#installation-archived

1 messages ยท Page 61 of 1

tender wren
#

Have you added the proxy's IP (range) to HA's trusted proxies?

# 
http:
  use_x_forwarded_for: true
  trusted_proxies:
    - 10.0.0.200
    - 172.30.33.0/24

in the configuration.yaml (IPs / IP ranges have to be changed accordingly)

polar flicker
sharp copper
#

i am figuring out how to properly setup a home assistant server for my parents that i can control from my own home

#

are there build in options?

#

or is the best way by using duckdns and access it that way?

long wyvern
#

Unless you've exposed the VM's network stack to the network with bridging, anyway. That's all configured with whatever you use for container management

long wyvern
#

Personally, I use a FQDN from moniker and public DNS is handled through afraid.org. OPNsense handles updating the DNS records when I get a new lease, and I can use a wildcard cert for HTTPS with Lets Encrypt and ACME client.

#

You could theoretically use a site-to-site VPN initiated by the client, then local resources would handle routing

sharp copper
#

Okay cheers thank you

versed vapor
#

can we shutdown ha without shutting down haos?

tender wren
versed vapor
#

thanks!

polar flicker
#

Well I made progress I think? It's not erroring out. But it is spooling for a /long/ time

#

Damn. It spooled for a solid 3 minutes and then gave the 502 error

#

when I log it I get 2024-03-17 16:56:51.636 ERROR (MainThread) [homeassistant.components.http.forwarded] Received X-Forwarded-For header from an untrusted proxy 192.168.xxx.xxx (not sure what this IP is to)

long wyvern
#

that IP address belongs to either your computer in front of you, or it's the internal IP address for the nginx proxy

#

either way, it's telling you that IP is proxying requests, and it's not in the trusted list on HA

#

FYI, 192.168.. are reserved internal IP's. No privacy concerns posting the entire IP address on the internet

#

same with anything in the 10.x.x.x block, or 172.16

polar flicker
#

Cool I just found its the ngnix ip. So I just need to find the subnets for it right?

THank you for all your help and info!

long wyvern
#

does it give you the full 192.168.whatever ip address? Just add that single IP to the trusted_proxies list

#

It's also smart to set up a DHCP reservation on your router to give dedicated machines (including VM's) the same IP address each time

polar flicker
#

yes 192.168.128.2. I put that in the reverse proxy and i'm still getting the 502. While researching I saw something about needing to update the upstream_app from home assistant to an ip.... is that what i'm missing at this point?

long wyvern
#

128.2, that's different

#

I'm not sure what 'upstream_app' is referring to. Maybe the host container?

#

Maybe it's referring to set $upstream_app <ip_address> in nginx proxy.conf

polar flicker
#

I think so yeah

#

What up address is it looking for? The internal one for my network, the ngnix, the bridge a different one?

#

I got it! I needed my machine up there.

potent perch
#

Is there a way to mount a usb disk for backups etc? or is it just samba/nfs?

potent perch
long wyvern
#

Look in the supervisor tab -> system under hardware.

#

I think the File Browser addon supports external devices

marsh dew
#

I'm new to HA, set up the compose file but after the container is created I have nothing on port 8123

#

2024-03-18 13:15:47 s6-rc: info: service s6rc-oneshot-runner: starting 2024-03-18 13:15:47 s6-rc: info: service s6rc-oneshot-runner successfully started 2024-03-18 13:15:47 s6-rc: info: service fix-attrs: starting 2024-03-18 13:15:47 s6-rc: info: service fix-attrs successfully started 2024-03-18 13:15:47 s6-rc: info: service legacy-cont-init: starting 2024-03-18 13:15:47 s6-rc: info: service legacy-cont-init successfully started 2024-03-18 13:15:47 s6-rc: info: service legacy-services: starting 2024-03-18 13:15:47 services-up: info: copying legacy longrun home-assistant (no readiness notification) 2024-03-18 13:15:47 s6-rc: info: service legacy-services successfully started

#

so it ran successfully? But nothing started

#

that's all there is in the log

tender wren
#

You might want to share your compose file

raven forge
#

What OS? How does the compose file look like?

humble scaffoldBOT
#

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

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.

marsh dew
# raven forge What OS? How does the compose file look like?

Windows Docker Desktop, I'm using a copy-paste from the HA tutorial:

services:
  homeassistant:
    container_name: homeassistant
    image: "ghcr.io/home-assistant/home-assistant:stable"
    volumes:
      - D:\Docker\homeassistant:/config
      - /etc/localtime:/etc/localtime:ro
      - /run/dbus:/run/dbus:ro
    restart: unless-stopped
    privileged: true
    network_mode: host```
#

nothing else added to it

#

I tried ports: - "8123:8123" and it didn't work either

raven forge
#

There are no official HA docs for docker on windows. It does not support host networking.

marsh dew
#

Oh I see

grand pivot
#

your only sane option on Windows is HAOS in a VM

marsh dew
#

I have a pi4 with docker, wanted to just try out HA locally before I dive deeper

grand pivot
#

which is the first option when you follow the Windows installation instructions on the main HA site

empty bloom
#

check out HA dev container

#

if you want to try it out on windows, but it requires VSCode

#

but it runs HA fine on windows

#

localhost:8123

raven forge
#

I'd recommend VirtualBox on windows. Reason is that it supports snapshots which VMware free does not and it supports USB passthrough which Hyper-V does not.

raven forge
empty bloom
#

no idea

#

Probably not

#

it doesn't discovery things on my network

raven forge
#

It's based on docker too so my thought was that it's no better than just using port mapping rather than host. But since you actually use it I thought it best to ask.

empty bloom
#

Let me check out the dev container settings

grand pivot
#

It's just a docker container

empty bloom
#

nothing special, just 2 ports

grand pivot
#

Same limitations

raven forge
marsh dew
raven forge
#

Yeah that overrides the port mapping. I think it's also logged/printed somewhere.

humble mirage
#

Keep in mind that many integrations won't work

#

Anything that needs mDNS, UPnP, or other broadcasts/discovery magic

marsh dew
#

yeah I'm rerunning a compose on my pi right now, it was fun to click around the UI to familiarize myself with it though

grand pivot
#

The VirtualBox install that Impact suggested is the best course on Windows

marsh dew
#

like the container takes less than 10 seconds to pull and set up on my desktop pc, and exponentially as long on the pi4b

rare valley
#

Hi, I'm having issues with the HA supervisor all of a sudden. I can't start addons, create/apply backups, update anything, and a bunch more things. Any ideas how I can get it working again? Logs from supervisor: https://pastebin.com/01A4kZZN

For context I was not even at home when the issue started and did not make any changes to my configuration so I don't think anything I did specifically casued this. I did make some changes to my network but it was working fine for at least a few days before this started. HA is connected via ethernet to my router and I can access it externaly with no issue so I don't think it's network related but I could be wrong.

Installed HAOS on a RPI 3

placid sage
#

I want to migrate my HA install from RPi to Odroid. Is there a guide on how to do that easily. Can't find it in the Help pages. Perhaps I'm looking at the wrong pages.

humble mirage
#

Take a backup, download the backup, restore it to the new install

placid sage
#

Thanks, it really looks simple.

rare valley
#

I see there's a CLI command but I can't start the SSH addon due to the supervisor issue so unless there's some other way I can access the CLI I don't think that's an option

merry marsh
#

hoping for some quick pointers on upgrading, I moved from a install that was over a year old, by just recreating my docker via portainer, as a result I need a updated zwave js server.

It so long since I set all this up ( and becuase it just works ) I have lost track of the moving pieces.

I was running docker for zwave, so I hit recreate and pull on that, expecting it to upgrade in place.

However, it has created a new zwave docker sat in created, and kept a copy of the old renamed to zwave-old sat in exited, that was unexpected.

#

Any tips appreciated

#

this is the issue I was trying to address via upgrade

"Invalid server version: Z-Wave JS Server version (1.24.1) is incompatible. Update the Z-Wave JS Server to a version that supports at least api schema 33"

#

Happy to read and work through anything, just looking for pointers of what to read / where, my google fu is weak today

humble mirage
#

#zwave-archived can help with the Z-Wave side, but it sounds like you need to update

merry marsh
#

Yeah, its the update process for docker based I want to sanity, for some reason the recreate and pull did not go well, so I want to go back to basics.

rare valley
# rare valley Hi, I'm having issues with the HA supervisor all of a sudden. I can't start addo...

Just tried booting into safe mode but seems like it's failing to reboot unless I manually unplug the pi

[22:38:08] WARNING: Maybe found an issue on shutdown
s6-supervise watchdog: warning: unable to open max-death-tally: Bad message
s6-supervise watchdog: warning: unable to write supervise/death_tally: Bad message
s6-supervise watchdog: warning: unable to spawn ./finish: Exec format error
s6-supervise watchdog: warning: unable to write status file: Bad message
s6-supervise watchdog: warning: unable to spawn ./run (waiting 60 seconds): Exec format error
#

I'm guessing a fresh install is my best bet at this point, but just trying to see how I can avoid losing data (see above comments)

raven forge
rare valley
raven forge
#

I meant that one should already have proper off-site backups for cases like this. You could try if ha supervisor repair and rebooting does something. You can access the CLI via keyboard and monitor too.

rare valley
#

Gotcha, unfortunately I think the only external keyboard I have available right now is Bluetooth. Is there another option for accessing the CLI without the SSH extension?

#

And yes I will definitely make sure to have a proper off site backup for the future, definitely should've done that before.

ivory gale
#

I am getting this error on fresh installation on RPI4, it is the very first time this has happened, everything was running smooth before.

'Updater.fetch_data' blocked from execution, no supervisor internet connection'

#

is there anyway to download an older version of RPI, the last update was working fine until it just stopped working.

Maybe 11

#

The LAN cable is connected.
port is working fine.
No issues from the backend.
The "wait for 20min" screen shows but it ends up with this error.

rare valley
#

Ok so the SD card is definitely somehow corrupted. I'm trying to erase and reformat it as a standard SD card so I can start from scratch but can't figure out how to do it from my computer (macOS). I tried a bunch of stuff in disk utility and it just does nothing. Anyone know how I can reformat it? Some 3rd party/CLI tool?

