installation-archived

[placid sage]

Luckily i didn't have a whole lot on ha

[frank perch]

I fixed my issues, I no longer have an angry lady

LMFAO

[placid sage]

😂

I'm not gonna delete it just yet... just in case.. never know it could come back 😂

[frank perch]

it's just like the angry lady, it's not coming back magically

[placid sage]

you're right but the angry lady will be more angry if i delete it

[placid sage]

shes already angry what difference does it make

[frank perch]

well I doubt your HA is angry, more like dazed and confused

[placid sage]

Sounds about right ✅️ 🤔

very confused

the real question is.. is it actually fixable?

i guess the better question is can I FIX it?

[frank perch]

take 15-30 mins, see if you can... if you can't at least figure out what's wrong, wipe it

[placid sage]

I wouldn't know the first thing to type in..

i guess your right.. sometimes it's not worth it

i value my time so wipe it it is

it has been wiped @_@

[placid sage]

Now that I have home assistant back up and running is anyone able to walk me through how to properly setup duckdns?

[fallen lichen]

Well, have you tried first following the instructions?

[placid sage]

No

where can i find the instructions

[fallen lichen]

Go to the Add-on store and find DuckDNS

[placid sage]

okay

done

install it?

[fallen lichen]

Seems reasonable to me

[crude inlet]

Install it. Then Click on its documentation tab

[placid sage]

i really hope i dont screw this up

oh

i'll try hopefully it works this time

hmm for some reason duckdns wont let me add another subdomain

error: reCaptcha failed, please try again

i'm assuming it's because i tried earlier on the same account and i deleted the subdomain

let me try on a different acct

still same issue

can you only create 1 subdomain with duckdns per day or something?

[fallen lichen]

DuckDns seems to be broken at least 23.79% of the time, probably just have to try later

[placid sage]

nvm it worked now

i had to use a different browser for some odd reason..

[placid sage]

yeah seems like it

[brittle sail]

Or just use cloudflare and buy a domain

And remove duckdns entirely

[placid sage]

yeah true, but i'm just trying to get by 😆

not super techhy, plus im too cheap for that lol

[brittle sail]

$5-10 a year is too expensive then you need to find a new hobby

[placid sage]

o

nvm then LOL

didn't know it was that cheap

[placid sage]

cloudfare.com?

welp i followed the instructions and it gave me this error

Failed to save add-on configuration, Missing required option 'domains' in Duck DNS (core_duckdns). Got {'domains': [None, 'for2.duckdns.org'], 'token': '**-4e28-8b21-806er756cfd7246e', 'aliases': [], 'lets_encrypt': {'accept_terms': False, 'algo': 'secp384r1', 'certfile': 'fullchain.pem', 'keyfile': 'privkey.pem'}, 'seconds': 300}

nvm i jusst removed the little popup before it and it saved

For duckdns do i add this to the yaml file?

http:
ssl_certificate: /ssl/fullchain.pem
ssl_key: /ssl/privkey.pem

i dont want to screw up the yaml file again

[fallen lichen]

That is what you add to HA's configuration.yaml

[placid sage]

with file editor yes?

[fallen lichen]

Sure, that'll work

[placid sage]

just to be clear

its not configuration.yaml right? or is it?

[fallen lichen]

I'm not sure what you're asking.... that section you posted needs to go into /config/configuration.yaml

[placid sage]

oh thanks

do i need to do anything with any of the files it mentions?

[fallen lichen]

such as?

[humble scaffold]

@placid sage I converted your message into a file since it's above 15 lines :+1:

[placid sage]

that is the section im referring to

is any of that important

or is it just telling you as reference

[fallen lichen]

accept_terms will need to be true as it says above

But yes, it's a reference

[placid sage]

okay thanks

well i guess i finished the documentation..

what do i do next?

i added the token and the subdomain from duckdns and made it changed the value to true... i also added http:
ssl_certificate: /ssl/fullchain.pem
ssl_key: /ssl/privkey.pem to configuration.yaml

is there anything else i would need to do get duckdns running?

[fallen lichen]

Sounds like you're ready to start the add-on

[placid sage]

oh

so just click start?

[fallen lichen]

That's the one

[placid sage]

okie now what do i do

[fallen lichen]

Click the Log tab

[placid sage]

o

says done at the bottom

[fallen lichen]

No errors?

[placid sage]

no

not that i can see

[fallen lichen]

Do you have SSH setup?

[placid sage]

probably not

what is that and how can i set it up?

[fallen lichen]

You probably should as a backup method if you ever lose UI

Another addon

[placid sage]

o

[fallen lichen]

Find the advanced SSH addon and install and configure it

[placid sage]

Advanced SSH & Web Terminal?

[fallen lichen]

That's the one

[placid sage]

gotcha

[fallen lichen]

then make sure you have an SSH client installed and are able to connect

[placid sage]

lost you at that part

oh you mean just install it?

[fallen lichen]

Correct

[placid sage]

right...

okay i've installed ssh

[fallen lichen]

Follow the docs

[placid sage]

okay

ok so i clicked the myhome assistant button..

showing me home assistant url

should i change this url?

[fallen lichen]

Yeah, point it to your instance

It'll be helpful in the future

[placid sage]

o

so just save it then?

[raven forge]

Sorry but are your shift and k keys defective?

[placid sage]

probably

Configure the username and password/authorized_keys options.

what does this mean

oh they mean change it in the configuration of ssh

[raven forge]

Yes. You need to configure one of them to be able to (remotely) log in

[placid sage]

oh i see

that's cool

[placid sage]

however i don't see a login page when i start ssh

it shows home assistant

but i guess i don't really need it now just trying to setup duckdns, but would be nice to know how it works

[fallen lichen]

You don't need it yet.... you'll want to make sure you can connect via SSH outside of the UI in case you need to fix any mistakes

[placid sage]

okay

so now that ssh is setup

what's really the next step?

or is already running?

[fallen lichen]

DuckDNS is running, yes? You've also started SSH?

[placid sage]

yes

[fallen lichen]

If you go to http://yourddns.duckdns.org:8123/

does that work?

Or have you forwarded ports yet?

[placid sage]

let me see

oh no i do not @_@

i have ubiquiti router i can easily open a port

so i tried going my dns

nothing

[placid sage]

i'm assuming i need to open a port like you mentioned?

[fallen lichen]

Yes

[placid sage]

ok

uhh might have to look up how to do this unless you can tell me... so theres an option called port.. forward ip.. and forward port..

[fallen lichen]

port is the external port you're coming in on... probably 8123 unless you want to change it. Forward IP is your HA instances IP address and forward port is 8123

[placid sage]

gotcha

so is my ha supposed to be homeassistant.local?

[fallen lichen]

That's one way to access it, that's not it's IP though

[placid sage]

oh

how would i find the ha ip?

i went into network and theres something called ipv4 and ipv6

and static and automatic or something

oh wait i think i found it

[fallen lichen]

You should set a static assignment to your HA instance so it doesn't end up changing

via your DHCP server

[placid sage]

oh

is it bad that i copied the ipv4 address and put a 8123 at the end? It still went to ha somehow

[fallen lichen]

No? That's your internal IP address for HA

[placid sage]

oh

right

so i should change that to static?

[fallen lichen]

I'd recommend creating a static lease at your DHCP server and leaving HA's network settings alone

[placid sage]

so i think i watched a video before on youtube and they changed the last 3 numbers of the static ip

why do they do that?

[placid sage]

yeah you're probably right i should ust leave it

to automatic

i don't really know what it does so

ok so i think i opened a port now

what's the next step

[fallen lichen]

Retry #installation-archived message

[placid sage]

nothing

[fallen lichen]

Nothing as in...

[placid sage]

as in this site can't be reached

[fallen lichen]

timeout?

https://www.yougetsignal.com/tools/open-ports/

Confirm w/ your external IP and 8123

[placid sage]

i see my external ip and port number for it

[fallen lichen]

type 8123 for the port

hit check

[placid sage]

port closed

[fallen lichen]

Back to router config

[placid sage]

okay

hmm did i forward the wrong ip?

i put my ha instance

port and forward port are both 8123 i believe

what did i do wrong @_@

[fallen lichen]

Do you know if you're behind CGNAT?

[placid sage]

what is that

i don't believe i am?

i think we pay for a dynamic ip

i'm not sure what cgnat is though

[copper steppe]

cgnat is where your internet provider puts all it's customers behind a NAT service, so the IP on your modem isn't a real one.

Check https://www.google.com/search?q=what+is+my+ip and compare to what's on your modem

[placid sage]

oh i don't believe so

o

[copper steppe]

I'm on starlink, which uses cgnat

[placid sage]

i believe i have xfinity

[copper steppe]

Do the comparison. Google result vs IP address in your modem/router

[placid sage]

okay

[copper steppe]

eg: google says my IP is 206.83.x.x but my modem thinks it's 100.103.x.x (which is reserved space, not internet address anyway)

[placid sage]

so would this be the ip address of my isps modem or would it be my dream machine router

[copper steppe]

Whatever is the "most outside" address. Not a LAN address..

