Installing Home Assistant OS using Proxmox 8

I’m struggling over the “GUEST AGENT NOT RUNNING” problem, with HAOS vm ramdomly stopping to work after 1-5 days of perfect Uptime. After this problem appears, the only solution is to physically disconnect the NUC and restart the device (It’s not possible to restart from proxmox itself).

I’ve tried several things over this past weeks. but the problem still appears every 2-3 days (and when I am out of home for more than 2-3 days, it becomes unusable as I need physical access to the device)

  1. I’ve reinstalled again following this guide
  2. i’ve manually installed qemu-guest-agent as suggested before.
  3. I’ve tried disabling every other VM on proxmox
  4. I’ve manually reinstalled proxmox itself multiple times
  5. I’ve tested different file-systems (ext4 vs zfs)

The problem is still there. Guest Agent is Enabled. DHCP is enabled on the router

I would really appreaciate any help as I’m really tired of trying to fix this and I’m seriously thinking about going back to the good-old raspberry pi which simply worked :frowning:

Tell us your machine specs.

It’s exactly this one.

NiPoGi AK1 PRO Mini PC Desktop Windows 11,11th generation Intel N5105 Mini computer(up to 2.9 GHz),Micro PC 8GB DDR4 256GB ROM with dual display,4K HD,Dual WiFi,BT4.2 : Amazon.co.uk: Computers & Accessories

I’ve used your script to install HAOS, thanks for your help.

N5105

Installing Home Assistant OS using Proxmox 7 - #786 by tteck

Installing Home Assistant OS using Proxmox 7 - #790 by tteck

Hello @tteck! Just to let you know that changing the Kernel to v6 solved all of my problems!

Doing a simple apt get for the Kernel, as shown on your post, did not worked, I had to use your Kernel script to changed it to V6, working great now. Even the “memory leak” is now fixed!

Thanks again and have a great day!

2 Likes

I Just wiped my old MSI GE60, installed latest ProxMox via same “old” script at the top here, and installed HA aswell … so far so good, not much time spend …
Then i close the lid (after dedicated this laptop to ProxMox) , and proxmox goes into sleep, and with that ha :frowning:
… or atleast both become unavalable
Anyone know howto fix this , i doubt i have features in Bios for controlling the lid, and can’t find anything in Proxmox

Here’s the fix you’re looking for:

nano /etc/systemd/logind.conf

[Login]
#NAutoVTs=6
#ReserveVT=6
#KillUserProcesses=no
#KillOnlyUsers=
#KillExcludeUsers=root
#InhibitDelayMaxSec=5
#HandlePowerKey=poweroff
#HandleSuspendKey=suspend
#HandleHibernateKey=hibernate
HandleLidSwitch=ignore
#HandleLidSwitchExternalPower=suspend
HandleLidSwitchDocked=ignore
#PowerKeyIgnoreInhibited=no
#SuspendKeyIgnoreInhibited=no
#HibernateKeyIgnoreInhibited=no
#LidSwitchIgnoreInhibited=yes
#HoldoffTimeoutSec=30s
#IdleAction=ignore
#IdleActionSec=30min
#RuntimeDirectorySize=10%
#RemoveIPC=yes
#InhibitorsMax=8192
#SessionsMax=8192

Save and exit the editor with “Ctrl+O”, “Enter” and “Ctrl+X”

systemctl restart systemd-logind.service

2 Likes

To receive better assistance, consider creating a new topic instead of continuing to post off-topic comments in this thread. This thread should remain focused on installing Home Assistant OS using Proxmox 7.

1 Like

Thx , fwiw, further config can be found in other topics:

Hi,

I am a bit lost at the moment and would appreciate any help :slight_smile:
Just moved my homeassitant from a raspi 3 to a NUC using Proxmox 7.3.4 and your install script (which is pretty awesome btw.).
Everything works fine, but after 4-5 hours I am not able to switch on/off my shellys or set the climate temperature and so on. Strange thing is that the status of the shellys is still updated and displayed correctly in hassio. Also all of my zigbeedevices and nodered flows etc. are working smoothly…

After rebooting the whole VM everything works fine for another 4-5 hours. In the first hours I am able to review the supervisor logs, after the “crash” I am not able to open the logs anymore. My guess would be that there seems to be a problem with the supervisor but I have no clue what to do. In the VM I can see that the allocated 4GB RAM are constantly fully taken.

When I move back to the old raspi 3b+, everything works fine with no problems…

Thanks for any hints and support

Here are my specs:

Proxmox 7.3.4 with only Homeassistant running in a VM
NUC-Kit NUC6CAYH, 4 x Intel(R) Celeron(R) CPU J3455 @ 1.50GHz, 8GB DDR3 RAM, 512SSD
Kernel: Linux 6.1.2-1

From the Homeassistant logs:
supervisor
supervisor_log

Proxmox specs:


Try increasing RAM

EDIT: is this OS 9.5?

wow, you’re fast.
I’ll give it a try with the RAM

Yes its 9.5 and Supervisor 2023.01.1

Increasing the RAM wasn’t the solution :confused:
Still encountering the same Problems…

And You get “request timeouts” + Bad gateway , when your on the “new” Proxmox … Sounds like you haven’t “transferred” your network-settings correct

I would agree if it wouldn’t work perfectly for 4-5 hours before it runs into some kind of network errors. Or am I wrong with this?

Most likely your logs + events are building up , until it hangs and crashes , so monitor your homeassistant.log and homeassistant_db

I think I will rebuild the DB and give that one a Chance :slight_smile:

This is a great guide, worked perfectly for me, so thank you for taking the time to make it.

One question: this script seems to allocate 32 GB of disk space for the virtual machine - is there any way to reduce it? I would like something more like 16 GB.

On line 320 of the script there is a number of 32 GB hardcoded:

SHOT6

I tried to run a version of the script with this number manually edited to “16G”, but the VM that resulted was still 32GB. Thanks!

If I remember near the start when running the script it asks if you want standard settings (or something like that), say no and it may let you change them in there, it lets you change memory size i know as I did.

By the way 32gb is the recommended size. so may not be wise changing it although my setup does not get anywhere near 32gb.

The minimum disk size for the .qcow2 disk image provided by the Home Assistant team is set to 32GB and cannot be altered during the virtual machine creation process.

3 Likes