#

I am wondering if there is physical damage to the SD card but want to try that before I just go buy a new one

ivory gale
vagrant plume
#

Hello, I am running home assistant on my raspberry Pi 4 Model B. Yesterday I updated the Home Assistant core from my phone and closed the tab. Later that day I noticed my Pi was boot looping. When I plugged it into a screen I got this log among others I have tried changing the power adapter & cable to no avail.

Card did not respond to voltage select! : -
raven forge
#

Try plugging the SD Card into your PC and see if it's alive. Assuming you use one, of course.
What other logs were there?

vagrant plume
#

Yes, I only have it as an image though

#

not sure how to copy them as text easily

vagrant plume
#

is it worth dual booting a linux distro to look at the sd card?

raven forge
#

You can just live boot a linux iso. I recommend putting ventoy on a USB stick for that. You can also use DiskInternals's Disk Reader on windows.

humble scaffoldBOT
#

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.

vagrant plume
vagrant plume
#

or computer

raven forge
#

On your computer. You wanted to boot linux, no? Ventoy allows you to create a USB stick with multiple bootable images.

vagrant plume
#

I see

raven forge
vagrant plume
raven forge
#

It's possible you can press E during boot to edit the kernel args and remove quiet if it's there but I'm not sure if that works with U-Boot. Unfortunately I have no clue what could cause this with the current information. Hmm seems quiet is not used anyways.

humble scaffoldBOT
delicate valve
raven forge
#

Please show the OOM log and check with glances what uses the memory.

delicate valve
#

Is it on the VM?

raven forge
#

In the system log of whatever ran out of memory. How do you know it ran out of memory?

delicate valve
#

It showed an error on the VM

#

Let me upload the pictures

delicate valve
raven forge
#

The process it mentioned was killed because it ran out of memory but it might not be the only thing.

delicate valve
raven forge
#

Yeah your instinct was correct. It's definitey a memory issue. OOMs are quite bad and can often leave a system in a bad state. How much memory is allocated to the VM? How much does the PC have?

delicate valve
#

2048 on the VM and PC has 8gb.

#

I can up the memory to like 5gb to VM or whatever amount that is green it allows me

raven forge
#

How much you need depends on what addons you use. I'd try 4G for now. I can tell you more if I see the glances output.

delicate valve
proud pilot
#

my HA web interface will not load now....help

#

HA has been unstable for me since the latest updates of OS ver 12.1 and core 2024.3.1 i've been nursing it by doing a repair to supervisor and restarting which seems to fix it, but now the web ui will not load at all, but the banner on the cli looks fine. I don't know what to do at this point...

empty bloom
#

if you have a backup, reinstall and restore

proud pilot
#

If i have to do this from cli, I'll need a step by step to do that...

#

or to even check if a backup exists..

empty bloom
#

You don't move your backups to a separate device?

#

Do you have samba addon set up?

proud pilot
#

no

#

and no i don't move my backups

#

i did a quick search and found the common tasks doc...i used a command "ha core update --version 2024.3.0 --backup" ...i don't know if a backup exists for 2024.3.0..but i tried it... at this point it says "Processing..." I presume that means it's restoring?

empty bloom
#

backup means it's creating a backup

#

you don't want to take a backup of your broken state

proud pilot
#

I cancelled it...

#

I found a backup list command, but the screen cuts off the full list of backups...

empty bloom
#

you can scroll up

proud pilot
#

how? I have the HA prompt, mouse wheel does nothing, and the up arrow cycles previously used commands...

empty bloom
#

are you plugged directly into the system? or are you SSHing in?

proud pilot
#

I use Oracle VM virtual box

empty bloom
#

Sorry, not familiar with that

#

but if you have the SSH addon installed, you can access it from a windows machine with putty and it will let you scroll

#

Can't speak for whatever system you're using

proud pilot
#

i do see a core_2024.1.3 (of what i can see)...is that too old?

empty bloom
#

it's not too old but you can attempt to restore that.

#

I'd try to get the backup off that device and on to something else.

proud pilot
#

If i can get to the UI I can work better to do that...thinking of switching to docker desktop, but I reaaaaaalllly do not want to recreate all this stuff again...

raven forge
proud pilot
#

The backup restore was only a partial, but i rebooted afterwards, HA core states it's 2024.3.0 , ui is still not accessible

raven forge
#

Check ha supervisor logs and ha core logs.

#

Sorry I just scrolled up. You can do this to be able to scroll.

# To get a proper shell
login

ha backups list | less
proud pilot
raven forge
#

Nothing interesting but it's only a part of the logs.

proud pilot
#

ah, wait, it does make a difference, but the output is strange looking... I am able to scroll now

raven forge
#

You can also run ls -l /mnt/data/supervisor/backup/.

proud pilot
placid sage
#

The problem is found. it's bc of sd card. I just changed it.

raven forge
#

Do share.

placid sage
#

I think the old one was slow

raven forge
raven forge
proud pilot
#

I presume i would need to know if it's a full or partial backup?

#

and the year...are these all core backups? ... I'm not fully sure what I'm looking at...trying not to make more of a mess...

placid sage
raven forge
proud pilot
#

I don't see any full backups...

raven forge
placid sage
raven forge
# proud pilot I don't see any full backups...

You said you only did partial backups. You can also run this which only prints out the slug and type of each backup

ha backups | grep -E "type|slug"

which should save some space on your display.

proud pilot
#

they're all partial backups... I picked one and did a restore, it says it's done but throws an error that it's only a partial backup. I didn't say anything about what backups i did..it's whatever is set in HA by default.

#

well...interestingly, I can get to the ui from the hass agent, but not the browser...

#

I can get to the web ui from my server, but not my office machine...that's progress i suppose

#

ok now it's accessible again from my office machine... I didn't do anything, just waited.

#

I have a pending update to go to the latest core again.... I'm f*ing terrified to touch anything right now..

#

I clicked on add-ons and it crashed...

#

and now it's back....wtf is going on!

#

It's crashed again, clicking on 'add-on's seems to make it crash

brazen herald
#

Have you tried booting into safe mode?

proud pilot
#

If i can get back to the ui again, i'll try that...

#

Restarted the VM and was able to get it to come back, now restarting in safe mode

#

and for some reason I can only get to the UI on my server, but not my office machine

brazen herald
#

I believe to do it via the CLI you need to set the supervisor option for debug, but I don't know for sure so don't trust me on that

brazen herald
proud pilot
#

i tried another browser and it works...caching issue probably, not related.

#

So i'm in the ui in safe mode on my server, now what?

brazen herald
#

Well if the UI stops going away, then your problem is with something installed ๐Ÿ™‚

proud pilot
#

well...ring mqtt kept coming up in the logs...maybe that...

#

After removing ring mqtt i am able to nav to the addons screen in normal mode now

#

I notice that the hacs icon is missing from the sidebar... I've been getting weird things like this happening lately... where things just randomly disappear.... Several todo lists, a flex horseshoe card, and now the hacs icon

grand pivot
#

Clear your browser cache

#

Beyond that, review the logs

#

Those apply to nearly any problem

proud pilot
#

well, that brought the hacs icon back, the horseshoe cards are still missing...the todo lists i consider a lost cause

#

ok...MQTT failed, i had to restart the hass.io agent on a few machines which brought the cards back

#

I think i'm cautiously good to go for now.... I don't think this is the end of it...but time will tell. Thank you all for your help

eager trail
#

Not exactly sure where to put this - since it is sort-of installation related, putting it here. I finally got my Dell Optiplex hardware all setup - after three messed up deliveies for a PCIe card to put my Intel AX210 WL 6e/BT 5.2 working. Proxmox sees the card and the WiFi - and I am guessing somewhat sees the Bluetooth (it is a Debian 12 type build). I am trying to follow these instructions but giving me a major headache -- https://www.home-assistant.io/integrations/bluetooth (specifically the https://www.home-assistant.io/integrations/bluetooth#requirements-for-linux-systems and https://www.home-assistant.io/integrations/bluetooth#requirements-for-linux-systems). I am using the TTeck script to install the HAOS VM to my system - and will pass though the WiFi and BT into the VM. How do I make sure the BT works? I never had a problem with the WiFi.

raven forge
eager trail
#

Thanks @raven forge - those I have followed. I just want to make sure there is nothing I need to do at the HOST level.

ivory gale
raven forge
#

Can you check ha resolution info and ha network info?

ivory gale
# raven forge Can you check `ha resolution info` and `ha network info`?

how to?

This has happened to me for the first time in 5 years, tried different devices, versions, everything was working find till last month, all of a sudden it stopped working, upon installing a fresh copy, I started getting this error

"WARNING (MainThread) [supervisor.jobs] 'GitRepo.clone' blocked from execution, no supervisor internet connection
23-04-04 15:58:18 ERROR (MainThread) [supervisor.store] Can't add repository core due to 'GitRepo.clone' blocked from execution, no supervisor internet connection"

#

I am able to access HA page thought the IP, the reason I get to see this error.

raven forge
#

These are commands. Run them with one of the SSH addons.

ivory gale
#

How can I when I am unable to even boot HA

raven forge
#

You did not mention that.

ivory gale
raven forge
#

So I assume you never set up SSH?

ivory gale
#

No

#

I always did so after installation,

raven forge
#

Can you run it via keyboard & monitor?

ivory gale
#

yes, I can do that,

raven forge
eager trail
humble scaffoldBOT
raven forge
#

I'm not familiar with bluetooth but hcitool is part of the bluez package. Doesn't really matter when you pass it through. The host will not have access anymore then.

eager trail
# raven forge I'm not familiar with bluetooth but `hcitool` is part of the `bluez` package. Do...

I figured that - I was just reading the HA Bluetooth help and it talks about enabling things. If you think I am good to do the VM install script and then enable PCI Passthrough of the Wi-Fi and USB Passthrough of the Bluetooth, should that be enough to allow HA to see them? and allow me to use them? This Proxmox box will be all HA. Going to do a separate VM (container for MariaDB, MQTT, etc.) .

raven forge
#

I'd simply try it. If you followed all in the links and enabled IOMMU in the UEFI you should be good to go.
Note that the HA devs recommend SQLite for HA.

eager trail
raven forge
#

If you use HAOS that should not concern you. Or in other words, HAOS should be set up for that already.

eager trail
#

Even HAOS in a VM on Proxmox?

raven forge
#

I don't see why not.

eager trail
#

OK - I will try it.