[placid sage]

oh

[copper steppe]

Eg: my lan is 192.168.* and my router has two IPs, one is 192.168.x.1 and one is the internet side, which is 100.103.x.x

[placid sage]

im assuming id have to log into my router to see this address?

[copper steppe]

You want the internet side of the router/modem

Yes.

[placid sage]

or can i just type in ipconfig in command prompt

[copper steppe]

NOPE

Your LAN isn't/shouldn't be real IP addresses. It will be a private network, with NAT on your modem.

ipconfig will show you the 192.168.* (LAN) address

[placid sage]

ok i logged into my router

not the gateway address right?

ipv4 address?

or is it wan?

[placid sage]

i see xfinity network the same as ip address as that google link you sent me

so does that mean i have a cgnat or not?

[copper steppe]

It means you don't have cgnat

(which is good)

[placid sage]

oh

thank goodnes LOL

[copper steppe]

the WAN address on your modem should be the same as what google said

[placid sage]

i'm still trying to setup duckdns lol

i opened a port.. but

wont go to the domain

perhaps i'm doing something wrong

[copper steppe]

I haven't got time to help sry

[placid sage]

no problem

dang it.. i got this far and still wasn't able to connect lol

perhaps i have to create it within my xfinity router itself

oh wait i think i see the problem

i can open a port on my ubiquiti router... but

on my xfinity modem it says...

"Managing your home network settings is now easier than ever.
Visit xfinity.com/myxfi to set up port forwards, among many other features and settings."

my isp must be blocking blocking it or something

[placid sage]

nvm i was able to enable it through the mobile app

[placid sage]

okay by some miraculous miracle i was able to log in to home assisstant with that duck dns domain (http://yourddns.duckdns.org:8123/)

is that pretty much finish the install? do i need to do anything else? also can i close the ports now or will that screw something up?

[kindred escarp]

just installed hass os on a nuc and its stuck on preparing home assistant, it has been like that for hours
any tips?

[fallen lichen]

The ports will need to be open as long as you want external access and to use DuckDNS

Restarting HA should make it available via HTTPS

[placid sage]

awesome thank you so much again for all your help!

[fallen lichen]

You're welcome

[placid sage]

i didn't think i'd be able to do it haha

i'm glad i didn't break home assistant this time 😂

Is there a reason why at the top it says not secure

[fallen lichen]

#installation-archived message

[placid sage]

oh right

restarted but still says not secure to the left of the url

what does this mean anyway

that the connection isnt secure enough?

[fallen lichen]

Requests over HTTP are subject to interception

[placid sage]

o

so it's not something i did wrong?

[fallen lichen]

If you added that section to your http: key in your configuration.yaml and restarted... it should be served via TLS

[placid sage]

which section?

[fallen lichen]

#installation-archived message

[placid sage]

oh i skipped that part as i didn't understand what it was saying

are you saying this? ssl_certificate: /ssl/fullchain.pem
ssl_key: /ssl/privkey.pem to configuration.yaml

i added that part

to the yaml

idk about the other part you are referring to

[fallen lichen]

🤨 What part are you saying you skipped then?

[placid sage]

oh

[placid sage]

this part

one sec

[fallen lichen]

Sure, that's a reference of the options as we discussed

[humble scaffold]

@placid sage I converted your message into a file since it's above 15 lines :+1:

[placid sage]

ah gotcha

oh no shit

@_@

i think i did add it but

im looking at the configuration.yaml now

its not there

I DIDNT SAVE IT @_@

this part ssl_certificate: /ssl/fullchain.pem
ssl_key: /ssl/privkey.pem

@_@

am i still able to add it?

wait where was i supposed to add this?

ssl_certificate: /ssl/fullchain.pem
ssl_key: /ssl/privkey.pem

i swear i added it somewhere

[fallen lichen]

Take a look at the docs again

[placid sage]

Additionally, you'll need to configure the Home Assistant Core to pick up the SSL certificates. This is done by setting the following configuration for the HTTP integration configuration in your configuration.yaml:

http:
ssl_certificate: /ssl/fullchain.pem
ssl_key: /ssl/privkey.pem

welp i screwed up AGAIN

JUST BECAUSE I DIDNT CLICK SAVE

UGH

please tell me i dont have to redo this again

i'm gonna go ahead and assume i have to :C

[fallen lichen]

Redo what... copy and paste? If it didn't save, yes

[placid sage]

oh

it wouldnt break home assistant or anything?

[fallen lichen]

Why would it

[placid sage]

LOL IDK

because it did last time

[raven forge]

Backups. Lots of them

[placid sage]

lol

if only i knew how to set that up

[raven forge]

https://www.home-assistant.io/common-tasks/os/#backups

[placid sage]

ok i will try to put the code in. hopefully it will still work when i restart

[placid sage]

and thank you 🙏

[placid sage]

it worked now thank you 😂

not sure why that was so difficult for me

literally the only reason i wanted duckdns was to add over my smartthings devices lol

[fallen lichen]

Aren't those all Zigbee?

[placid sage]

hopefully that will be less issues

zigbee and zwave i believe

[fallen lichen]

Sounds like next steps are to get those local

[placid sage]

and how would i do that @_@

[fallen lichen]

#zigbee-archived & #zwave-archived

[placid sage]

is smartthings cloudbased?

[fallen lichen]

Check out pinned posts

Yes it is

[placid sage]

i thought they moved to edge drivers though

How can i tell if my devices are locally controlled or cloud based @_@

why is networking so hard

[fallen lichen]

For one, Integrations will tell you: https://www.home-assistant.io/integrations/smartthings/

Its IoT class is Cloud Push.

[placid sage]

o

so i have to generate another token.. interesting. Seems like a lot things in networking have tokens 🤔

I have a question. if i add more devices to smartthings will they automatically appear in home assistant?

like with this integration?

[fallen lichen]

Entities automatically synchronized upon restart of Home Assistant when changed in SmartThings.
I'd assume this means upon restart

[placid sage]

thats cool

if i can even get the integration to work.

its saying i need something called ssl

[fallen lichen]

That'd be the HTTPS you set up

[placid sage]

Oh thank GOD

that freaked me out a bit lmao

what is ssl anyway

[fallen lichen]

https://www.ssl.com/faqs/faq-what-is-ssl/

[placid sage]

oh

so its just an extra layer of security

essensially

sorry i know i ask way too many questions, prob getting sick of me by now 😒

"Set the external URL in the Home Assistant configuration to the URL that Home Assistant is available on the internet"

is the external url the duckdns that i setup?

[raven forge]

Yes

[placid sage]

sent me to this page https://www.home-assistant.io/docs/configuration/basic

i'm assuming i would put these 2 lines external_url: "https://www.example.com"
internal_url: "http://homeassistant.local:8123" in my config.yaml

[raven forge]

I think you can also set it here: https://my.home-assistant.io/redirect/network/

[placid sage]

says homeasssistant.local:8123

i assume thats the internal url

why do i even need to add it to my config file

cant i just add a new integration and type in smartthings

[fallen lichen]

Following the instructions is usually the best approach instead of skipping steps

[placid sage]

yeah true

i hate reading i'm sorry @_@

i need to get used it

bad habbit i guess

how do i find my internal ha ip

[manic bane]

What do you mean by that? Do you mean your internal IP for HA on your network?

[placid sage]

https://www.home-assistant.io/docs/configuration/basic

its saying to add internal and external url to the conguration.yaml file

what does this mean

[manic bane]

Url isn't IP

[placid sage]

oh

oops lol

then how do i find my internal url?

or is it just http://homeassistant.local:8123 by default

[wooden elm]

What is wrong with the following build command of docker:-
docker build --build-arg BUILD_FROM=homeassistant/amd64-homeassistant-base:3.16 . -t homeassistant:local

[raven forge]

Are you gonna tell us why you think it is? Also wrong channel

[wooden elm]

It is throwing error as: manifest for homeassistant/amd64-homeassistant-base:3.16 not found: manifest unknown: manifest unknown

[fallen lichen]

That tag doesn't exist, what are you trying to do

[placid sage]

do i have to use nabu casa cloud to add smartthings into home assistant?

[fallen lichen]

No

[placid sage]

o

well when I tried adding my external url (duckdns) and internal url into configuration.yaml and checked the config it would say "Not a directory @ data['allowlist_external_dirs'][0]"

[fallen lichen]

Sounds like you incorrectly edited the file

[placid sage]

do i need all the extra stuff like lattitude and longitude... time zone?

basically i just copied and pasted this into my yaml and edited the external and internal url

[humble scaffold]

@placid sage I converted your message into a file since it's above 15 lines :+1:

[fallen lichen]

You can't just blindly copy and paste things and expect it to work

All of that is likely incorrect for you

[placid sage]

o

[fallen lichen]

I'd suggest reverting that entirely, and set it via the UI: https://my.home-assistant.io/redirect/network/

[placid sage]

can you elaborate a bit more, what do you mean by that?

