Added a second instance of Home Assistant on same network and its messed up my original one?

Installation Home Assistant OS
1

Hey,

All was fine with my RPI4 running Home Assistant (not docker) and i decided I wanted to kind start fresh but copy most of my config over bit by bit and make it more organised.

Well I didn’t get that far, I installed Home Assistant OS on my Intel i5 Unraid as a KVM. After it installed I tried to log into the webgui using homeassistant.local:8123 but forgot the RPI4 was also running a the same time. At first it was rejecting root and no password until it clicked and I typed in the RPI4 details and it logged in but either then or very soon after (not doing anything to the RPI4 at all) it was in safe mode and I cannot get out. I tired using a backup but it still doesn’t work.

The error when trying to restart is:

The system cannot restart because the configuration is not valid: Not a directory @ data['allowlist_external_dirs'][0]

I also get this above the log:

[547390486464] The system cannot restart because the configuration is not valid: Not a directory @ data['allowlist_external_dirs'][0]
02:59:18 – (ERROR) Home Assistant WebSocket API
The system cannot restart because the configuration is not valid: Not a directory @ data['allowlist_external_dirs'][0]
02:59:18 – (ERROR) Home Assistant Supervisor
The 'entity_id' option near /config/configuration.yaml:117 is deprecated, please remove it from your configuration
02:59:18 – (WARNING) template - message first occurred at 02:59:18 and shows up 2 times

This is the pastebin of the error logs: HA Log - Pastebin.com

EDIT: I found this comment from someone about a person having quite the same issues but I dont know what it means really? Here is what they said:

Since one is in a container I'd bet that you never exposed the container port to the outside world/host.

Ie you need eitger host networking and binding the port to the broadcast address or adding a '-p xxxx:xxxx' flag when you do docker run. Post your docker command here to check.

Can also verify by running an Nmap scan on that ip of the docker host/rpi

Could this be something to with it maybe?

2

The comment seems to be rather related to port accessibility issues on Docker. In your case it’s a problem of HA refusing to even start. Were you running Home Assistant OS on the RPI4 as well?

1 Like
3

Hey, many thanks for the quick reply.
I was indeed running Home Assistant OS on the RPI4 before which is still in safemode while the Unraid box is running super! How do I go about fixing this please?

I would like ot get the RPI going again to help move the config over etc before wiping and then using the Hacs Integration: Remote Home Assistant and having the Unraid server as Master and RPi4 as slave or experimental.

4

In Safe Mode HA should start and run bare bones, at least allowing you investigate. Is it aborting startup altogether?

5

If you have 2 HA’s running, which one is homeassistant.local ?
I use http://10.0.0.63:8123 and http://10.0.0.66:8123 (replace 10.0.0.xx with the correct IP-addresses for your configuration), that way there is no confusion which one you are accessing.

6

Yeo, just loads into safe mode. Like I said I also used a full backup and its still not worked?

7

I dont use homeassistant.local:8123 anymore, just ip address and port for each one. Unraid (new one works fine) the RPI is in safe mode,

I use http://10.0.0.63:8123 and http://10.0.0.66:8123 (replace 10.0.0.xx with the correct IP-addresses for your configuration), that way there is no confusion which one you are accessing.

I dont understand what you want me to replace? My ip address is the 192.168.1.180 for Unriad and 192.168.1.120 for RPI4

8

Just wanted to say, don’t use 10.0.0.X but your own IP’s, as 192.168.1.180:8123, but you understood :).

1 Like
9

Ha, cheers!