#

So far - so good. HAOS VM built and it is on the Preparing Home Assistant web page - now I wait.

#

Thanks for the guidance.

eager trail
humble scaffoldBOT
#

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.

raven forge
eager trail
#

In HA under SETTINGS >> SYSTEM >> NETWORK -- the WiFi is there.

#

If I try and add the BLUETOOTH Integration - I get: No unconfigured Bluetooth adapters found. There are 0 ignored adapters.

raven forge
#

AX210 is not on the known working list.
WiFi is actually handled through PCI-E on the AX210 while bluetooth goes through USB. You shouldn't need to pass through WiFi.

System Interface Type: Wi-Fi(PCIe), BT(USB)
https://www.intel.com/content/www/us/en/products/sku/204836/intel-wifi-6e-ax210-gig/specifications.html
Lots of people seem to have bluetooth issues:

eager trail
#

On this same box - I too was having all sorts of Bluetooth issues - the PCI card was the issue. You have to have a cable from the card to the USB header on the system board - The first card I ordered did not have it. Then they sent me another with wrong type cable to connect to system board. 3rd card was the charm - it connects. Did a pure Debian install and it set everything up with no issues - I had W-F and BT both from the installation of DEB12 - did not have to install anything.

wooden smelt
#

Bluetooth proxy is the way.

eager trail
#

Is it the '[man]' himself with all these wonderful scripts?

wooden smelt
eager trail
# 
For anyone else that may stumble across this issue, The host was 'taking control' of the Bluetooth device causing the (-19) error. The fix was to create a blacklist file within /etc/modprobe.d/ on the Proxmox host.

While the filename doesn't really matter, I named mine disable-bt.conf so that future me knows/remembers what I did and why. The file itself simply needs to contain the following line:

blacklist btusb

This will blacklist the bluetooth-usb driver from being loaded by the host. From there I was successfully able to pass-through the BT device and HA immediately detected the new device once a reboot of the VM was complete.
raven forge
eager trail
ivory gale
raven forge
#

I don't know.

eager trail
#

Thanks @raven forge and @wooden smelt - Great Resouces.

#

Now to figure out how to use it and get it to see my phone and stuff. ๐Ÿ™‚

#

6 BT speakers in the house too.

eager trail
#

Got any pointers on getting this Bluetooth to see things? It is there. I enabled 'passive' scanning. Added the bluetooth: to the configuration.yaml file and rebooted. Not seeing anything.

humble scaffoldBOT
placid sage
#

nevermind, I had to change the encryption mode on cloudflare. it works now

delicate valve
cobalt moss
#

Hi! I'm trying to install HASS OS in Synology's Virtual Machine Manager.
The installation is pretty successful but I'm unable to get past the onboarding page. It's not even loading for me.

I'm able to ping HA, but the onboarding page just won't load.
I've tried different browers and pcs and all return the same issue.
Firewall on the NAS is disabled. I am also able to ping other devices and the internet via the HA CLI. I have ensured I'm connecting to http instead of https as well.

I've already gone through every forum post related to this issue with no success and have run out of ideas what to try and look at and would appreciate some help.

OS Version: Home Assistant OS 12.1
Home Assistant Core: 2024.3.1

manic bane
cobalt moss
#

Yep, I've ensured the port is correct.
The observer url works fine and says everything is okay.
I've also used nmap to ensure that my ports are open and reachable.

#

Going to that address, it redirects me to http://{ip}:8123/onboarding.html just fine, but it's coming up blank and content on this page isn't loading at all.

humble mirage
#

You're using Chrome?

#

No content blockers?

manic bane
#

Try using a private browser mode

cobalt moss
#

Tried, different pcs, different browers, incognito mode already.

#

Chrome, Edge, Firefox, Safari. These are the ones I've tried so far.

plucky talon
#

I had this during my setup, only solution I found was legit just smashing control-R and it worked

delicate valve
raven forge
cobalt moss
eager trail
brave musk
grand pivot
#

No

humble mirage
#

No, because HACS is a custom integration

#

Just follow the HACS docs to install it

brave musk
#

It would appear it gets installed in the actual run time, so if I pull a new image, it would get nuked?!

grand pivot
#

No

#

Your config isn't in the container

#

Well, shouldn't be

#

Containers are ephemeral

humble mirage
hazy mica
#

Running HAOS on a MacPro, last update lost access to Home Assistant, no port open at 8123. addons still working. Only have ssh access. supervisor and core apparently check out ok. Im stumped to fix. any suggestions please. tia.

jade berry
#

sounds like firewall is blocking it

#

google how to change firewall options on macos and allow the port

#

i assume you can use it on mac itself by specifying localhost:8123

hazy mica
#

System was running fine. Was playiing with drag and drop sections and had a terminal window open. Doorbel rang and came back to chaos on the screen from the 7 month old cat apperently trampling the keyboard, so I have no idea what ws done

#

Not running macos, running HAOS on the mac hardware

jade berry
#

oh, then that simplifies things

hazy mica
#

been running fine for years until day before

jade berry
#

do netstat -tlpn and check if you see 8123 port open there

#

if so, do ufw allow tpc/8123 (or something like that), there is some status command or something as well

hazy mica
#

tcp 129 0 0.0.0.0:8123 0.0.0.0:* LISTEN -

#

thats in the results

jade berry
#

yeah, thats good. now try allow thing and you should be able to access it from outside

hazy mica
#

however, when I scan for the port in network radar, it does not show.

#

I also have this in the supervisor log:
24-03-20 12:55:52 INFO (MainThread) [supervisor.auth] Home Assistant not running, checking cache

jade berry
#

iv enever tried haos, not sure whats the default config. ping might be disabled

hazy mica
jade berry
#

from ssh curl localhost:8123 should result in something something
and if you do it from some other machine curl <IP address of HA instance>:8123 it probably will result in error that it is not reachable

hazy mica
#

it spits this out with the -v flag and seems to get stuck
โžœ ~ curl localhost:8123 -v

  • Host localhost:8123 was resolved.
  • IPv6: ::1
  • IPv4: 127.0.0.1
  • Trying [::1]:8123...
  • Connected to localhost (::1) port 8123

GET / HTTP/1.1
Host: localhost:8123
User-Agent: curl/8.5.0
Accept: /

reef relic
#

Good morning to everyone. Thanks for allowing me to join. I received my Home Assistant Yellow the other day day and I was successfully able to put it together and install I believe the software needed to do the first boot. Unfortunately, I used MacOS w/ Silicon chip and can not get the Home Assistant App do download so I can preform the boot up and go live. Iโ€™ve spent all evening and morning reading trying to figure it out. Iโ€™ll be honest a bit overwhelmed with the Terminal and GitHub coding as Iโ€™ve never been taught any of it. Please help me if you can. I would hate to think Iโ€™ve spent all this money just to get to this point and that be that. Thank you in advance.

tender wren
tender wren
#

<@&330946878646517761> Spam alert - user Lanten101

waxen glade
#

looking for some help, have it installed on a pc x86, its installed on nvme drive, but i have a SSD drive in there as well that i wanna use for storage, trying to figure out how to mount the drive,
[16:04]
if i got to settings and system, under all hardware it shows the drive there, but under storage it doesnt show

#

ive tryed ssh into it, and went under etc/fstab/ and i tryed manully mounting the drive there, then reload homeassistnat and it clears the mount point that i put in there

raven forge
#

That's not really supported by HAOS other than using it as data disk. You can use udev rules though. You likely did you edit in a container.

waxen glade
#

im not using a container

#

using generic x86_64 install

#

or is there still a udev folder somewhwerE?

raven forge
#

That's not what I meant. You likely ran your commands via the SSH addon, right?

waxen glade
#

yes

raven forge
#

Addons are docker containers.

waxen glade
#

oooh ok

#

learn something everyday

raven forge
waxen glade
#

ok, thank you

rare valley
# rare valley Hi, I'm having issues with the HA supervisor all of a sudden. I can't start addo...

So update on this issue. I bought a new SD card and attempted to restore my backup on the fresh SD card. However, when trying to restore it, it completely breaks HA. I checked the logs (via monitor and keyboard) and there are a ton of errors about being unable to set up dependencies, including http, network, frontend, etc. I tried restarting HA multiple times, both by unplugging the pi and with various CLI commands but the same thing happens every time. Any idea what's up this time?

raven forge
#

Share what you see.

rare valley
#

(output of ha core logs)

#

Ok after being able to see the full logs I can see this line which I think might be the cause:

2024-03-20 15:24:16.719 ERROR (MainThread) [homeassistant.config] Invalid config for 'http' at configuration.yaml, line 19: not a file for dictionary value 'http->ssl_certificate', got '/ssl/fullchain.pem', please check the docs at https://www.home-assistant.io/integrations/http
Invalid config for 'http' at configuration.yaml, line 20: not a file for dictionary value 'http->ssl_key', got '/ssl/privkey.pem', please check the docs at https://www.home-assistant.io/integrations/http

Is there some way I can edit the configuration over SSH (22222) since I can't access the FE or anything?

grand pivot
#

seems like you don't have the cert files there

rare valley
#

Yeah I think the backup I have doesn't have the cert file. I can fix that part later, I just need to fix the config so I can get it running again.

#

Ok I think I was able to edit the file in the docer container, lemme try restarting

grand pivot
#

the config is failing because the files aren't there

rare valley
#

yea i got that part, I just commented out the lines for SSL which hopefully will do it

#

Alright frontend seems to be running again. Need to change a few things but I think I am back up and running.

rare valley
#

Got everything working again finally. And yes I made sure to actually create a full backup and save it offsite this time. Not goin through that again lol.

delicate valve
raven forge
#

Still not sorted by memory. I already answered earlier so I'm just being a stickler ๐Ÿ™‚

#

Oh, btw, if you press / you get more information about the processes.

delicate valve
raven forge
#

Nothing sticks out to me at first glance.

delicate valve
#

thanks...