[fallen lichen]

Remove the homeassistant: section you added to your configuration

Set your external (Internet) URL via the link I pasted above

[placid sage]

oh gotcha

[humble mirage]

This ... this was all explained yesterday (or the day before) when you were asking about it

[placid sage]

i'm very bad at following instructions @_@

[humble mirage]

Time to learn to be better at it 😉

[placid sage]

i can't get better at it if my brain doesn't comprehend @_@

[humble mirage]

https://tenor.com/view/youre-doomed-youre-all-doomed-doomed-doom-end-of-the-world-gif-10248096

[placid sage]

Invalid config for [http]: [external_url] is an invalid option for [http]. Check: http->http->external_url. (See /config/configuration.yaml, line 12).

This is what my yaml looks like

automation: !include automations.yaml
script: !include scripts.yaml
scene: !include scenes.yaml
http:
ssl_certificate: /ssl/fullchain.pem
ssl_key: /ssl/privkey.pem
external_url: "https://.duckdns.org:8123"
internal_url: "http://homeassistant.local:8123"

[humble scaffold]

To format your text as code, enter three backticks on the first line, press Shift+Enter for a new line, paste your code, press Enter again for another new line, and lastly three more backticks.
```yaml
example: here
```
Don't forget you can edit your post rather than repeatedly posting the same thing.

[humble mirage]

Why are you setting those there?

[placid sage]

IDK @_@

[humble mirage]

#installation-archived message

[placid sage]

i thought thats where i'm supposed to put it

[humble mirage]

You're not joking about being bad at following instructions

#installation-archived message

it is done in the UI

[placid sage]

OH

Soooo

where it says configure network interfaces?

i see something that says host name

at the top

@_@

[humble mirage]

Home Assistant URL

[humble scaffold]

Various settings are hidden unless your user profile has Advanced mode enabled. These include some add-ons, the Lovelace Resources menu, server control options, and more.

[placid sage]

your instance url??

OH

i see now

configure my home assistant by entering url of your home assistant instance

im assuming i would put the duckdns followed by the port number

[humble mirage]

Yes

[placid sage]

ok i have done that part LOL

i assume i should probably delete that other section from the yaml file

[humble mirage]

"that other section"?

[placid sage]

umm well you said it was done in the ui didn't you

this section

external_url: "https://.duckdns.org:8123/"
internal_url: "http://homeassistant.local:8123/"

remove that part

[humble mirage]

One day ... one day you'll use code markup

Yes, those two lines

Only those two lines

[placid sage]

gotcha

[placid sage]

that will be a while from now the way i'm going 😂

oops sorry forgot not to ping

so at this point i assume i would just be able to add smartthings integration

and add the token i copied

but problem is.......

when i try to do di

would you like to add smarthings integration? click yes

This pops up Error
Config flow could not be loaded: 500 Internal Server Error Server got itself in trouble

[humble scaffold]

If you're having problems with your updates to your configuration:

• Check the troubleshooting steps
• Check your log file - remembering you may need to set logger to info or debug
• Explain what the problem you're having is - sharing configuration, errors, and logs

[placid sage]

il y a plus de deux heures que j'ai le message "préparation de l'assistant" normal ?

[humble mirage]

English please @placid sage

[placid sage]

Hello, sorry for my English! it's been more than two hours since I got the normal "preparing assistant" message?

I'm waiting on the address http://homeassistant.local:8123/

[empty bloom]

refresh the page

[placid sage]

anyone to reassure me or help me?

[humble mirage]

And if the refresh doesn't work, click the blue dot

[placid sage]

done several times

OK ! It was enough to cry!!! Thanks for the advice, it works! Finally !

[safe stirrup]

Is it possible to downgrade HA without disasters happening?

I'm on 2023.8.1 but want to go back a few versions as I think it could be the cause of my zigbee device not configuring

[humble mirage]

Depends, the official answer is to restore a backup from the older version

[placid sage]

Hikvision integration with ONVIF missing entity for smart event

[warped timber]

Heya!

Quick question; I've instaled a HA OS dev build a while ago
Now i do not get the 10.4 OS update presented in the UI.

Is this expected behaviour?

[empty bloom]

What channel are you on for system?

I can't remember the exact term

but there's the normal channel, beta channel, and dev channel

if you're on the dev channel, you'll get all updates that are current and the current could be the latest release depending on when dev channels are cut

[warped timber]

i manually instaled a specific dev channel

I've just pressed the leave the beta channel option

But still no update offered by the ui.

I run Operating System 11.0.dev20230613

This is a nigthly build

[empty bloom]

so what's the problem, you want to go back to 10.4?

[warped timber]

Yep

[empty bloom]

I'm not sure you can

[warped timber]

Actualy this devbuild is from in between 10.3 en 10.4

[empty bloom]

IIRC OS and Supervisor have no way to install a specific version

not according to what you posted above

it's listed as 11

which is > 10.4

[warped timber]

Mhhh fair point

[fallen lichen]

ha os update --version 10.4 should work

[warped timber]

Would i be able to do that one over SSH?

[empty bloom]

I don't use OS so I wasn't sure if you could install specific versions of OS via that method

that doesn't work for supervisor

[fallen lichen]

Supervisor does also have an update method w/ a specific version

[warped timber]

In specifcly OS that i want to update to 10.4

Gonne give that a try over SSH

If not, gotta grab some monitor and keyboard somewhere

😂

[fallen lichen]

Over SSH should be fine

[empty bloom]

worse case is make a backup

reinstall then load the backup

[warped timber]

Well, its running. Lets grab some popcorn, and wait for i guess 15 minutes (odroidM1)

[frank perch]

you just caused me to spit monster on my screen

15 minutes is... aggressive at best

LOL

[warped timber]

?

If you ment that it would go faster

well it was more like 1,5 minutes

But it worked perfectly. Thanks for the tips gents!

[frank perch]

oh wait it was a backup, not a restore

my bad

[plush quartz]

hi all. I'm trying to connect my samsung watch to HA. I installed the Android HA app on my phone and when I set the address to my HA, I get: home assistant hostname certificate mismatch
why is that?

[fallen lichen]

I'd imagine you're trying to go to the wrong hostname when using a HTTPS URL to connect

[plush quartz]

when I connect to it via my PC, I use https://homeassistant.local:8123

when I open the android app to configure the watch, it asks to put the address

it I use http://192.168.1.182:8123 I also get an error

[fallen lichen]

What's the actual domain name the SSL cert was issued to

[frank perch]

if you have SSL enabled you have to use the hostname your cert is for or you'll get an error

[fallen lichen]

DuckDNS presumably?

[plush quartz]

I get error code: -1 Description: net::ERR_EMPTY_RESPONSE

[plush quartz]

yes

[fallen lichen]

So use that domain name

[empty bloom]

with https

not http

[frank perch]

^this

[plush quartz]

oh

got it

ok

[empty bloom]

notice lack of s and presence of s

that stands for Security

(Not true, bad joke)

[frank perch]

well it stands for 'secure'

so you're not far off

[empty bloom]

ya, i knew it was something like that

[frank perch]

I honestly don't understand why folk still use duckdns when it's down so much

[plush quartz]

ok, with the duckdns I gete connection refused

[fallen lichen]

Some people go through a lot for free things

[empty bloom]

I legit switched a month ago because of that

I'd wager people still use it because it's what's the most documented

and the other half of the people are too lazy to switch

[frank perch]

perhaps I should document dyndns and LE

because that's what I always used before I switched to cloudflare

[plush quartz]

I use it because that's what I saw documented

[empty bloom]

or a proper cloudflare tunnel tutorial

[plush quartz]

I have no understanding of how to issue a secure certificate to use https

[frank perch]

those exist, that's what I used to setup my tunnel 😄

[plush quartz]

eitherway, no connection setting works

[empty bloom]

you don't do that, the DDNS addon does that for you

[plush quartz]

yeah, which is why I use duckdns

[frank perch]

wait, did you enable SSL in your install ?

in the config

[fallen lichen]

Yeah, they did since it was responding over HTTPS

[empty bloom]

I'd bet he's not using nginx, so local won't work

i.e. you have to use the ddns url at all times

[frank perch]

yeah this is why I asked cuz this shouldn't work

[empty bloom]

well, only he will know that

[frank perch]

my tinfoil hat stops my mind reading abilities

[fallen lichen]

Desktop browser at least allows you to click past those pesky TLS warnings

[frank perch]

TLS? you're old.

