I have removed haveged from several Devuan systems with no ill effects.
Haveged was made  default in the Debian Distro to overcome performance issues after a change to systemd using random to create unique ids.
There was a bug report somewhere if anybody cares to follow up and search for it, of course systemd was not at any point at fault.
Despite having restricted resources (cpu,memory) havaged was even carried forward onto the Raspberry Pi which has a 'real' number generator and therefore can never run out of random numbers.

NB. for real randomness security use a Pi, not your £2K workstation.

@spliskin

I read that as "rust that causes memory corrosion"

Amazing how the subconscious mind tries to warn us of the truth.

@gdstew, either I have misunderstood your reply or you have borked your private network configuration. If you want to fix it I am willing to try but I will need a full breakdown of your current network topology. Is there a reason you are not using the Laptop's and Raspberry Pi's built-in Wifi ?

You cannot assign 10.1.10.1 to both eth0 and wlan0.
If you want your wireless network to be a continuation of your existing ethernet network then you need to bridge.
Or
Make a new subnet for your wireless network e.g. 10.1.11.1/255.255.255.0
You will also need a DHCP server for this new subnet and routing configured.

Nope. I am wrong, key management is possible in ad hoc.
For AP mode go with the 'hostapd' package ('apt show hostapd')

Outputs are capable of carrying digital sound as well as video, so consequently show up in ALSA as individual sink devices.

ls -ld /proc/asound/card1/pcm*
dr-xr-xr-x 4 root root 0 Nov 18 13:07 /proc/asound/card1/pcm3p
dr-xr-xr-x 4 root root 0 Nov 18 13:07 /proc/asound/card1/pcm7p
dr-xr-xr-x 4 root root 0 Nov 18 13:07 /proc/asound/card1/pcm8p
dr-xr-xr-x 4 root root 0 Nov 18 13:07 /proc/asound/card1/pcm9p

@greenjeans, like'ish your script...wont work for me, running a corner-case here

Suggestions...
Have a default when 'sed' fails/ .asoundrc is empty ore doesn't exist.
Have the window size itself to the number of entries in the list (its annoying to have to keep resizing to make a selection).

@stultumanto, I have learnt from your summary, thanks for posting in detail how you resolved it.

Here is another concerning trend

git grep -l '\<systemd\>' linux-stable.git/Documentation/ | wc -l
23

Twenty three kernel documents referencing (IBM/RedHat's lets not forget the commercial interest source from whence) systemd.
As far as I am aware only one would be legitimate, SYSFS.

@ralph.ronnquist, the mdadm report looks normal to me (example from Daedalus of four SATA disks raided in pairs)

mdadm --detail --scan
ARRAY /dev/md1 metadata=1.2 name=localhost:1 UUID=bbfe098d:4d1afad0:46f8d336:6e554bce
ARRAY /dev/md0 metadata=1.2 name=localhost:0 UUID=ad7eb025:c0a54024:836db957:a0985a5b

@fuchur
Is there a specific reason you are raiding individual partitions ?
Performance will suffer.

check /proc/mdstat to confirm raid integrity '[UU]' (healthy example from Daedalus)

cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdd[0] sde[1]
      499975488 blocks super 1.2 [2/2] [UU]
      bitmap: 0/4 pages [0KB], 65536KB chunk

md1 : active (auto-read-only) raid1 sdf[0] sdg[1]
      499975488 blocks super 1.2 [2/2] [UU]
      bitmap: 0/4 pages [0KB], 65536KB chunk

unused devices: <none>

These two issues are serious and should be addressed

In my experience USB device failures can cause other errors especially when a power cycle is triggered.
Disconnect this device if you can.

Cannot be helpful without more to go on, sorry.

@stultumanto exactly, it has taken me a long time to learn how to not answer ambiguous questions.