#

probably better to do it when i get errors on the VM huh

placid sage
#

Anyone know why I'm getting the error: "Unable to fetch auth providers. https://hass.example.com/?auth_callback=1" after setting up home assistant in an nginx reverse proxy? All of my other websites work fine, it's just home assistant. I'm sure it's something simple that I'm missing as usual

placid sage
#

nevermind, I figured it out

raven forge
#

Good manners are to share how.

white stone
#

hey, im using HA docker in host mode, and i want to use pihole too, is there a way to run both, cause both use port 5353

brave musk
white stone
#

so i gotta do pihole

eager trail
#

@raven forge - Still trying to figure out how to use the Bluetooth you helped get recognized on my new HA build. This morning I happened to check the console screen in Proxmox for the HA VM and it was filled with these type errors. I went to web page for HA and rebooted it - and this is what I am seeing within 5 mins. https://imgur.com/cWlgAa6

barren glade
#

Hello my home assistant is completely blocked after starting, I am connected in cli after a few seconds it is also completely blocked.

robust cairn
#

on which hardware you start? raspberry pi? it can take up to 20 minutes to get a login

barren glade
#

Nop its ont FREEBOX DELTA :/

raven forge
cloud depot
#

I am having an issue updating from 2024.3.0 to 2024.3.1, usually in the past it was related to disk space but I have freed up a bunch (for me) 1.2gb free and also rebooted. The core and matter server will not update. Here is the error message details: https://pastebin.com/Q9Wyixxc

raven forge
#

1.2gb free
That is nothing.
Please check the supervisor logs and also share some context surrounding what you shared.

humble scaffoldBOT
cloud depot
#

File "/usr/src/homeassistant/homeassistant/helpers/service.py", line 905, in entity_service_call
single_response = await _handle_entity_call(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/src/homeassistant/homeassistant/helpers/service.py", line 975, in _handle_entity_call
result = await task
^^^^^^^^^^
File "/usr/src/homeassistant/homeassistant/components/update/init.py", line 161, in async_install
await entity.async_install_with_progress(version, backup)
File "/usr/src/homeassistant/homeassistant/components/update/init.py", line 465, in async_install_with_progress
await self.async_install(version, backup)
File "/usr/src/homeassistant/homeassistant/components/hassio/update.py", line 305, in async_install
raise HomeAssistantError(
homeassistant.exceptions.HomeAssistantError: Error updating Home Assistant Core

raven forge
#

Supervisor logs please.

humble scaffoldBOT
#

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

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.

modern basalt
#

Not for HA, but for some other containers that I want to run: What's a good, easy to administer (for a linux newbie with little time to learn), light weight OS that I can run as a pure docker/dockge host? RancherOS worth using, or just stick with Debian?

hexed ridge
#

Hi there Id rather avoid gettijng another pi or messing with my pi I have a server with VMS, any idea how Id use a hyper v VM to install home assistant?

#

I have a linux install with ubuntu but idk if that will work

raven forge
raven forge
modern basalt
vast river
#

Hi Community,
I need help please. This is my first try to install the Green. However, itโ€™s not working.

  1. I cannot access the home assistant welcome screen.
  2. On trying to access the home assistant device, I get an error message โ€œtoo long to respondโ€.
  3. Iโ€™ve tried using the various addresses to access the welcome screen but I get the same message.
    I have also confirmed the IP address that Iโ€™m using Wifiman app. Itโ€™s the same IP address detected by my Home Assistant iPhone app.
  4. When trying to access it using Home Assistant for iOS, I get an error message โ€œThe request timed out.โ€
  5. I donโ€™t see a green light at the back of the device where the ethernet cable is connected. I only see yellow.
  6. I can see the front yellow led light heartbeat pattern.
  7. Iโ€™ve tried doing a reset of the Home Assistant Green OS using a microSD card following the installation instructions in the website. The procedure went as expected, e.g., fast blinking then led lights turned off. I still get the same error message and problem.
  8. Iโ€™ve checked that my ethernet cable has internet by connecting it to my laptop.
  9. Iโ€™ve checked my ISP modem and the LAN ports are live.

The device is supposed to be plug and play so this is very frustrating.
Could you please help me resolve this? Or is my device broken and will need to be replaced with a new one?

tender wren
#

Connect a keyboard and a HDMI display to the Green and look for error messages

hexed ridge
empty bloom
#

specifically hyperv? no

eager trail
empty bloom
#

Hyper v is terrible at non normal usb devices

raven forge
fickle oak
#

Hello there!

I have an old laptop that I use as a media server with Ubuntu Server on it. Any chance on installing Home Assistant with Add-ons?

grand pivot
#

in a VM

fickle oak
#

Ok. THX

fickle oak
#

I checked if I can run VM and return 0, that means that I cannot run VM. Any other soluศ›iona?

hexed ridge
#

I also like the docker Wyze bridge support but idk how to get that to work with hyper v

grand pivot
#

there's nothing magical about addons

fickle oak
#

Didnt know you can install stuff whitout addons. I need the salus integration from hacs

grand pivot
#

integrations aren't addons

#

and HACS has no addons

#

you an use HACS and any integrations regardless of how you install HA

#

it says it in the docs

fickle oak
#

Ok. I need to read more about it

#

Thx

proven mauve
#

are there any major downsides or benefits to running HA on either a VM or docker container in unraid vs a dedicated box?

raven forge
hexed ridge
#

do we need to port forward for docker wyze bridge

storm raven
#

Core update 2024.3.3 seems to have a problem, just updated and it brakes my Home Assistant OS VM...

#

Tried it twice and it's stuck on loading groups...

raven forge
#

Breaks in what way? Did you check the supervisor logs?

storm raven
#

Itโ€™s just stuck on loading groups. No hadnโ€™t the time to check what went wrong and just restored from a snapshot.

proven mauve
lilac dirge
#

Hello I installed proxmox

#

I have a single SSD

#

In a mini pc and idk what I should take

#

Xfs or ext4

raven forge
#

I'd stick to the default EXT4 unless you have a good reason not to.

lilac dirge
#

Ok thank you

sleek escarp
#

I'm moving from a pi 4 to a proxmox server. I wouldn't wanna waste resources on a full on home assistant OS VM
Would a supervised install in a debian 12 LXC be recommended?

raven forge
#

Very much not:

  • #general-archived message
  • #general-archived message
    Yes supervised is documented but it's not recommended and if everyone gives you heat if you ask about it is it really supported?. I'm not sure if anyone's tried to run it in a CT but not all software works that well in them.
    Proxmox staff also recommends against docker in LXC so this is just a mix of bad ideas. If you want to save resources I recommend the pure docker installation method in a debian VM. You can re-create addons on your own with other docker images.
sleek escarp
#

Docker in LXC works well with nesting enabled
I think i could try it as i have some solid experience with linux, but if it anytime i have an issue i'll try either a VM or docker method you suggested

#

Honestly the only add on i care about is esphome

grand pivot
#

No need to use an addon for that

raven forge
sleek escarp
raven forge
#

Now personally I'd be too curios to try it anyway. Just be prepared for having to scrap it.

sleek escarp
#

It will surely work, but i don't wanna put a lot of time every week into maintaining it

raven forge
sleek escarp
#

My 4 core first gen ryzen can probably take one more 1 core HAOS VM

raven forge
#

I wouldn't worry much about CPU. It's pretty normal to over allocate it and HA(OS) doesn't use that much during normal use. As long as you have 2G of memory or so to spare it's fine.

sleek escarp
#

Wait I can over allocate cores on proxmox?

#

I already have a 2 core windows vm, keeping one for proxmox and CTs, and the HAOS vm will take the remaining one
Could i give the HAOS vm two cores and run it off prayers?

raven forge
#

Of course.

It is perfectly safe if the overall number of cores of all your VMs is greater than the number of cores on the server
https://pve.proxmox.com/pve-docs/pve-admin-guide.html#qm_cpu
At the end of the day a VM is just like a process. Your PC's processes have access to all cores too and the OS manages that, no?
You only have to be careful with memory. KSM can stretch it a bit but it's generally adviced not to over allocate it or go over 80% host usage.

ocean sandal
#

I am lost, I have set up home assistant, and all great, but now I wamt to add the dashboards to a monitor/tv. I want to cycle a couple of dashboards
Is this possible or do I have to stick with the view I get when i log in with the ipaddress
I searched, maybe I got the terminology wrong, but not finding what i look to do
thanks all

raven forge
#

You want to plug in a monitor into the same machine running HAOS and get a browser windows showing HA?

sleek escarp
slim geode
#

I just installed HA Supervised and I'm not able to connect to the main HA page in my web browser. I can ssh into the machine, I can connect to the Observer page (port 4357), but when I try to connect to the main HA page (port 8123), it times out.

I've tried restarting HA, I disabled an Apache server I had running, I manually changed the port in the config file. None of these worked.

In the iOS app, my server is found, but it times out there, too

#

From the observer page:

Home Assistant observer
Supervisor: Connected
Supported: Supported
Healthy: Healthy

raven forge
#

Which OS?

slim geode
#

Debian GNU/Linux 12 (bookworm)

raven forge
#

I'd check the supervisor and core/docker logs.

slim geode
raven forge
#

Can you run the commands from the link too?
Is ufw installed or iptables configured?

slim geode
#

I'm not sure if that link is working. It brings me to a different message each time I click it

#

ufw is installed

#

And iptables is there, too, but I haven't done any sort of configuration on it

raven forge
#

Check ufw status.
What does nc -vz localhost 8123 and ss -4ltpn | grep 8123 say?

#

UFW does not come with debian by default. I'd disable it.

slim geode
#

ufw status:
Status: active

To Action From

OpenSSH ALLOW Anywhere
WWW Full ALLOW Anywhere
43/tcp ALLOW Anywhere
53/tcp ALLOW Anywhere
53/udp ALLOW Anywhere
43/udp ALLOW Anywhere
OpenSSH (v6) ALLOW Anywhere (v6)
WWW Full (v6) ALLOW Anywhere (v6)
43/tcp (v6) ALLOW Anywhere (v6)
53/tcp (v6) ALLOW Anywhere (v6)
53/udp (v6) ALLOW Anywhere (v6)
43/udp (v6) ALLOW Anywhere (v6)

raven forge
#

Yeah so likely firewall issues.

slim geode
#

nc:
Connection to localhost (::1) 8123 port [tcp/*] succeeded!

ss:
LISTEN 0 128 0.0.0.0:8123 0.0.0.0:*

raven forge
#

Please use code blocks. Run ufw disable and see if you can reach HA then.

humble scaffoldBOT
#

To format your text as code, enter three backticks on the first line, press 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.

slim geode
#

Disabling ufw fixed the issue

#

Thank you for your help

raven forge
#

Yay.

slim geode
#

Will disabling ufw likely to cause me any other issues?

raven forge
#

This isn't permanent. I think. Either uninstall UFW, disable the service or configure it to allow access.