[empty bloom]

he's a cloud breaking free from a painting

or is that a window?

I hate how you can't zoom in on avatars

[frank perch]

you can tho

take a screenshot and zoom in

😛

[fallen lichen]

Yeah, I think it's a cloud... I'm not quite sure, some server required an avatar to join and so... random avatar

[empty bloom]

https://tenor.com/view/enhance-gif-10953787

[frank perch]

I was thinking cotton candy

[fallen lichen]

Did you go through the port forwarding set up and all of that? Alternatively perhaps there's a NAT reflection issue w/ your router causing you to not be able to access HA via your external IP from your LAN

[plush quartz]

I was just about to write that I resolved it. Yes, there's a port forwarding in my router.

thanks!

now I got it sorted and this whole thing is to try and get ESPresence working with my galaxy watch

it's impossible to get the BT MAC address in the logs with the associated device ID for espresnese

[kindred escarp]

need assistance troubleshooting an installation

this is the second time installing and im stuck on the "preparing home assistant" screen. when i look at the logs i see this:

23-08-11 15:25:36 WARNING (MainThread) [supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30sec```

i have a monitor, network and keyboard available

[tender wren]

Check your internet connection

[kindred escarp]

ive confirmed that connection has internet access, im not sure how to check it on the console

should home assistant OS be available locally if I dont want to use internet? I can ping the HA box and i can browse to the site but is stuck on that step. this is the second time i flash the hdd

[fallen lichen]

Internet access is required for Home Assistant to come up, as there are required downloads during the initial process

[kindred escarp]

i running a pihole on my network, i suspect that is blocking the connection then.

[uncut comet]

Unicorn sneezing through a screen door

[frank perch]

ha net info, or net info from the ha> prompt

[kindred escarp]

One line says “host internet: false” should that value be false ?

[frank perch]

if your internet connection is working (or at least the supervisor thinks it is) that will say true

[kindred escarp]

How do I manually setup the dns address via the cli ?

I think that’s where it could be

[vague hill]

Hi everyone,
I'm running HA container and am surprised to see that some files are owned by root.
I am running compose and am running as user, not root.
Anybody have an idea why some files are owned by root?

[humble mirage]

Because the container runs as root

[vague hill]

Even though I up-ed as user?

[humble mirage]

Docker itself runs as root

The user you run the commands as only matters in that it has to be in the docker group

[vague hill]

😦
I was hoping not to have to run rclone as root but I am seeing error messages of files that are changed to root.
So either I need to chown before calling rclone or call sudo rclone

[humble mirage]

You generally want your backup daemon to be root or otherwise able to read all files anyway - that way it can also preserve ownership etc

(fun fact, some commercial Unix versions have a special user for backups that can read any file)

[vague hill]

Since I was running as the only user, I figured that user can access everything anyway.
But since HA returns everything to root, I will run the backup as root then 🙂

Interesting fun fact 🙂

[frank perch]

nmcli> set ipv4.dns 1.1.1.1

[kindred escarp]

Didn’t work, also noticed the time is not synced properly

Wish I could take a picture

[humble scaffold]

Please use imgur or other image sharing web sites, and share the link here.

Image posting is blocked in most channels to discourage people from sharing text as images. Sharing text as images assumes that everybody sees the world as you do, which isn't the case. Some people are colour blind, or have visual impairment that means they can't make sense of an image of text.

[runic venture]

@fallen lichen @humble mirage Heyo! Thought I'd let you know how I went. Turned out I didn't have a spare 500 GB NVMe sitting about, and when I got set up to clone to a 256 gb disk (resizing the unused space away in the process) it told me the disk was unreadable and may have errors. Now I don't know if that's because Proxmox pre-provisions (or what the correct terminology here is) disk space, which I assume it does based on the disk % allocated when viewing the pve LVM (which is an acronym I don't know still) OR if it's actually erroring. I ran through it with a logical HDD sector checker in read-only and it came up without errors so I can only assume it's just all allocated and hence the actual used data is indistinguishable.

So all that aside I set up a brand new pve and HA using the two tteck scripts (thanks you legend) and it's been up for about 24 hours now without spitting any console entries out at all. On both my previous setups I resized the disk for HA and I'm worried/wondering if that's what caused this, so if it's a possibility I can try it - otherwise I might just leave the damn thing at 32

[low notch]

resizing disks demands that you inform the guest os about the change and also resize the partition table in the guest beforehand. only expanding should work without trouble

[obtuse geyser]

I need help for first time installation.

In logs it is showing that version 10.4 is installed but not going anywhere.

What should I do next ?

[raven forge]

Share the logs

[humble scaffold]

Please use a code share site to share code or logs, for example:

• https://dpaste.org/ (select YAML for the language, and consider picking a longer expiry)
• http://pastie.org/ (select YAML for the language)
• https://paste.debian.net/ (you guessed it, select YAML as the language)

Please don't use Pastebin, since it can randomly add spaces to the main view. Please also don't share text as images since it makes it harder for people to help you. Remember that others may have colour blindness, impaired vision, etc.

[floral igloo]

um i got a dumb question, the ''custom_components'' folder should be put in ''config'' folder.
but i dont see one inside the file browser.
adding things in ''C:\Users\mumu0\VirtualBox VMs\Home Assistant\custom_components'' wont work.
(im installing this on windows 11)

[obtuse geyser]

I got the access !!

[runic venture]

Yeah; in my experience in VMWare, I've only ever had to expand a disk, and in Windows system's it's fairly straight-forward (you can generally even shrink it if you're sure you're not going to cut into any allocated data) so I would assume it's the same here, but Prox and Linux devices are very new to me and likely quite different

Thank you for chipping in, every extra bit I can learn is great

[manic bane]

Virtualbox doesn't mount a directory on disk to the VM, it will have created a virtual harddrive that you can't access in windows directly

[runic venture]

When I ran VirtualBox, I ended up using the SAMBA share add-on to expose the HA files to the network, setting a username/password, and then just editing the files in HA through that share as if it were a network drive

It was pretty reliable, it never didn't work - it was more that VirtualBox HA is unreliable and seems to run into scheduling/allocation problems a lot

"it never didn't work" might be technically applicable but boy does that wording sound off...

[wraith flicker]

Thanks for this suggestion. I finally had time to attempt a fresh install and restore and it has worked and I’m now running 10.4

[ornate spear]

Hello Team, I am migrating HA from a VM to a mini PC (Geekom). I created a backup successfully. When I installed HA from scratch on it, the restore of the backup is not executing. I have googled, followed YouTube video advice to the tee, no luck. It will take me a very long time to start building HA up again from scratch. Any tips or suggestions would be helpful.

[humble mirage]

Are you uploading the backup at the onboarding screen, or using Samba?

How big is the backup?

[ornate spear]

I am using my Synology NAS / NFS

255MB

[humble mirage]

That's small, so you should be fine to upload it any way

[ornate spear]

Ya, Is there a way I can place the backup on a USB key that directly plugs into the mini-pc for restore?... I have not tried that yet.

[humble mirage]

No

(at least, not that I know of, until after you've onboarded anyway)

[ornate spear]

Thanks,.... that saved me time with trial and errror.

[humble mirage]

https://www.home-assistant.io/common-tasks/os/#restoring-a-backup-on-a-new-install

[crimson furnace]

Thinking about upgrading my old HA hardware to a newer OptiPlex 10500T - kinda overkill for HA but I'm thinking of starting fresh and wondering what method of installation I should go for? I primarily want to use Frigate but I wonder if I'll expand later to include Plex etc too.

I've never used Proxmox but is that something I should look into?

[raven forge]

Subjective, of course, but absolutely

[crimson furnace]

Would that generally be the best/most supported option? I've not set anything up like that before 😄

[raven forge]

The most common is HAOS: https://analytics.home-assistant.io/
Here's a list of supported installation methods: https://www.home-assistant.io/installation/#compare-installation-methods
You can run HAOS in a VM on proxmox.

[little swift]

somehow my trusted networks wont work anymore...https://www.home-assistant.io/docs/authentication/providers/#trusted-networks I keep seeing the dropdown, but when I click that, I get the error message. Even before any user is selected

I do Not have any trusted_proxies configured, so am a bit lost as to the why

[raven forge]

What error message?

[little swift]

"Your computer is not allowed", as explained on that link

[ornate spear]

I got a successful restore. It took multiple attempts. Made sure that the target mini-PC had the same network IP as the backup I was using. Also, I let the restore run over night - 10hrs as there is no feedback on the restore status. Most everything came up great. I restored from a Synology NAS with NFS successfully. I tried to restore from other locations unsuccessfully.

[little swift]

ignore me... found the cause, and its a good thing..... the device was logged in on the wrong wifi, which I had not trusted in the trusted_networks settings.... my apologies

[versed vapor]

Does the "setup failed" on a device is actually anything different from "device unavailable"

[frank perch]

on the device or integration ?

