I can try that with new VM. Do you think HAOS 8.3 and latest pve kernel will solve this ?
Have no idea. This is the first time I’ve heard about a possible kernel issue. Good investigative work though.
I have tried everything I could think. there are some posts on forums for possible problems with pve kernel above that I posted earlier so I noticed that when I did new proxmox install with my old downloaded proxmox (after I messed up something) everything was normal. when tried to upgrade proxmox HA restarts came back. Reverted to old kernel fixed the issue.
I will try latest kernel and then HAOS 8.3 when I have time to play.
Thank you @tteck !
1 Like
I also downgraded the PVE kernel to the last 5.13 and still had these types of issues. In my case the VM is/was only rebooting/hanging after a longer time usually. I’m back on 5.15 currently.
I tried a lot of different things already and so far the only situation without any of this is where I’m only using n-1 cores for all containers/VM.
Now using 3/4 cores (1 for Adguard container, 2 for haos VM) and I guarantee you if I start my homebridge container (1 core) that haos will crash after a few hours 
In the rare case where it hangs, I get somewhat these errors: rcu: INFO: rcu_sched self-detected stall on CPU | Proxmox Support Forum
If it reboots, nothing usable in the logs. Not ruling out that the reboot and crash are two different problems.
I’ve been using Proxmox edge kernels, running 5.18.10-edge now.
Some have suggested that the edge kernel has fixed all of their 5.15 issues.
I’m using it on one server, I like that it’s keeping up with new kernel releases instead of LTS.
EDIT: There’s a Edge Kernel Tool
2 Likes
Installed PVE Edge Kernel and HAOS 8.3… fingers crossed
Edit: still running after about 35 hours…
1 Like
is anyone having random restarts installing using this method? HAOS restarts but proxmox stil shows that it is still up
That’s normal, Proxmox does not know what is going on “inside” your VM unless you add stuff so the two talk to each other. Proxmox is not a method to check if Home Assistant is “running” to me.
For the random reboots and crashes, I had the same, so I installed Home Assistant OS 8.3 (VM) and switched to the Edge Kernel (Proxmox). Have not had a crash since 5 days now after doing this.
Edit:
To update to 8.3, in the VM shell: ha os update --version 8.3
To switch to PVE Edge Kernel: run script found at Proxmox Helper Scripts | Proxmox Scripts For Home Automation
Just accidentally rebooted to get you this information 
Thanks a lot, I found the guide from the previous few posts. Will monitor.
Both are using the Intel Celeron N5105 Processor
Hello, I have the problem that you are dealing with in the last messages. I have a VM with HA OS, core version 2022.6.7, operating system version 8.2. If I update to core version 2022.7.6, the HA VM loses connection and cannot access my instance in any way. I have tried updating the operating system to version 8.3 and updating to the latest version of the core, with the same result, the VM keeps losing the connection. I have installed kernel 5.18 edge, along with operating system 8.3 and core 2022.7.6, also with the same result. I’ve even done a cleanup of old kernels. I have loaded a backup with HA core version 2022.6.7, operating system 8.2 and the kernel I have returned to 5.15 and everything works fine here. I don’t know what else to try, my knowledge is not very extensive. My proxmox is installed on a NUC i3. Sorry for my English, I’m using a translator.
That sounds like a totally different issue. May wanna start a new topic?
Home Assistant OS 8.4
- Linux: Update kernel 5.15.55
Changed
-
Home Assistant OS VM
- Present the drive to the guest as a solid-state drive rather than a rotational hard disk. There is no requirement that the underlying storage actually be backed by SSD’s.
- When the VM’s filesystem marks blocks as unused after deleting files, the SCSI controller will relay this information to the storage, which will then shrink the disk image accordingly.
Thanks for these updates!
Are these settings we can manually change in an existing installation or does this require a reinstall?
1 Like
Click on VM > Hardware > Hard Disk > Edit
enable the following
Then reboot VM. I don’t know if previous unused blocks will be removed.
Changed
-
Home Assistant OS VM
- Set the real time clock (RTC) to local time.
- Disable the USB tablet device (save resources / not needed).
1 Like
Edge only was not fix for me. 8.3 and 8.4 did the job. I am stable now too!
1 Like
v4 Script - Whiptail menus
The goal is to get input from the user in a friendly way.
3 Likes