#

I don't think so. I don't use it.

tidal hatch
#

good day/night, I would like to ask you about how can I fix some pulseaudio issues i'm having with my setup (orangepi 4 LTS + debian 12), right now I don't have alsa or pulse installed in baremetal, just the hassio-audio image. I have installed the voice assistant combo (wyoming using assistant microphone+openwakeword+piper+whisper). other addons like spotify connect works perfectly, but when I try to use the voice assistant, pulseaudio seems suspended and doesn't wake up W: [alsa-sink-ff880000.i2s-ES8316 HiFi ES8316 HiFi-0] alsa-sink.c: Resume failed, couldn't restore original sample settings.

cyan granite
#

Has anybody noticed any issues with docker-compose eorroring out on the latest update?

#

It has been working fine for AGES and now something has gone haywire when I do the docker compose

cinder marsh
#

It may have better hardware support

raven forge
raven forge
humble scaffoldBOT
merry wigeon
#

Any idea what I did wrong to have both my Zooz Zwave and Sonoff Zigbee dongles show up as the same device in Proxmox? Plugged into different usb ports, obviously. They show up with the correct manufacturer, product and speed under my HAOS VM, but show the same device address.

Edit: If anyone sees this, it works anyways I guess. Strange.

raven forge
tidal hatch
grim parcel
#

Still having tje issue at a mates install HA won't work on some devices like 90% of the time but other devices works and external access via nginx works OK too but internal network via ip isn't working on some devices just says can't be reached

keen bloom
#

Error installing Home Assistant
An error occured while installing Home Assistant, check the logs below for more information.

#

how to solve this?

tender wren
#

By checking the logs

humble scaffoldBOT
#

This forum post covers the things we need to know to help you. Key things are the version of Home Assistant you're running, how you installed it (Home Assistant OS, Supervised, Container, Core), the exact component you're having problems with (eg Nest climate rather than climate), and details of the problem (ideally with any error message in full).

grand pivot
zinc raptor
#

is this the best place to ask about errors seen while upgrading, as well as installing?

I have HAOS's supervisor detecting crashes and rolling back when trying to take HA from 2024.2.5->2024.3.3. HAOS on a Pi5 on SDcard.

I saw some forum posts that mentioned empty YAML files as a possible issue; I had some of those and removed them. -rollback.log is full of python errors being unable to reach the internet, from lots of different integrations, but I don't know what to make of it -- HA seems to be running fine, everything else on my network is fine, no obvious reason for the errors I can see. What should I look at?

balmy solstice
#

Hi guys, just had my esxi server shut down unexpectedly, tried to do a soft power off (to recable) but the server froze and I ended up having to unplug it, after booting it back up all my vms boot fine apart from home assistant. VMWare complains saying "Object type requires hosted I/O", and says the vmdk doesnt exist when it does?

#

Tried a disk scan in ssh, doesnt look great

#

Using vmkfstools -x check haos_ova-12.1.vmdk returns "Disk needs repair"

#

Is there anything I can do other than a full re-setup? I got quite comfortable with everything set up I really dont want to spend half a day setting stuff up again

#

So im not sure it will actually do anything

#

Update: I did vmkfstools -x repair and it fixed it ignore me

#

:P

subtle sun
#

If I was to upgrade from a Pi 4 to Pi 5 what would be the easiest way to do it?

Install HAOS for Pi 5 on a new SD Card then reload a backup from the Pi 4?

tender wren
#

Yes

restive phoenix
#

Hi, I'm migrating from home assistant core on docker (unraid) to home assistant OS. I can't get into HAcore UI to make a backup file...is there another way to move all my files from Core in the config directory over to my standalone haos box? I have access to the ha console on it

grand pivot
#

Copy the whole config directory, including all hidden and . files

#

Use tar or zip or whatever to create an archive, copy it over to HAOS, stop HA core, extract the files into the new config directory

restive phoenix
brazen gull
#

Hey all! I see there is now support for RPI5, currently running HA on an older RP4 and want to migrate, is there any documentation to follow etc or is it a case of a new build?

tender wren
#

It is basically the same install procedure as a Pi4, just selecting the Pi5 specific image in the Raspberry Pi Imager. Make a backup on the old device, download it and restore it on the Pi5

outer wadi
#

Is there anyway to debug why an odroid n2+ doesnt boot. Tried flashing 12.0 and 12.1 with belenaEtcher both via an emmc usb adapter, or by directly attaching to the device. It flashes and verifies but then it doesnt boot, gives red power light but no blue sys light.

I had this issue about a week ago, but 12.0 worked so i thought it was a 12.1 issue. After a few days I tried to move the device and upon plugging it back in, it wouldnt boot again.

If i flash it with ubuntu it boots straight away, any ideas?

empty bloom
#

Not sure how you can debug something that doesn't boot

outer wadi
#

Me either , was hoping there might be a log file or a way to view an error that the N2+ might be getting

#

Oo I just updated petitboot and now I have some screen output at least

outer wadi
outer wadi
#

So i just flashed ubuntu again and it went straight in, i dont see how this can be anything but an issue with the images now. I dont know enough about linux or boot partitions to know whats up but if I can boot into another OS using the same emmc and device, i dont see how it can be a device fault any more, almost has to be the HA image, or compatibility of that image with the device

empty bloom
#

write up an issue against HAOS with your findings

outer wadi
#

Yeah ill update my existing ticket

#

No idea what to do now though ๐Ÿ˜‚ guess i try older versions to see if they work.

outer wadi
#

Yeah same with 11.5 i have no idea what to do now, its like something is different on my device to the other ones or something. I cant see how this is an issue for everyone

outer wadi
shy pond
#

I'm facing an issue relating to file and directory permissions when running HA via docker. I'm trying to change the permissions on the HA config directory, but each time I start the container, the permissions on certain files change to root. I have PUID and PGID set to 1000 within my compose file, and this is the user to which I am chmod'ing the HA directory to (recursively) . What am I missing?

humble mirage
#

The HA image runs as root, and only as root

#

Nowhere in the HA docs does it say it supports changing that

#

There's third party Docker images that do support that, but you'll get limited support for those images since that potentially breaks many things

shy pond
#

Ah, well that explains it then. I was under the impression that it could run as a non-priv user.

#

Thanks for confirming that it's not possible for me.

ocean sandal
humble mirage
#

Plug the monitor/TV into a computer that runs a web browser

shy pond
shy pond
ocean sandal
shy pond
shy pond
tidal hatch
outer wadi
#

Well I am giving up. I have been trying for hours to get HAOS running on my odroid N2+ and I can only assume theres an incompatibility between either my N2+ Revision or my mmc card revision and HAOS. Tried 12.1, 12.0, 10.5 and 9.5. None of these work for me. Flashing and booting ubuntu works fine. I have raised an issue for it, I can see two other people also raised issues similar to this. I guess i am just unlucky, for now going to have to live in my smart house that's very dumb owing to having no brains behind it ๐Ÿ˜ญ

Really nothing more I can do but hope someone with more knowledge that me can work it out.

plush folio
#

Hi, I have HA on a promox VM. I tested on both sata ssd and on a nvme and I have the same result. I also run memtest86 (https://imgur.com/a/bOfTXxW ) but rams seem ok. This are my journal logs https://paste.debian.net/1311977/. I don't really see any reason that says "crashed because of this". It crashes often, every 1/2 hours and then it restarts itself but after 5/6 tries it just freeze and It doesn't work until I completely restart the VM. When it crashes I have still access to ssh and sftp. Thanks

#

I'd like to understand how can I see why it crashes

raven forge
outer wadi
placid sage
#

Hi. I'm installing HA supervised (because I need it on x86_64 and need addons which aren't available in the container). I have onboarded successfully and am seeing the HA home page. However, it says my installation is "Not privileged". I ran everything as root on the debian host. The help page isn't very helpful, it just tells me to update my OS (it is up to date). Any hints?

fallen lichen
#

What's all been done on the host machine apart from the supervised installer

#

and what version of Debian

placid sage
#

debian@debian:~$ cat /etc/debian_version
12.5

fallen lichen
placid sage
#

Yeah sorry, that's the parent link

raven forge
#

Can you share the complete error?

placid sage
#

I clicked "Ignore" and it went away. It's not under System > Logs > Show full log

#

Maybe it fixed itself when it restarted, not sure

cyan granite
#

Has anybody noticed any issues with docker-compose erroring out on the latest update?

#

I was trying to do an update on HA and it failed on the docker compose

grand pivot
#

nope. please be more specific about how it failed

cyan granite
#

It's long, do I make a thread about it?

grand pivot
#

just describe the problem

#

if you must:

humble scaffoldBOT
#

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

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.

cyan granite
raven forge
#

Please share the compose file and also show the command you used, not only its output.

grand pivot
#

That doesn't really look like the output of a recent version of docker compose

raven forge
#

The .py gives it away. Docker compose v2 is based on go.

cyan granite
#

I used: sudo docker-compose up -d

raven forge
cyan granite
#

just a moment its comin

raven forge
#

I tried to reproduce this with 1.29.2 but can't.
I'd recommend you upgrade and try again. It's possible you already have V2. Try sudo docker compose up.

cyan granite
#

<breaks out in cold sweat>

raven forge
#

How did you install docker?

cyan granite
#

I did a manual install ages ago

raven forge
cyan granite
#

trying to remember

#

it looks like the docker compose worked

#

do I need to use the -d feature

#

if I do it at the command line it just keeps running

#

I think that is a yes. the -d sets it to background mode

grand pivot
#

If you want your containers to keep running in the background, yes

cyan granite
#

Thank you for the help folks!

#

that seemed to work.

raven forge
paper pulsar
#

Hello, I have installed the Virtual Box and i installed the windows installation for Linux but it is denying me permission when i boot up the machine. I am trying to eventually download the ESP home intergration on the HA interface after. Any help?

versed vapor
#

Without you telling us what you did try and what about it isn't working, this will be difficult

paper pulsar
versed vapor
#

Esphome is an addon + integration combo.you will need both

#

With not connecting in the first place do you mean you to ha?

paper pulsar
cosmic lynx
#

hello i am planning to install HAOS on an Intel x64 tablet that would be wall mounted after 3 years on the RPI3. For this id probably need to make some firmware modifications (/usr/lib/firmware) to make the touchscreen work and if something else.

Is if its ok to perform this system-level modifications without messing with HAOS upgrade cycle? The idea is mostly to make the touchscreen work and for more interactions access using SSH

humble mirage
#

HAOS has no local UI

#

So... that's not gonna work

tender wren
#

And most system parts are read-only

cosmic lynx
#

ah makes sense, coming to my mind manually adding local UI but i assume thats gonna mess up things

humble mirage
#

You can't add a local UI to HAOS

cosmic lynx
#

in this case i assume best would be to keep HAOS on RPI3, a linux distro on the tablet while accessing through browser to the HAOS

humble mirage
#

Well, anything with a web browser for the tablet

cosmic lynx
#

hmmm comes to my mind a HAOS container in the tablet while its host using some distro but i dont think its gonna be as bullet proof as in the RPI3

#

Ill be messing around and check, thanks!

raven forge
#

You'll be having a great time trying to containerize HAOS.

versed vapor
#

Docker in Docker hehe and twice the supervisor pain

cosmic lynx
#

ops my bad meant Home Assistant Container, 3 years since my last install need to do homeworks again ๐Ÿ˜‡

steel blade
#

I need some advice on upgrading from RPi4 + USB-connected SSD for data store to RPi5. The full backup is huge, I would prefer to move config to new hardware and plug in the SSD to move the data

foggy jackal
#

Hi. I'm using HA in a virtual machine on my Synology 1019+. It was running fine for years, but since the update to the march release of HA, the system freezes every 2.5 days. All I can do is restarting the VM. After that, it is running fine for the next couple of days.
I can't find anything causing the freeze. Is someone here, who could help me finding the cause and solve the problem?

haughty elbow
#

Hi everyone, newbie here. I'm trying to install HA on a Pi4. I cannot access homeassistant.local.

old sorrel
#

I run ZHA on an HUBZ Zigbee stick combined with ZWave. Last night I got an odd message that something had changed my ZHA config and did I want to restore. I said OK and as of tonight most of the 65 devices I have are offline. I have a SkyConnect t hat I've wanted to migrate to for a year so now sounds like the time to do it. Q: is it worth doing the radio migration before reconnecting the other devices? Or should I just start fresh?

daring moon
#

I'm not sure this is quite an installation problem, but I want to securely access HASS externally and am having some trouble getting it set up

I have the following:

  • standalone running on my NAS, which I can access internally, and is all working
  • domain name, correctly connected to HASS via cloudflared tunnel, can access via URL

currently anyone could access the domain, I'd rather lock it down a bit tighter. Initially I had it working with zero trust and google auth, only allowing whitelisted email addresses, this works fine in the browser, but breaks the comapnion app.
I recently just tried to set up a WAF rule with client certificates, but I can't get the browser to request a certified and the companion app just fails to connect with a 403 error.

I'd love to know how people are securing HASS externally, and if anyone has the companion app working with any of the above (maybe I made a mistake and just need to tweak some config?), ideally I'd love to utilize the cloudflared tunnel and some part of cloudflares firewall if possible, and I really want to "roll my own" so don't want to just pay for the cloud services.

drifting osprey
#

I just recently secured myself successfully via cloudflared and WAF rules. Happy to help but I need to be in front of my desktop and not on my phone watching TV!!!

raven forge
#

I use tailscale to access my servers/services from outside of their local network.

drifting osprey
#

I added wireguard for laptop usage in addition to my cloudflare tunnel access which is secured via mTLS cert on my phone and WAF rules in cloudflare.

daring moon
#

I have the WAF rules set up and the certs but the phone won't ask to use them (now will the desktop, I never get a popup or anything). Really just not sure to get the cert to be used in the first place

daring moon
#

got the thing working...I think there's some significant delay in flushing out changes, probably caching in the DNS and browsers

cinder marsh
lean depot
#

okay now i try it here. i updated my Homeassistant today, and after the update i cant access the webui.. - now i get this error: "ERROR (MainThread) [homeassistant.components.http] Failed to create HTTP server at port 8123: [Errno 98] error while attempting to bind on address ('0.0.0.0', 8123): address already in use"

#

And its not already in use, idk what to do

#

i can access the CLI and automations are also working

#

anyone got an idea for me? dont want to install everything agian

torpid beacon
#

hello.

I installed home assistant qcow2 into a KVM following instructions on their site.

I'm having what's maybe a networking issue I guess I would call it.

The host system has access to the homeassistant.local:8123 but I can't ping any devices on my network from the HA cli

any suggestions? it's using a USB dongle and I have the correct bus ID.device ID from lsusb

i set the ip/mask in the HA Setting -> network to the static IP i want, and it shows in ip a from the cli

any suggestions to try something next?

#

nmcli connection shows two names Supervisor (device)

but i can't see more about those

#

let me know if there's a different place here to ask about this, because it's maybe not an installation question.

earnest karma
latent hinge
#

Hi, I am sure this is not a new question but after installing homeassistant in my old Raspberry Pi, when I try to connect to it I get the error, "the site cant be reached". I tried using the device IP address, http://homeassistant:8123/ and http://homeassistant.local:8123/. When I plug a screen to my Raspberry Pi I can see everything is loaded ok and the last shown output is "openhabian login":
Does anyone know how to proceed from here?

humble mirage
#

openhabian?

#

homeassistant.local isn't working because you didn't use HAOS and somehow (Container? Core?) added HA to an existing OpenHAB install.

#

Use http://192.168.0.42:8123/ instead, with the IP of the OpenHABian host

latent hinge
#

The IP of the device is 192.168.1.188 and I tried to use IP:8123

raven forge
#

How exactly did you install it? We can't help that well with non-official or not as common installation methods.

latent hinge
#

I flashed the image using Raspberry Pi imager and then slotted the flash card into the device and switched it on

#

It went through different steps including some updates and then it stopped at the login line

#

I assumed following the instructions from raspberry pi website was official, but maybe I am wrong

raven forge
latent hinge
raven forge
#

You mentioned openhabian though which doesn't seem to come with HA so we need to know how you installed it.

latent hinge
#

yes I think I understand the problem. I used openHAB image assuming that was the correct one since there was no other Home Assistant image available when using raspberry pi imager

#

I assume there is no image available for an old rasbperry pi ๐Ÿ™‚

raven forge
#

Maybe I missed it but which pi version do you have?

latent hinge
#

I believe it is the first one based on the layout of the device

raven forge
latent hinge
#

ok, I will take a look at it, thanks for your help. I will also try to flash the image for raspberry pi 3, in case I have that device

raven forge
latent hinge
#

you are right I didn't notice it, it is version 1

#

thanks

raven forge
#

There's also the choice to pick different hardware ๐Ÿ™‚

fallow aurora
#

hello I am trying to install homeassistant using generic image on tiny PC but the result is:

#

Not able to add screen but jumping to emergency console

#

and /dev/root is full....

#

why? .... there wasn't any interaction in process of installation with user

fallen lichen
#

The installation is the flashing of the image. After that, it's just onboarding/configuration.

fallow aurora
#

I have image on flash usb, in tiny PC I have SSD, sdbx are visible

#

booting from usb, processes are standing up, making grow file system, after that jumping to emergency

fallen lichen
fallow aurora
#

am I doing something wrongly

#

yes I was doing that

fallen lichen
#

Sounds like you did not do that. You explained you installed HA to a USB drive but your PC has an SSD

fallow aurora
#

so what I don't understand? I should flash image using flashing tool directly to ssd?

fallen lichen
#

Correct

eager shale
#

i just installed duckdns and set up port forwarding on my router. after rebooting HA i tried connecting via duckdns url with https and it works fine. however, when i try using the local address it loads the login page then when i log in it shows the home assitant logo and says "unable to connect to HA. retrying in 60, 59, 58..."

#

Anybody know how to fix this?

empty bloom
#

you have to have a proxy to access HA locally

#

it's either local or remote if you don't use a proxy

#

don't enable ip_ban if you mess up your password frequently

eager shale
#

seems im not so savy with this. i installed nginx but log output is just saying a bunch things stopped. also im not sure what you mean by adding trusted proxies in the configuration.yaml file. im not sure how to do it. additionally im not sure what you mean about the piort forwarding. on my router i already setup a rule forwarding 443 to 8123, thats how the external duckdns is working.

#

is there a more deatailed guide to help me with this?

#

not sure if this helps, but when i try to log in from the local ip address i get a notification in HA saying login attemt failed, invalid authentication. seemigly indicating i am able to connect to HA localy, there is just some kind of authentication error

#

also under network settings its showing me:
"Invalid local network URL
You have configured an HTTPS certificate in Home Assistant. This means that your internal URL needs to be set to a domain covered by the certficate."

#

to clarify my issue - when accessing https://<domain>.duckdns.org i am able to login, regardless where im doing it from. the problem is when trying to log into https://<ip_address>:8123 it loads up the login page but then when i click login it just says unable to connect to home assistant

empty bloom
#

i.e. you will edit your configuration.yaml file for home assistant, you will use yaml to define what I described. You can use the post I linked as a guide

#

when you add the proxy, you'll need to forward 443 to 443

#

and remove your previous 443 to 8123

#

if you want a guide, I'm sure you can find one on google. But Everything I've siad is all you need to do and that post covers all the yaml

empty bloom
#

to get around that, use a proxy so you can use both internal and external

buoyant beacon
#

Hello people.
I installed HA on QNAP NAS.
I struggle with finding a way to install Zigbee2MQTT add-on.
When I go to Settings, there is no Add-ons section.
I installed HACS and enabled it according to the howto. Still no luck.

#

Can you advise?

eager shale
empty bloom
#

did you configure the proxy correctly and is it running?

#

I'd wager you didn't expose 443 on the nginx addon

eager shale
#

im not sure

eager shale
empty bloom
#

in the configuration page, there's a port section, expand it and type 443 in the 443 field

eager shale
#

oh, yeah its already populated 443

empty bloom
#

are you sure, or is it faded

#

you literally have to type it out

#

if you didn't type it out, it's not being used

eager shale
#

it says it there

empty bloom
#

screenshot on imgur pls

eager shale
#

btw i changed my router back to 8123 to be able to log back in

empty bloom
#

that's fine

eager shale
empty bloom
#

Yep, and you filled out the domain?

