Migrated from Core (Dockerized) to Supervised - errors

Hi,

I’ve searched for incompatibility settings when migrate from a core in docker to supervised. Cannot say that I’ve found so many posts that do what I’m doing, mostly it is the reverse to a core installation.

My installation is a Proxmox version with VM which hosts the Home-assistant supervised installation. I did running it for a week before migrate the actual installation and I also moved some other docker containers that I did not want to or could have in the supervised setup. All of this was well working and even all the mqtt and influxdb container move did work very well before the actual core migration.

However, I did a backup of my /config incl all . folders (hidden folders) and removed all files (copy to another folder) at supervised /config folder. When started up, there was some warnings but not errors directly. Found out that the http: setting had to be altered since using supervised version (watchdog related errors). I updated to latest version (2020.12.7 ) on both installs before migrate to make sure the config was working.

So to my issue. I’m not sure if it is the combo with proxmox or the home-assistant version or what. but I get major issues with stability. Sometimes the installation has just halted or freezed, sometimes there is automation that has been working for ages that does not work anymore (did work before migration) and other very basic things.

When I look at the deconz container that is hosted in another vm/docker at same proxmox host, that works well and respond etc so I can be quite sure that the proxmox working quite well or at least does not have own issues, rather impacted by home-assistant install.

My questions.

  • is there a “compatible” list of settings that does not work in supervised that does in a core/docker version?
  • what logs would be good to look at? I tried to look at the journal logs, but kind of hard to get what could cause the halt or freeze.
  • any recommendation of debug mode or settings to capture the issue?

Thanks,
Peter

System Health

version: 2020.12.2
installation_type: Home Assistant OS
dev: false
hassio: true
docker: true
virtualenv: false
python_version: 3.8.6
os_name: Linux
os_version: 5.4.84
arch: x86_64
timezone: Europe/Stockholm


GitHub API: ok
Github API Calls Remaining: 5000
Installed Version: 1.9.0
Stage: running
Available Repositories: 777
Installed Repositories: 14


logged_in: true
subscription_expiration: January 17, 2021, 1:00 AM
relayer_connected: true
remote_enabled: true
remote_connected: true
alexa_enabled: true
google_enabled: true
can_reach_cert_server: ok
can_reach_cloud_auth: ok
can_reach_cloud: ok


host_os: Home Assistant OS 5.9
update_channel: stable
supervisor_version: 2020.12.7
docker_version: 19.03.13
disk_total: 36.7 GB
disk_used: 8.1 GB
healthy: true
supported: true
board: ova
supervisor_api: ok
version_api: ok
installed_addons: Nginx Proxy Manager (0.8.2), MariaDB (2.2.1), Check Home Assistant configuration (3.6.0), Mosquitto broker (5.1), Samba share (9.3.0), InfluxDB (3.7.9), Grafana (5.3.6), AppDaemon 4 (0.3.2), Git pull (7.12), Portainer (1.3.0), Terminal & SSH (8.10.0), File editor (5.2.0)


dashboards: 1
mode: storage
views: 10
resources: 6


api_endpoint_reachable: ok

Did not receive any response on this so I tried further on.

It looks like that move all usb transmitters/receives out to an USB2 compatible and powered hub solved some parts. The ones that I’m a bit concerned about and feels it strange is the different configurations that previously have worked and now randomly starts giving me errors and after awhile does not have any complains in the logs…

Anyway, hope this can help any others out there.

Hi, I’m in the same cenario here. Did you solved the migration?

Hi,

I’ll think I had two type of problems if I remember. Initially when mowed to Proxmox I used an old laptop that I’ve spared. This laptop gave me hw issues and segments faults. You will see those in the proxmox console.

I switched over to use an NUC and the stability was now 100%.

What I cannot remember is if I switched over to the USB2 separated powered hub before switched to NUC or after.

Any case, still using the USB2 hub and working very well, so would say that if you do only have USB3 connections, the hub would probably solve your issue. However, if it’s segments fault, most likely it’s the underlying hw you using proxmox with.

Thanks,