[versed vapor]

on the integration

the devices have already been added but after every coreish update it goes back to that message when the device is unavailable

[frank perch]

ok yeah that usually is caused by the device being unavailable. can't integrate an unavailable device.

[versed vapor]

i see so its probably save to ignore

[frank perch]

it is indeed

mine tells me that for my Denon receiver for example, I know it's unavailable because... it's off.

same thing on the samsung tv's

so safe to ignore unless you know the device should have been on

you can always turn the device on, and then reload the integration

[versed vapor]

i figured, just wasnt sure if it actually is trying some sort of needed migration

[frank perch]

and some things are cloud so for example when my internet goes down, I get the same thing on my weather stuff

migration or mitigation ?

[versed vapor]

migration as i kinda just saw it after updates of the os/core package

[frank perch]

nah it's just cuz it was off

after the restart

[versed vapor]

makes sense

[frank perch]

I have harmony hubs to turn on anything I can't do otherwise so sometimes I just flick them on and reload the integrations to make sure but that's because CDO

[heavy sleet]

i want to move my HAOS instance over to Proxmox for when i screw something up, so its easy to just restore a snapshot

i'm currently running HAOS bare metal on a thin client with a few add-ons:

mqtt
Z2M
VS Code
and a few others

what would be the best way to migrate my setup?

i would have to install proxmox on the same machine that is currently running HAOS now

i'm kinda scared that when i restore a backup everything will be broken

[humble mirage]

Take a backup, download the backup, fresh install of HA, restore the backup

[frank perch]

you're going to put proxmox on a thin client?

[heavy sleet]

i've read online that some add ons will break

thats why i was asking

[heavy sleet]

yes

only for the snapshots tbh

[frank perch]

out of curiousity, which model of thin client?

[humble mirage]

Ah... good old context free FUD

[placid sage]

Nedd Help by installation.:( after type name and passwort i cant chose home location. Always Error "NetworkError when attempting to fetch resource" also a new reinstall doesnt work show the same Error

[heavy sleet]

idk

some low end old one

[frank perch]

have you confirmed if it even supports virtualization?

[timber lagoon]

virtualization smurtualization...

[trim halo]

I'm trying to install HA via docker on a x86 system and after installing I don't get any connection on port 8123

The host OS is debian

No errors in log

[humble mirage]

Are you using Docker Compose, or a run command?

(the answer should be compose, but not everybody has seen the light)

[trim halo]

I've tried both, latest one was compose

[humble scaffold]

Please use a code share site to share code or logs, for example:

• https://dpaste.org/ (select YAML for the language, and consider picking a longer expiry)
• http://pastie.org/ (select YAML for the language)
• https://paste.debian.net/ (you guessed it, select YAML as the language)

Please don't use Pastebin, since it can randomly add spaces to the main view. Please also don't share text as images since it makes it harder for people to help you. Remember that others may have colour blindness, impaired vision, etc.

[humble mirage]

Please use one of those to share the compose file

[trim halo]

https://dpaste.org/qNbz6

[humble mirage]

Were files created in ${USERDIR}/docker/homeassistant/config

[trim halo]

Yes, I just checked. automations.yaml, bluprints/ configuration.yaml etc

[humble mirage]

And a log file?

[trim halo]

home-assistant.log
looking at it now

home-assistant.log home-assistant.log.1 and home-assistant.log.fault are all empty

[humble mirage]

Check the docker log for the container

And if that's empty stop it and set logger to info

[humble scaffold]

If you're having problems with your updates to your configuration:

• Check the troubleshooting steps
• Check your log file - remembering you may need to set logger to info or debug
• Explain what the problem you're having is - sharing configuration, errors, and logs

[trim halo]

Docker log:

https://dpaste.org/dhSPi

[humble mirage]

Yeah, effectively empty

[timber lagoon]

I appreciate your service sir...

🫡

[trim halo]

OK I'll set logger to info and restart

[timber lagoon]

yeah! you do that...and thing about what you have done...

[humble mirage]

Are you feeling ok?

[trim halo]

Are you talking about changing the logging driver docker is using?

[humble mirage]

I'm talking about #installation-archived message

[trim halo]

https://dpaste.org/pmcyf

[humble mirage]

Look in the log file

[trim halo]

I included the log file at the bottom of that. one linie

*line

[humble mirage]

There should be more, a lot more

Try debug maybe?

[trim halo]

OK now there is a lot more

https://dpaste.org/2b4nT

[humble mirage]

It looks like it's starting ok... what hardware is this?

[trim halo]

Its a NUC

I see a lot of references in the logs to 10.0.2.100 but this machines local ip address is 192.168.68.100

[humble mirage]

Last time I did a fresh Docker install on decent hardware onboarding was available in under 3 minutes

[trim halo]

If I curl 192.168.68.100 (this machines local address) I get "(7) Failed to connect to 192.168.68.100 port 8123 after 0 ms: Couldn't connect to server"
If I curl -v 10.0.2.100:8123 I get "Trying 10.0.2.100:8123..."

[humble mirage]

I'll do a fresh one now

[trim halo]

I think it has something to do with networking

[humble mirage]

That seems probable, maybe some host firewall

Ok, sub-15 seconds and I can access onboarding

[trim halo]

I used rootless-kit from docker to run rootless containers on debian and I think that is what it messing it up

I haven't seen it before because I wasn't running anything with 'network_mode: host'

[humble mirage]

Ah...

[trim halo]

Thank you for your help. I'll see if I can track this down'

[humble mirage]

The official HA container expects to run as root

It needs to run as root

[limber sinew]

When I try to update to the new core update my whole internet is stopping only local is working and no external internet need to unplug the raspberry or vm then i have internet again also with an new install

[quick forum]

Hi there! I am getting an error on macOS intalling HAOS on Virtualbox

[toxic quartz]

Hello there!
I seem to have locked myself out from HA after installing NGINX so I would like to reroll to a backup. I see that my backup is protected in the list and fortunately I've found the password but for some reason I can't get it to run.
I should type something like 'backups restore 35bg69pm 1234badpasword' but I get an error, about receiving 2 arguments.
If I don't use a password, it gets an error and I could see in supervisor logs that I needed to put in a password.
Is there anyone that can point me in the right direction?

[raven forge]

Try

ha backups restore 35bg69pm --password 1234badpasword

You can get help for a (sub-)command by prefixing it with help like this

ha help backups
ha help backups restore

ha is not needed inside the ha > cli/prompt

[toxic quartz]

So I should use --password in between? I will try. 🙂

It's processing so that did the trick, stupid of me not to think off that. Not really used to cmd lines. Hopefully this will get me back in to the system. Thanks @raven forge !

It's restored, I used help btw but was confused about the flag thingies. Went from ha snapshot restore (filename) to snapshot restore (slug) to backup restore slug and got stuck there for some time. Then realized that I needed the password and really didn't knew where I put it. Guess I'll make a note in onenote for next time. 😄

[limber sinew]

Please can anyone help me with this

[manic bane]

So when you update you're saying your entire network stops working?
Are you running adguard or anything else on the pi?

[limber sinew]

Yes entire outgoing internet stops only local works and I don’t have anything else running on the pi

First I thought it was the raspberry so I tried an other one also on a vm but every time he updates it then the internet stops and I need to unplug it and reset my network

[manic bane]

How did you connect to the VM one? If you did homeassistant.local you might've still connected to the pi

[limber sinew]

I reset the internet unplug the pi and then I tried on a vm so as far as I know the pi was not connected because with ip scan only one comes on

[humble mirage]

And you're not on dialup/some other very low bandwidth connection?

[sudden canyon]

why is home assistant in docker so needy?

[humble scaffold]

@sudden canyon I converted your message into a file since it's above 15 lines :+1:

[humble mirage]

lsiopy

You're not runnng the HA Docker image, are you?

[sudden canyon]

transitioning to docker after my vm died this morning and now getting this

[sudden canyon]

i think so, is there a go to image?

[humble mirage]

Try using the official image

You mean... the one in the documentation?

https://www.home-assistant.io/installation/linux#docker-compose

[sudden canyon]

i dunno mahn, i just installed the one in the app store in unraid 😄

[humble mirage]

Well, you get to blame unraid yourself for failing to RTFM

[sudden canyon]

lel

[limber sinew]

No I have gigabyte internet

[humble mirage]

Never heard of HA causing a problem like that, except where you're using HAOS to provide DNS/DHCP to the network

[limber sinew]

No me neither but no haos is not used as dns or dhcp version

[limber sinew]

If I don’t do the update Iam fine

[humble mirage]

HA is wired or WiFi?

[limber sinew]

Wired

[humble mirage]

How are you testing that your whole Internet goes out?

[limber sinew]

Static ip from router

Ping to google

[humble mirage]

From?

[limber sinew]

From my laptop

And also check it on my other server I have running

[humble mirage]

Tried another device, one wired to the router?

Then I'd check your router

[limber sinew]

Yes all mine devices stop with it

[humble mirage]

Because that's the likely problem

[limber sinew]

Also tv and radio stops

I will check it out

[humble mirage]

It could be that the bandwidth shaper is mis-configured

[limber sinew]

I will look in to it thanks

[barren fulcrum]

Hey!
Is there anyone here who would like to help me?
I'm trying to deploy HA with VMware on my PC, but there seems to be some problems.

It seems to have reported it wrong.
It reads: Waiting for the Home Assistant CLI to be ready... [WARN] Home Assistant CLI not starting! Jump into emergency console...

[ 547.6615791 rcu: INFO: rcu preempt detected stalls on CPUs/tasks:

[ 548.7820311 rcu: 01-...: (1 GPs behind) idle-ed9c/1/0x4000000000000000 softirq-53759/53760 fqs=9

[lean skiff]

Hello! I'm moving from a Pi4 to an intel NUC, HAOS on both ends. I'm able to get to the onboarding page on the new device and have a backup ready to restore. Backup is about 600MB. Backup seems to import successfully, but when I run a full backup restore the connection times out. Observer says everything's healthy but clearly not since all requests fail. Let it sit for an hour, no dice. Hard restarted and plugged in the box, ha backups shows the backup. Attempting restore stops the core, gets through docker config and folders, starts the core restore, then nothing. No additional logs in ha supervisor logs or ha core logs. Looking at the code, it clearly gets to https://github.com/home-assistant/supervisor/blob/main/supervisor/backups/manager.py#L325 but then fails somewhere before https://github.com/home-assistant/supervisor/blob/main/supervisor/homeassistant/module.py#L395 which seems to write the first log message when restoring the core. Where would those core restore logs be logged if not the core or supervisor logs, and if it is stalling somewhere there how would I fix it? Thank you for your help!

[sudden canyon]

Is it possible to roll back to an older version of HA in docker by just using the older version tag or something?

[humble mirage]

Yes

https://www.home-assistant.io/common-tasks/container/#run-a-specific-version

[frank perch]

dammit the pigeons

[humble mirage]

And... you didn't even read the question

[frank perch]

whoops I read HAOS

not container

and that's not the same peson

lol

[sudden canyon]

Thanks Tinkerbell

[barren fulcrum]

So are there anyone could help me to deal this problem? Thanks

[shadow zealot]

Hi, my HA keeps crashing every 1-2 days. How can I check if the SD Card is the point of failure here?

[raven forge]

Logs, mostly. Is it just HA? How is it installed?

[tender wren]

The easiest would be to use a new SD card and restore a backup. If it keeps crashing, is wasn't the SD card, otherwise it's most likely. For stability I would suggest a migration to a USB SSD and a powered hub or another platform.

[shadow zealot]

What would I look for in the logs?

[shadow zealot]

Yes, SSD is planed in a next step.

[raven forge]

All kinds of IO related warnings/errors but please answer the questions 🙂

[shadow zealot]

It's installed on a 32GB micro SD from SanDisk. Running on a Raspi 3B.

[raven forge]

So HAOS? Can you describe what happens during those crashes? Assuming HA becomes inacccesible, can you still SSH in or reach the observer URL?

[shadow zealot]

I access HA via App on my Android phone. After rebooting, all works fine.
When the crash happens, of course I no longer reach HA via App nor via Browser and IP Adresse. The LED on the Raspi is red.

[raven forge]

Please answer all my questions 😦

[humble mirage]

I've learned that when you ask more than one question people tend to only answer the first

can you still SSH in or reach the observer URL?

[raven forge]

I usually ask two or three times and then give up

[humble mirage]

Looks like they're offline... maybe an answer will come later

[vast turtle]

hey guys, how does one go about reinstalling HA (supervised) via docker/portainer?

[empty bloom]

you don't

HA supervised is separate from docker/portainer

you run HA supervised or you run containered

[vast turtle]

So i believe i have everything setup in docker (which i can see in Portainer)

[empty bloom]

ok

Then what's the problem?

[vast turtle]

Basically i want to reinstall HA, e.g. be reset to factory settings

[empty bloom]

delete your configuration

[vast turtle]

We moved to another property and everything is cluttered. i saved config.yaml

[empty bloom]

and restart your container

or stop your container, delete your configuration, start container

make sure you delete hidden files/folders as well.

[vast turtle]

i have no idea if i would have those

[empty bloom]

you will, your config folder will have a hidden folder called .storage which stores all the UI integrated stuff

[vast turtle]

i would really love to start with a fresh install though, i never re-did the those install and ive learned alot since then

[empty bloom]

anyways, if you delete (or move) your config folder to an empty folder, HA will recreate everything

[vast turtle]

hmmm

the whole folder? lets see

[empty bloom]

well, the folder still needs to exist, the contents need to be gone

[noble crater]

Hello, could you help me? I've been stuck on this for 3 days:

root@RaspberryMaison:~# apt install os-agent_1.5.1_linux_aarch64.deb
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
E: Unable to locate package os-agent_1.5.1_linux_aarch64.deb

[humble mirage]

Mostly ... don't use Supervised. But if you must then you have to use Debian 12 itself

[noble crater]

I'm on debian 12

[humble mirage]

And you read https://github.com/home-assistant/architecture/blob/master/adr/0014-home-assistant-supervised.md?

[noble crater]

I'm going to give up

[humble scaffold]

@noble crater When using Discord's Reply feature it defaults to pinging the person you reply to, which can get frustrating for the target. Use Shift + click on the Reply option, or click @ ON to @ OFF to stop this - on the right side of the compose bar.

You have to change this every time (thank the Discord devs for that).

[humble mirage]

Using HAOS or Container is the correct answer

[noble crater]

i want add-on ... and install other programme

[humble mirage]

@noble crater you don't have to ping me on every reply you make

#installation-archived message

You don't get add-ons and other programs, even with Supervised

[noble crater]

ok sry

[sinful siren]

I currently have a Raspberry Pi running PiHole (Adblocker), would there be any conflict running that and Home Assistant on it?

[humble mirage]

You'll be fine, with Container anyway

[tropic hill]

I'm migrating from supervised to docker. Is the contents of /usr/share/hassio not relevant outside of the /usr/share/hassio/homeassistant for the container?

[cosmic olive]

how long is the "preparing home assistant" going to take?
installed it on vmware, pretty good PC, not high end but alot better than raspberry.
Been in "preparing" state at least 2h now

[tropic hill]

@humble mirage Thank you. Anything else I should be aware of before this migration? I understand addons go away.

[lean skiff]

Update to this, a partial restore of everything except home assistant itself (aka just folders and addons) works (ha addons shows addons) but obviously isn't enough to get me fully working again. Going to try ripping stuff out of the ha core backup until the core one works but gosh the logging isn't great as to when it fails

[humble mirage]

Tinkerer Thank you Anything else I

[strong skiff]

Good day everyone!

I'm stuck with a weird issue that showed up about a week ago (i think) where the Hass UI shows a different amount of ram in the system than there actually is

My setup is as follows:

• 64gb RAM bare metal
• Running Proxmox
• HASS OS inside a VM with 8gb assigned to it

Now when i fully restart HASS it shows 8gb (in the UI) at the start but it quickly goes down to 1gb (using about 800mb of it)

This (probably) results in high swap memory usage turning the whole system in a performance nightmare

Where should i start to look for issues? I'm not that familiar with HASS OS subsystems

[tender wren]

Is Proxmox configured to use Dynamic Memory for that VM?

[strong skiff]

I dont think so, how can i double check that?

proxmox --> hass vm --> hardware --> memory shows: 1.00 GiB/8.00GIB

[tender wren]

In the VM settings should be a Memory tab. There are fixed memory or automatically allocated memory settings.

[strong skiff]

i will be dammed, ballooning is enabled, and minimum is 1024MiB

[lean skiff]

Removing the database got me restored. Kinda sucks to lose the energy data but at least I'm back up and running. Seems like the only thing that didn't come over was the Network storage settings

[strong skiff]

I assume it should be disabled @tender wren? Still weird since it has been this way since i setup everything

[tender wren]

It would explain why the memory is shrinking.

I'd go with a fixed memory setting

[strong skiff]

Agreed, there is enough memory available anyhow

[raven forge]

It should only really take effect if the host/node's memory usage is >= 80%

[raven forge]

Can you show a screenshot of that memory behavior?

[strong skiff]

Cant upload images here

[humble scaffold]

Please use imgur or other image sharing web sites, and share the link here.

Image posting is blocked in most channels to discourage people from sharing text as images. Sharing text as images assumes that everybody sees the world as you do, which isn't the case. Some people are colour blind, or have visual impairment that means they can't make sense of an image of text.

[strong skiff]

I have just checked it out and that is indeed occurring, some other VM is using way more than it should

[strong skiff]

Plus its nothing special, just that the memory "installed" in HASS UI goes down very quickly

Now that i dissabled ballooning, it stays steady at 8 GB

[raven forge]

I don't consider that a proper fix but I can't tell much without those screenshots. See 10.2.6 in proxmox's docs

[strong skiff]

Agreed, but its probably shrinking it because HASS doesnt use enough, like even when it went down to 1gb, it only used 700-800mb

So maybe Proxmox thought it doesnt need all of it (since HASS kept the usage well below max)

[shadow zealot]

I'll need to wait for the next crash, then I'll try to access the Raspi via ssh.

[raven forge]

You could leave balloning enabled and simply set both min and "max" to 8GiB.
Interestingly if I assign 8GiB I can reproduce the less-than max behavior but not if I give it 16GiB 🤔
I tested with 1024-8096GiB and when it was booted free -h showed about 2G total memory.
I used the munch.c from here and it was too fast and got killed (at 4G IIRC) before the memory could be given to the VM: https://www.linuxatemyram.com/play.html
So from what I can tell the VM gets as much memory as it needs within min and max but it can take time. Usually not a problem.
I'm too tired (so expect bogus information) to test this further right now

[strong skiff]

"gets as much memory as it needs" doesnt really work properly it seems, because it needs more but doesnt get it

and it has been running on 1gb for days

[copper steppe]

Unless you're running a crazy number of addons, 2GB is probably fine for HAOS on proxmox

Mine has 3.5GB allocated, but only is about half of it.

https://imgur.com/zlMIx6b

HA docker container (within haos) is using about 500MB

https://imgur.com/h4dHyon

[late fable]

Anyone have any advice for pulling files from a broken install which reboots every 20 seconds?

The supervisor runs and the addons seem to run but that's about as far as I can get

[raven forge]

Does the OS reboot or does only HA restart? On what kind of storage is it installed?

[raven forge]

I don't trust proxmox's memory stats. They are often misunderstood, include caches which over time tend to use what's available, depend on configuration and so on.
Without ballooning the memory stats will become kinda useless.
But yeah its memory usage mostly depends on addons and can be very low. I mean people are running it on old raspberry pis.

[copper steppe]

Um, in the last few days (the graph) it's probably only been restarts due to me changing yaml config. I don't think there's been a haos update (or powerfail, or proxmox update) in a few days

[raven forge]

It's possible I might just be wrong 😁

[copper steppe]

You can always login to haos and use free etc

➜  ~ free -h
              total        used        free      shared  buff/cache   available
Mem:           3.3G        1.0G       97.4M        4.7M        2.2G        2.3G
Swap:          1.1G      419.0M      701.9M

[raven forge]

That's what I'm testing right now with various settings like balloning off, dropping caches in-between and so on. I don't use HAOS in production so I don't have long-term data.
Ah, I see. Yours does use most of the memory it has

[copper steppe]

buffers and caches. not real usage.

[raven forge]

It is to proxmox's stats hence why I wouldnt use them on their own

[copper steppe]

only 1G actually used. The 2.2G are sauce

Yes, it is to proxmox, and the host, but the VM thinks it's spare/free so uses it.

Hence the reason it will consume all you give it, eventually

My ~2 months of proxmox graph shows 2.0-2.5G all the time

I don't really know how proxmox categorises usage vs total..

Looks like I've got ballooning on. No idea impact of that.

i've got min=max=3500MiB

[raven forge]

I'm not sure how the balloning driver will allocate memory. how often, how much, etc either. I was planning on researching that today

[copper steppe]

Didn't you say you had 64GB of ram? Who cares then? 🙂

Give it 4, or 6, or 8. You'll be golden.

My proxmox boxes only have 16, and also run opnsense, omada, pihole, plus a few dashboardy-type containers.

And box is <50% used ram, and ~6% CPU

https://imgur.com/bCUBvH4

[neat sapphire]

Hi, i have a problem doing a update from Home Assistant Operating System 10.3 to 10.4
im using a generic x86-64 system, i'm new to home assistant.
im running a hp-microserver with 8gb ram and 1 tb ssd i have a corel tpu for frigate and zigbee3.0 stick.
can some guide me how to setup my log file so i can see why the system reboots if i try to upgrade to 10.4

[raven forge]

You can install the community SSH addon: https://github.com/hassio-addons/addon-ssh
With it you can run ha supervisor logs or ha host logs.
You might have to enable Advanced Mode in your profile: https://my.home-assistant.io/redirect/profile/

[neat sapphire]

and how will this help me, do i have to try to install de update and after it crashes and the system reboots , run the ha host log command in the terminal window?.

[copper steppe]

Is there no screen?

It is expected to reboot.

[neat sapphire]

after reboot the update is still there 🙂 and i'm still on version 10.3

[copper steppe]

Are you upgrading from web button? Try from haos cli?

[left parcel]

Had a power cut last night and when power was restored about 5 mins later all my Zigbee lights came on. If I login using homeassistant.local:8123 should I be able to control all my Zigbee devices without an internet connection?

[neat sapphire]

@copper steppe and what's the command for update from cli?

[copper steppe]

There’s help built in.

I assume it’s: (ha) os update

If you’re in the cli shell thing you don’t need the “ha”

[strong skiff]

Correct, thats why i just gave it 8gb to start with, had enough to spare anyhow. But since a different VM started using way more than normal the total host usage exceeded 80% and thus (according to @raven forge) ballooning started to go into affect and shrinked the HASS vm to 1gb (set as min) why it went down so low i dont know since more was definetly needed

[neat sapphire]

yes its the os 10.3 to 10.4 problem.
i'm using the terminal ssh windows addon to login to ha
Just adding i'm a noob to this.
if you tell me what to type in the terminal window, i can try that

[raven forge]

I did. ha supervisor logs and ha host logs

[neat sapphire]

i don't think there is anything special in the logfiles that shows why it reboots.
can i just past the outcome from the logfile here ?

[raven forge]

Sure but verify there's nothing personal in there

[humble scaffold]

Please use a code share site to share code or logs, for example:

• https://dpaste.org/ (select YAML for the language, and consider picking a longer expiry)
• http://pastie.org/ (select YAML for the language)
• https://paste.debian.net/ (you guessed it, select YAML as the language)

Please don't use Pastebin, since it can randomly add spaces to the main view. Please also don't share text as images since it makes it harder for people to help you. Remember that others may have colour blindness, impaired vision, etc.

[raven forge]

The idea is to have a limit that is big enough to run all the VM's services. If the host has enough resources it can get more if needed

[neat sapphire]

https://dpaste.org/TREVg
this is the sup log file

https://dpaste.org/oettw
the host log

[raven forge]

It might make sense to tail the log during an upgrade with docker logs -f hassio_supervisor.
Or you can check if there's older logs if you do docker logs --since 24h hassio_supervisor 2>&1 | less

[barren hedge]

Any idea how I can debug what's going on with my haos installation on rpi4. Using correct power supply, 32gb sd card, ethernet connected and the haos_rpi4 image - flashed the sd card using balena. When turning on the rpi i get no digital output on my monitor, the monitor keeps turning on/off. not seeing the device on my local network either, kinda stumped at what to do

nevermind, discovered it was an issue with using nespi 4 case with haos

[winged vale]

Hi, I try here...Can I use/add a webbrowser on a HAOS Pi4 on a local display? I would like to use it for a simple display and switch function. So far the display I have is useless... And I would like to have a display permanently on. The Pi is running for HAOS, so why can't I use the attached display for that? Running another device for that seems to me a waste of recources. Any ideas?

[neat sapphire]

@raven forge it said installing of home assistant operating system 10.4 Succes
initialize host reboot using logind

[echo compass]

Does anybody know how you install the MQTT broker on docker

[frank perch]

Not with HAOS no

[manic bane]

Look at the mosquito image

[echo compass]

What mosquitto image?

[manic bane]

https://hub.docker.com/_/eclipse-mosquitto

A quick google can help

[echo compass]

And that's the official ✌️ App ✌️ Store?

[manic bane]

What’s with the emojis lol

Also it’s not an app store, it’s docker hub. There is no official app store

But that is one mosquito image, made by the eclipse foundation

[echo compass]

Yeah and how bout Node Red and Home Assistan? I don't find any - official - oner

Just node.js

[manic bane]

Home assistant is linked in the installation guide

Node red just look at their installation guide too

https://nodered.org/docs/getting-started/docker

[echo compass]

Awesome. And Pi Hole as well?

[manic bane]

I presume they have an install guide you can look at

Docker requires you to go out and investigate a little yourself

[echo compass]

The plan is using docker on my pi, so it doesn't mess up with each app

Do you know any good backup image if we're already talking?

[manic bane]

Backup image?

[echo compass]

A backup service, for example for home assistant, I can use.

[manic bane]

That can be whatever you want. It’s your own personal choice, docker you mount the directories onto the local drive

I just have rsnapshot installed and backing up to my nas

Not installed as a docker image, just a standard install

[echo compass]

I really have no fcking idea which backup service would be useful for a priovate smart home.

[manic bane]

Well it depends on how you want to back up

Whether you want to use a cloud storage, whether you want it stored to a nas or to an external drive, etc

[frank perch]

no disrespect but are you sure docker is the way you wanna go then?

[empty bloom]

full disrespect, goodmorning

Joking aside, IIRC container has a backup option

[frank perch]

well dang, full?

[empty bloom]

ok ok partial

[frank perch]

petro please go have coffee

lol

[echo compass]

What's the most safe one on a budget?

So, as in DSGVO-safe?

Obvious it would be NAS but it's expensive.

[manic bane]

Dsgvo?

[empty bloom]

https://allaboutberlin.com/glossary/DSGVO

[echo compass]

In English it would be GDPR. It's about data safety

[empty bloom]

that's something you'd have to research based on what you're buying

I woudl assume anything offered in germany would be compliant

[manic bane]

Pretty sure even google drive is compliant

But if you're worried just encrypt the data before uploading it

[hazy vine]

if i make a backup from a current release version(docker), can that be restored with version(os) that might be behind?

[grand pivot]

All that matters is the core version

And if there are things in the new config that aren't valid in the old config

[hazy vine]

im not sure what version the haos core version is as i set it up a couple weeks ago and has just been waiting for me to do the restore with my ha container backup. ive updated the ha container to most recent since then.

[grand pivot]

"A couple of weeks ago" was either 2023.7 or 2023.8. Both should be fine with a 2023.8 config.

[worn ore]

Hey everyone, goodafternoon! Just wanted to check something real quick. I'm currently running Home Assistant Supervised, but I'd like to move towards Home Assistant OS just to not bother too much with the whole Debian aspect of things.
I was wondering, if I were to re-install my server with Home Assistant OS, would I be able to restore my backup from Supervised or is that not possible?

[tender wren]

Yes

[worn ore]

Ah perfect! Thought I'd ask because it would be quite a hassle to redo all the configuration. Thanks Jorg, much appreciated! 😄

[sand sorrel]

hi. Just to double check. If I install HA as TrueNAS app, I can't have addons? So it would be easier to run the whole VM?

[frank perch]

yeah TrueNAS uses container, which you won't have 'addons' but you can still get the same software, and yes most people find HAOS easier

[sand sorrel]

like node-red, zigbee2mqtt?

[empty bloom]

You have to build the containers yourself unless truenas is one of those BS UI's that force you to use their pre configured containers

[sand sorrel]

ok, haos it is then

[raven forge]

Yeah if you want addons you want HAOS

[worn ore]

Just a quick follow-up. So I installed HAOS and was indeed able to restore a supervisor backup. Partially. Had to fiddle a bit to restore most of my addons. My NUC went haywire 😂. That being said I think I’m almost there. All in all the basis of the installation worked as expected

[tender wren]

The host usually needs a full reboot for the add-ons to work properly

[frank perch]

I find restarting HA then rebooting works best

[worn ore]

Yep that's indeed how I did it as well 😄

Fixed now. NUC is nice and calm and everything is A-Ok

[neat sapphire]

@raven forge https://dpaste.org/Aw16M

[raven forge]

Strange. So it installs 10.4, reboots and it's 10.3 again? https://my.home-assistant.io/redirect/info/

[neat sapphire]

yes

[gritty peak]

https://tenor.com/view/timewarp-gif-6237247

[spare halo]

Hi! I could use some help regarding my almost new HA installation.
I am running a docker container of HA on my Synology NAS and want to include some add ons like InfluxDB. Therefore I need to change the configuration.yaml.
https://www.home-assistant.io/docs/configuration/ says, the easiest option to do so is by using the "Studio Code Server add-on". After following the link and accepting to open it with my HA installation I get the following (german) error: "Diese Weiterleitung wird von deiner Home Assistant-Installation nicht unterstützt. Benötigt entweder das Home Assistant-Betriebssystem oder die Home Assistant Supervised-Installationsmethode." or in english: "This redirection is not supported by your Home Assistant installation. Requires either the Home Assistant operating system or the Home Assistant Supervised installation method."
Isn't the HA OS installed when running the docker version?

[high token]

no, the docker version is literally just the container for HA itself

HAOS is an Operating System that is installed on the device itself or as a virtual machine

if you're familiar with docker and have HA container installed, you can then run the addons you want standalone as containers (addons are anyway special containers)

[spare halo]

I am not sure if I got this right. Could you please explain this to me using the InfluxDB as an example? I already got another container with InfluxDB up and running. Do I just have to edit the configuration.yaml through the file system and add the necessary parameters like explained at https://www.home-assistant.io/integrations/influxdb/ ?

[limpid yew]

Hi, somehow HAOS upgrade from 10.3 to 10.4 shredded my "A" slot. What can I do to recover it? Is there e.g. a mechanism in place to copy over the B slot or sth?

[high token]

Yeap, if you already have an influx container, you just need to create a database there and then point HA to it by adding the right information on your configuration.yaml as per instructions (db name, IP, user, pass etc)

[high token]

Shredded as in what? It's no longer booting? It should have moved to the other slot if A is not working

[limpid yew]

@high token yes, no longer booting. the other slot still works but I don't want to risk trying an HAOS update there while A is not working in case that breaks as well. So I'm looking for a way to recover A before I proceed

[burnt prawn]

hi anyone using HP 400 g1 mini for your HA setup's?

i cant boot them without monitor! any help?

[loud hemlock]

about to bail on a pi4 for haos, having too many issues with my new sonoff p stick

i have apu2s available

uses an msata hdd

whats the best way to install?

[final socket]

Ok so it looks like you might run into some issues since the APU boards use coreboot firmware instead of uefi? But there's a fix for that.
https://community.home-assistant.io/t/add-support-for-apu-devices/449610/3
Boot it from a Live USB, use dd or etcher to write HAOS to the SSD, then mess with the grub config as described in that forum thread.

[plush quartz]

hi.
reading the guide on using duckdns to get https, how I can I get that to work while using http on local network?
seems I can't get http anymore on my local

[humble mirage]

Use a proxy server for SSL

[plush quartz]

ok. trying the nginx proxy manager and when I select a dns provide, duckdns it shows Credential file content and then dns_duckdns_token=your-duckdns-token what do I do here?

[humble mirage]

No clue, don't use it

Traefik for me

If that's an add-on there's a channel for help

[loud hemlock]

im about to try this

i already used etcher to write the ssd

im going to try the grub stuff, not my strong suit but will try

i dont understand though, if i boot from live cd, how am i viewing the OS?

its headless

this might be too advanced for me

i was able to boot to the live cd but there's no interface, ill try again with basic display drivers i guess

ok i cmdline on the rescue disk but i dont see the 60GB SSD that has the HAOS on it when i df -h

Mount /dev/sda1 (EFI boot partition) to /mnt

i dont know how to do this

any advice

holy shit i did it

jesus rollerblading christ

im gonna make an image of this disk because whenever it fails i wont remember how i did this

i guess i really just need backups of my home assistant, in a pinch i could restore it to anything

it works, wow

both zwave and zigbee antennas are running

this is nuts

[crimson furnace]

Wondering a few things about a new install for Proxmox / HA / Frigate with Coral USB

I’ve got a new machine for Proxmox. I was going to install HA with the Frigate Addon (same as I have now on my old setup) and pass through the Coral USB.

But I realised I probably want to have a larger drive for frigate, can you install HAOS on one drive and then have Frigate save to another drive?

Then I started to wonder if that was the best way to have this setup. I saw some mentions about installing Frigate in a separate VM but this would be something I’d have to read more about. Never done anything like this before.

Just wondering if anyone had any recommendations/advice on what the best way to go about this? 🙂

[humble mirage]

But I realised I probably want to have a larger drive for frigate, can you install HAOS on one drive and then have Frigate save to another drive?
With a separate VM for Frigate

[crimson furnace]

I saw this thread - https://community.home-assistant.io/t/google-coral-usb-frigate-proxmox/383737

Is this normal or is this a lot of workaround stuff I need to fully understand? To my eyes it seems very complicated but perhaps thats just because I don’t understand it yet.

[humble mirage]

#cameras-archived knows about Frigate. I don't use Proxmox

[crimson furnace]

Ah okay thanks as always Tinkerer!

[loud hemlock]

seems my luck ran out with the HAOS image on an APU2

i need to have wireless and the card wasnt identified by HAOS