#

when you start nignx you should be able to connect to HA with http://homeassistant.local:8123

#

or http://<ip>:8123

eager shale
#

yes, duckdns domain name is there

eager shale
#

cannot connect to page

#

only https

#

then it says cannot connect to HA

empty bloom
#

post your configuration.yaml file

eager shale
empty bloom
#

and you restarted HA?

eager shale
#

yeah

empty bloom
#

Then it should work locally with or without the proxy running

#

are you sure your port forwarding rule is 443 to 443?

eager shale
empty bloom
#

and when you turn off nginx proxy and duckdns does it work?

#

locally

#

if it does, that's your url to use

#

then, turn on nginx proxy manager

#

accessing it locally should still work

#

if it does

#

then turn on duck dns

#

then external should work

eager shale
#

trying now

#

should i also remove ssl info from conguraiton file?

empty bloom
#

are you actually putting the ssl crap in the ssl folder?

#

if yes, then you probably want to keep it

#

it needs to match what's in duckdns's ssl

eager shale
empty bloom
#

that's a link to your port forwarding

eager shale
empty bloom
#

yes, but tha'ts for HA

eager shale
#

i turned off duckdns addon and no change still able to access externally and not internally

empty bloom
#

those values need to match what you put in duckdns addon

empty bloom
eager shale
#

lol i know. i just dont know what..

empty bloom
#

if the ddns addon is off you won't be able to access externally

#

you're probably getting fooled by cache

#

use an incogneto window

eager shale
empty bloom
#

it doesn't help

eager shale
#

incognito is giving same results

empty bloom
#

if you turn off duckdns and nginx, but keep the 443 to 443 port forwarding rule, you 100% should be able to access HA locally.

#

http://<ip>:8123

#

you aren't adding an S to http are you?

#

http locally, https via duckdns

eager shale
#

correct. locally when i go to http://<ip> it says cannot load page at all. when i go to https://<ip> then i am able to login in but then it says unable to connect to HA

empty bloom
#

i think you need to brushup on urls, internal and external

#

you never set up SSL internally

#

so https://<ip> should not work

#

it will not work

#

so you're ither filling out the information wrong on the duckdns addon or I'm not getting the full picture or you're using the wrong lingo

#

tell me if you get to the page

#

that's the ONLY internal url you should be using

eager shale
#

doesnt work

empty bloom
#

dude, stop using dns

#

click the link above

eager shale
#

i also tried in incognito

empty bloom
#

literally click the link I posted

#

if that doesn't work, then you have the wrong IP for home assisstant

eager shale
empty bloom
#

Then you have the wrong IP for home assistant or your router is lying to you about the port forwarding being off

eager shale
empty bloom
#

remove the ssl things from http section in configuration.yaml keep the rest

eager shale
#

so wierd my laptop isnt working

fallow aurora
#

I have installed generic image on my tiny pc. I have ui and Iam trying to push backup, but I have answer 'upload failed - 400 Bad Request'. Whats wrong? I am reading on internet, there are many topics but no clear answer

hard magnet
#

anyone help?

i have burned countless images to a pi 5, it doesnt flash any kind of network driver as no lights flash on the ethernet port, ive imaged using balena and pi imager and nothing, im loosing my mind, i have plugged into a monitor and can get the command line, but i dont have an ethernet cable long enough to get it close to a monitor.

i dont know the commands to contect to wifi that sounds like alot of hard work

#

i have thought about compiling my own image as could get a wifi config on, there has to be an easier way surely

grand pivot
#

what image are you flashing?

hard magnet
#

is PI 5 Faulty on arrival? im really not sure

#

the linked image in the getting started thread

#

the official

#

haos_rpi5-64-12.1.img

#

is there a better stable image roll back and update after install?

grand pivot
#

I don't use any sort of Pi, so somebody else will need to help. My feeling is that if the Ethernet port isn't working, moving the whole thing to a monitor with Ethernet disconnected is probably fine for this debug activity

hard magnet
#

kk just waiting for it to boot up

hard magnet
#

okay i actually fixed it

#

was easier than i thought connected through CLI

#

just need to make sure it saves details some how and update it. would be good to see if its faulty or sofware / firmware

#

there were both disabled! in the fresh install

#

might want tocheck the compiled images.

#

Failed to change network settings
Can't create config and activate end0: Connection 'Supervisor end0' is not available on device end0 because device has no carrier

sterile scroll
#

I'm still stuck in 2023.12.4 due to Aqara motion sensor issues with the newer builds. I'm running core install on Python 3.11.4 and was going to update to 3.12.2. just wanted to confirm that 3.12.2 will work with 2023.12.4 as I've found conflicting information.

humble mirage
#

What makes you think 3.12.2 won't work?

empty bloom
#

It has too many 2s in it

grand pivot
#

When did HA start officially supporting 3.12? I can't find the release

#

Looks like 2024.2?

humble mirage
#

Sounds right, it was mentioned in the release notes

sterile scroll
grand pivot
#

so based on that, I don't think that running 2023.12.4 with Python 3.12 is a supported config

humble mirage
#

You need to be on 2024.2 for Python 3.12

#

Or just move to Docker

grand pivot
#

right, just answering the question they actually asked ๐Ÿ™‚

#

ideally, don't stay on an old build indefinitely to avoid some specific issue, report it and try to get it fixed

sterile scroll
#

I saw your message after Rob and agreed. Going to attempt the upgrade to 2024.3.x after the python upgrade completes and see if I can get some zha logs this time.

agile mason
#

Hi all, is it possible to have someone mentor me on the configuration? I'm will to pay. I have a raspberry pi5 and have HA up and running. I just need help on yml for frigate NVR.

humble mirage
shy yew
#

Hello, I just bought a skyconnect and connected to my NAS running latest HAOS. I did not detect anything, I installed Matter and Thread integrations and I want it to work as Thread Border Router. Any idea?

raven forge
#

Did you pass it through to the VM? You are using a VM, right?

shy yew
#

Yes and yes

distant mirage
#

Help! ๐Ÿ™‚
After a modem reset by my provider I cant access my Home Assistant via my Duckdns url anymore. I can reach my HA on the internal IP, but on the Duckdns url I get a page with the HA logo and a text; Unable to connect to Home Assistant. Retrying in x seconds.

I've already re-added a static IP for my HA device (same as before the modem reset), set port forwards for ports 8123, 443 & 80 to my HA's internal IP. REbooted HA and modem afterwards, but no dice. What am I missing?

#

Nothing changed on the HA setup, it was working fine before the modem reset.

fallen lichen
#

That sounds like your IP changed... perhaps your DNS server is caching the old value... or DuckDNS wasn't actually updated

distant mirage
#

Ip is the same, checked that on duckdns to be sure

fallen lichen
#

Run these two commands

See if the Address: in the Non-authoritative answer section match

nslookup yoursubdomaingoeshere.duckdns.org resolver1.opendns.com
nslookup myip.opendns.com resolver1.opendns.com
distant mirage
#

where do i run those?

fallen lichen
#

Windows command line? Any terminal

distant mirage
#

k

fallen lichen
#

Assuming your PC is running from the same internet connection as your HA

distant mirage
#

yes they match

fallen lichen
#

By reset your modem... did they just "reboot" it, or was it a hard reset where you had to reconfigure everything? I'm assuming it's more of a modem/router combo since you mentioned port forwards and whatnot

#

Also... you mentioned port forwards for 8123, 443, 80.... you usually pick one, not 3 and that one should be 443

distant mirage
#

Issue was slow connection, so I did a manual reboot first, didnt help, contacted provider and they did a full reset back to factorty defaults.
When I set it up like 4 years ago I remembere I did some port forwarding, and an IP reservation for the HA machine.

fallen lichen
#

Ah, alright, so your issue is in your reconfiguration there

#

You have SSL configured for HA? Or have a proxy in front?

distant mirage
#

good question, not sure tbh.. I set it up years ago with Duckdns and letsencrypt

fallen lichen
#

Take a look at the http: integration in your config.yaml. I'm pretty sure that just configures HA w/ an SSL cert, no proxy

distant mirage
#

http:
ssl_certificate: /ssl/fullchain.pem
ssl_key: /ssl/privkey.pem
ip_ban_enabled: false
login_attempts_threshold: 5

fallen lichen
distant mirage
fallen lichen
#

So then you just need 8123 externally forwarded to your internal IP:8123

distant mirage
#

I did that port forward

fallen lichen
distant mirage
#

nope that doesnt see it, so it looks like the port forward is not working although its in the modem, ill remove, reboot and re-add it and try again

#

Rebooted modem, tried again, but still not working, removed the port forward and re-created it, but nope..

#

looks like the modem is not actually forwarding the port, without he port I reach the login page so Duckdns is uptodate

night lava
#

[supervisor.resolution.checks.base] Run check for dns_server_failed/dns_server [supervisor.resolution.check] System checks complete [supervisor.resolution.evaluate] Starting system evaluation with state running [supervisor.resolution.evaluate] System evaluation complete [supervisor.jobs] 'ResolutionFixup.run_autofix' blocked from execution, system is not healthy - setup [supervisor.host.info] Updating local host information [supervisor.host.services] Updating service information [supervisor.host.network] Updating local network information [supervisor.host.sound] Updating PulseAudio information [supervisor.host.manager] Host information reload completed

#

It has been installed for 2.5 hours. This is the log of the installation. How long will it take to install?

distant mirage
#

@fallen lichen found it! just creating and saving the port forward didn't actually enable it. Totally overlooked a small checkbox to enable it like a billion times..
Thanks for your time trying to help me, very much appreciated and educational โค๏ธ

fallen lichen
#

Sweet, glad you got it working

pine coyote
#

TL:DR: New install on pi3b via docker. Have changed configuration.yaml file for philips hue emulation with view to controlling alexa.... not showing up when I add new light>philips hue>No bluetooth>Bridge v1 (circular)....stressed out trying to fix!

Hi folks, I've installed home assistant on a raspberry pi 3b running the latest 64bit lite raspbian OS. My goal was alexa integration. I've got home assistant running in a docker container and have given it permission to use ports normally reserved for root on starting it via --cap-add=NET_BIND_SERVICE ...

