HAOS refusing connection irregularly but daily

Hello,
I have HA installed for about a year now and the connection to the frontend was always reliable. So far I didn’t find the time to dive in deep, but I really liked the overall experience and the progress in the development! Thanks to all envolved!

I added a NEWS section below

My problem:
For about 2-3 weeks the connection to the frontend has become very unstable. Sometimes it just works normally, sometimes I lose the connection while being connected (via Browser or App) and sometimes it’s just unreachable for hours.

So far I didn’t manage to find a pattern.

What I checked:
The IP-address is: 192.168.178.11

ping 192.168.178.11 works

192.168.178.11:4357 works

Home Assistant observer sees no issues

I had a look at System → Logs when it was accessible but didn´t see anything that seemed related, but also didn’t know exactly where to look.

When I first discovered the problem I tried to restore a backup which was 3 days old which didn’t solve the problem. Didn’t want to go further back than that, since I had just finally managed to integrate my solar DTU before and didn’t want to lose the data.

What can I do? Can someone hint me in the right direction? Thanks!

NEWS
Just after finishing this post I was able to connect again and I saw some error message about the backups. Turns out the thumb drive attached to the router was full. After fixing this HA was accessible for several hours in a row and I already thought I found the culprit but: No!

I checked ha net info via ssh:

docker:
  address: 172.30.32.0/23
  dns: 172.30.32.3
  gateway: 172.30.32.1
  interface: hassio
host_internet: true
interfaces:
- connected: true
  enabled: true
  interface: enp0s18
  ipv4:
    address:
    - 192.168.178.11/24
    gateway: 192.168.178.1
    method: static
    nameservers:
    - 192.168.178.11
    ready: true
  ipv6:
    address:
    - fe80::8b7b:39e5:ae72:6fc8/64
    gateway: null
    method: disabled
    nameservers: []
    ready: true
  mac: 02:21:D3:xx:xx:xx  
  primary: true
  type: ethernet
  vlan: null
  wifi: null
supervisor_internet: true

but the CPU activity got up, just before frontend got unreachable:

The Setup:
I have ProxMox 8.3.2 running on a Dell Wyse ThinClient. HAOS is the only VM running and (to me) it’s seems to run normally: