that capable.  They even made a cript
for runit to run as a supervisor/parallel of openrc/sysvinit, a bit redundant but a security
measure for stability.  It still boots up like a bullet and gets me wifi connected on console which
I have had a terrible time in deb/vuan.

I had no idea that the testing and unstable of devuan was not screened by Devuan.  I could also not have imagined that buster was skipped all together and unstable is sid, while ascii is stretch.  So this makes it a further miracle that ceres is running trouble free since yesterday while ascii is still broken.  I do have old installations of buster and sid and neither have ever resembled the problem in the login screen freezing my 2 input devices.  Out of curiosity last night visited and kate and retext were destined to be removed there as well.
So it is beginning to make some sense now that you threw some light into the mystery.

Also no wonder the mixmatch of miyo installation fell apart when brought up to ceres.  I wonder what miyo maestro will find in his late endeavor.

Scratch this idea/mystery
Both dmsetup and libdevmapper are the same version on ascii now (looking at ascii now from ceres) but uuid is upgradeable from ceres to ascii
2.27.1 in ceres 2.28.1 in ascii ?????  No, it is not a typo

I don't see how uuid has anything to do with mice and keyboards freezing up on xorg.
But what is now in ascii may not be what was there yesterday and today 12hrs ago I did not see any updates.

------------------- edit this note so I don't add another ---------------------------------------------

Did check on ascii, nothing yet to upgrade and the suspected 3 pkgs are of the same edition on ascii and ceres.
So what the heck made a difference in ceres?  Only difference is the kernel I can think about, but this hasn't
changed recently in relation to the problem and after last update of Linux412 the problem was not affected.
While the problem started first in ceres, I kept switching DMs to see whether it was DM/login screen related.
So the only thing different is in ceres I had last stayed with LXDM, instead of my usual lightdm.
So I installed lxdm in miyo-ascii but it did the same thing there too as it did on lightdm.

Note:  It only freezes up when the DM/login screen "first" comes up.  If I logoff to login to another account
the mouse and keyboard never lock up a second time.  So it all happens when X starts and passed control to
the DM.

Also, while on Openbox, the logout breaks up the dm, and as openbox is shut-down and no longer exists,
I am stuck on a half broken session where a terminal and other windows can still be initiated (like task
management).  If I then kill the lxdm-session then it exits out to the DM/login screen.

I guess I will have to wait for September when the dev-dev-team returns from vacating

So who comes up with 3 massive upgrades a day in ceres?
I am not being sarcastic, but have the non-vacationing americans taken over?  ;D

https://dev1galaxy.org/viewtopic.php?id=1534

As I wrote on the above, due to updates not my own involvement, my problem with freezing mouse/k-b has been magically cured in ceres.  Still exists on ascii.

Update of my problems with xorg freezing mouse and key-b.
Since yesterday's update the next logon did not have a problem in ceres
Ascii still has the problem now.  I have intentionally not tried much of anything to see
what update fixes what.  So now I have to wait till the specific update rains down
on ascii.  This makes this problem a week long, and appeared mysteriously on
ceres first and 2 days later on ascii.

Whatever the dev1 technicians are changing, for me at least, it seems to be working.
I wondered if it is linux-kernel related and in an arch based distribution I had tried
Linux4.13 in its early 2 steps and the mouse would freeze.  I never tried then to
unplug it and plug it back to see whether it gets fixed.  But I am wondering whether
a patch to fix something else would break something different and it flowed down
the kernel development assembly line.

People who haven't used xfce/lx*/ob find it illogical (and maybe it is) why wouldn't a user without root rights NOT be able to browse and search available and installed software.  It maybe the gnome based, or kde, or mate, that only ask for pass when you attempt an upgrade or installation.
People have used linux for years through a desktop and have no clue where and which programs without a gui exist in their own system.  They only see what is on the menu.
If you try $apt -search nunit  do you get a result?  Should you?

you are king openboxer
how d'ya get lxpanel without the rest of the desktop?  Did you manually delete the rest of the stuff and left lxpanel and dependencies?
I bet the same questions applies about xfcepanel

On the last miyo 64 on pcmanfm the list of partitions DO come up on the left column, which I haven't been able to get done on my ascii or ceres
Devuan openbox.  If I log into lxde they work.
But the list is there but even with complete sudo rights none can be mounted.
Now even on lxde if I run sudo pcmanfm the list does not come up but if the user has mounted them then root has access in the /media/user/** mounted volumes.
I know I had asked before and I am still researching this problem. 
eudev is now on all my installations but the problem pre-existed, it is not related.
All the necessary gvfs tools are installed.

/etc/systemd/system/hook-for-my-shutdown-jobs

I wonder on a commercial server, that RH services, do they charge for removing this "feature"?

Hey, watch it with those gray beard jokes.

I always liked AMD but my last precious all AMD machine was stolen.
I only buy what I can get between 50-100 and I am happy.  Usually
a refurb Dell with a year warranty.

$ lspci -tv
-[0000:00]-+-00.0  Intel Corporation 82Q35 Express DRAM Controller
           +-01.0-[01]--
           +-02.0  Intel Corporation 82Q35 Express Integrated Graphics Controller
           +-02.1  Intel Corporation 82Q35 Express Integrated Graphics Controller
           +-03.0  Intel Corporation 82Q35 Express MEI Controller
           +-03.2  Intel Corporation 82Q35 Express PT IDER Controller
           +-03.3  Intel Corporation 82Q35 Express Serial KT Controller
           +-19.0  Intel Corporation 82566DM-2 Gigabit Network Connection
           +-1a.0  Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #4
           +-1a.1  Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #5
           +-1a.7  Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #2
           +-1b.0  Intel Corporation 82801I (ICH9 Family) HD Audio Controller
           +-1c.0-[02]--
           +-1d.0  Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #1
           +-1d.1  Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #2
           +-1d.2  Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #3
           +-1d.7  Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #1
           +-1e.0-[03]--
           +-1f.0  Intel Corporation 82801IO (ICH9DO) LPC Interface Controller
           +-1f.2  Intel Corporation 82801IR/IO/IH (ICH9R/DO/DH) 6 port SATA Controller [AHCI mode]
           \-1f.3  Intel Corporation 82801I (ICH9 Family) SMBus Controller

Miyo Linux and Vuu-do video presentations