I SSHed in and edited the configuration.yaml file to include the emulated_hue and specified port 80 as per the instructions. I even added the host IP when I was attempting to debug...

Unfortunately despite now being able to access <PI LOCAL IP>/description.xml, suggesting emulated hue may be working following my changes.... I get a '404: file not found' error when going to the other test page of: <PI LOCAL IP>/api/pi/lights. I don't have any hue lights but my understanding was I don't need them...

Anyway on a smartphone connected to the same local network as my pi I tap to add a new device on the alexa app, pick light, philips hue, No to bluetooth, then pick the bridge v1 circular version...and nothing happens...

I've been tinkering around and asking AI to help me try to debug it. Time for far superior good old fashioned human intelligence. Discord I beg your assistance! ๐Ÿ˜„

#

ps thanks so much in advance! Home Assistant rocks. Can't believe I never tried this project out before!

#

Specifically what I want to do is make my alexa speak out announcements I specify when scripts I've written tell it to...but getting thwarted on the set up hue emulation step.

mental lily
#

i dont remember where i read this but i read somewhere that its better to install HAOS on bare metal than use a VM like proxmox

grand pivot
#

it doesn't matter, it just depends on whether you want to do other things with the device

mental lily
#

ah i see

timber parcel
#

Trying my first installation of Home Assistant on a Raspberry Pi 4.
The Raspi connects fine to wifi and also using a CAT5 ethernet cable in Raspbian.
But when booting into Home Assistant from the SSD it doesn't show up on my router when using either wifi or ethernet cable.

The error I'm seeing during the boot sequence is a FAILED on the kernel time synchronised.
It keeps booting and eventually gets to the same screen as shown in the image :
https://imgur.com/cvRMPge

Any help would be appreciated.

mental lily
#

try giving it a static ip or increasing your dhcp range

timber parcel
mental lily
#

oh, huh weird

timber parcel
#

indeed

mental lily
timber parcel
#

it's on a new SSD

mental lily
timber parcel
#

i'll try that now

mental lily
timber parcel
#

How do I determine what the "HA instance is called" ?
edit
Seems that a long list is presented when using "nmcli", but only the ethernet one with "nmcli show" command.
Can't save when on wifi - and haven't a long enough cable for wired connection.

timber parcel
#

Looked over the 2 links, but with no success there either.

long wyvern
#

@timber parcel what guide are you following that's using that phrasing? "What the HA instance is called"
My HA instance doesn't have a "name" but the system it's running on has a hostname

#

I've never tried to run HAOS on bare metal, but I would guess it may not have much support for handling wifi

timber parcel
#

I guess Iโ€™ve been following a multitude of guides.
First I tried the โ€œsupervisedโ€ method, but it told me my version of Raspbian wasnโ€™t a true Debian OS so wouldnโ€™t install.
Yesterday I spent about 4-5 hours trying different things to use the Pi Imager tool to install HAOS onto my SSD, but no matter what I tried, I get the Time Sync Failed issue message during boot.
I think today Iโ€™ll try to understand what Docker is and try that method.

timber parcel
timber parcel
olive elk
#

hi im reinstalling home assistant from a backup, and on my old instance i used duckdns so right now its showing Ripristino in corso but after it use duckdns the page will stop working so how do i know when everything has finished

crude inlet
olive elk
#

and it's not working

#

after restoring backup like 1hour is passed

crude inlet
#

Your ducksdns config would need the router forwarding it used in the past.

#

You should continue to try to access locally using the ip address:port

olive elk
#

fixed thanks, how do i check if the bluetooth is working on my home assistant instance?

crude inlet
#

It should show up as a discovered integration if detected. You can restart ha if a dongle was recently added

long wyvern
# timber parcel Can I ask what method you use ?

Glad you got it solved - I do everything in my power to hardwire machines that need to be remotely available 24/7. Use either a static IP, or a DHCP reservation to make sure it's internal IP doesnt change.

long wyvern
crude inlet
#

@olive elk ๐Ÿ‘† this is the way to go if you want practical bt for the whole home. One or two BT Proxies can be placed for best coverage. More if needed.

ornate adder
#

uhh for some reason i cant access home assistant

long wyvern
#

make sure you're using the httpS:// prefix when accessing the homeassistant URL

ornate adder
#

yeah i am

#

wait would i need to specify the port anywhere in the configuration?

long wyvern
# 
http:
  server_port: 8123
```I wonder if there's some default firewall on debian that's preventing outside access to the docker port
ornate adder
#

might be iptables

#

ill open up the port using it real quick just incase

#

neither options fixed it

long wyvern
#

FYI, remember earlier that /etc/letsencrypt/live/{domain} on the host is most likely a symlink, and docker won't resolve symlinks

ornate adder
long wyvern
#

for me:

$ ls -al /etc/letsencrypt/live/supernerd.me/
total 12
drwxr-xr-x 2 root root 4096 Mar  1 12:46 .
drwxr-xr-x 3 root root 4096 Dec 22  2020 ..
lrwxrwxrwx 1 root root   37 Mar  1 12:46 cert.pem -> ../../archive/supernerd.me/cert20.pem
lrwxrwxrwx 1 root root   38 Mar  1 12:46 chain.pem -> ../../archive/supernerd.me/chain20.pem
ornate adder
#

its not, fullchain2.pem and privkey2.pem are not symlinks

#

only the other ones are

long wyvern
#

fullchain2.pem and privkey2.pem are the ones you need?

ornate adder
#

yep

# 
  server_port: 8123
  ssl_certificate: /ssl/fullchain2.pem
  ssl_key: /ssl/privkey2.pem```
#

thats what ive done for my config

long wyvern
#

that SSDP error should not keep HomeAssistant from starting - what does 'docker logs homeassistant' show?

ornate adder
# 
s6-rc: info: service s6rc-oneshot-runner successfully started
s6-rc: info: service fix-attrs: starting
s6-rc: info: service fix-attrs successfully started
s6-rc: info: service legacy-cont-init: starting
s6-rc: info: service legacy-cont-init successfully started
s6-rc: info: service legacy-services: starting
services-up: info: copying legacy longrun home-assistant (no readiness notification)
s6-rc: info: service legacy-services successfully started
2024-03-31 17:22:36.945 WARNING (MainThread) [homeassistant.components.ssdp] Could not set up UPnP/SSDP server, as a presentation URL could not be determined; Please configure your internal URL in the Home Assistant general configuration
long wyvern
#

and your docker-compose?

ornate adder
#

docker compose only showed the last line

long wyvern
#

no, the file itself

ornate adder
#

oh

# 
services:
  homeassistant:
    container_name: homeassistant
    image: "ghcr.io/home-assistant/home-assistant:stable"
    volumes:
      - /home/homeassistant:/config
      - /etc/localtime:/etc/localtime:ro
      - /run/dbus:/run/dbus:ro
      - /etc/letsencrypt/live/redacted:/ssl
    restart: unless-stopped
    privileged: true
    network_mode: host```
long wyvern
#

okay, do this
docker-compose down
then
docker-compose up (no -d this time)

#

it should spit out the error if there is one

ornate adder
#

nope no error

long wyvern
#

ugh. May as well put it back

#

ctrl-c
then
docker-compose up -d

ornate adder
#

yea done

long wyvern
#

wait

#

why would the guide leave out the port in docker-compose.yaml?

#

that's right, cause it's using network_mode: host

ornate adder
#

uhh so would i need to change anything or na

long wyvern
#

not yet, I"m spinning up a vm

ornate adder
#

alr

long wyvern
#

this is a new install, it shouldnt have any configuration issues - all you've changed was adding the ssl key stuff in configuration.yaml

#

do me a favor, copy the entire configuration.yaml file to a pastebin - so I can have identical sources

ornate adder
long wyvern
#

I have a Debian 12 VM running, installed with docker-compose, and copied all your stuff except for the ssl

humble mirage
#

Did you run the config check command?

ornate adder
long wyvern
#

and it loads correctly

humble scaffoldBOT
ornate adder
humble mirage
#

You need to:

  1. Check the config
  2. Remember to use https:// after restarting HA
ornate adder
# 
Testing configuration at /config```
#

thats my output

ornate adder
long wyvern
#

and using the full domain (that's what the cert is signed for)

ornate adder
#

wdym the full domain

long wyvern
#

when you did the letsencrypt cert, you specified a domain

ornate adder
#

yeah

#

and im using that domain

long wyvern
#

You said it worked earlier before adding the ssl stuff into the config file - Were you also using the domain name to access then?

ornate adder
#

no, i was using 192.168.0.192:8123

sinful bobcat
#

Hey guys! So Im going to be moving from ESXI to Proxmox, What would be the best way to go about that? Convert VMDX or install Home Assistant on Proxmox and restore a back up from the ESXI home assistant

long wyvern
#

I'm not sure - but I think you're dealing with a split DNS issue - the domain (and cert) is configured to be accessable from external, but may not resolve internally.

ornate adder
long wyvern
#

@sinful bobcat make a full backup of HomeAssistant, and restore that backup to a new install

ornate adder
sinful bobcat
#

@long wyvern Awesome! I have both a full and VMDX / OVF. Shouldnt have an issue with cloudflare tunnel

long wyvern
#

@ornate adder Open a windows command prompt, and do nslookup <domain>
whatever IP that reports is the IP HomeAssistant needs to be listening on

ornate adder
#

ive got it tunnelled through cloudflare

long wyvern
#

okay, I've never touched cloudflare

sinful bobcat
#

Ill play around worse case is I have to install esxi back for a bit

long wyvern
#

since you're running multiple services on the same box, I would set up a reverse proxy and have it communicate with the internal stuff - then you only have to deal with 1 SSL cert and the internal proxy would use HTTP for internal communication on the same host

#

I'm pretty sure it's DNS-related, although I feel you should get a 'certificate mismatch' error instead of failing to load completely. Unless cloudflare is verifying SSL when it tries to connect to the backend host

ornate adder
#

ill try setting up a reverse proxy then

#

hopefully that works

long wyvern
#

OpenMediaVault is using the other cert already installed and working?

#

You would ideally set up a wildcard cert for the reverse proxy, then set up the services as subdomains. The subdomains would automatically be included in the wildcard cert. When you connect to any subdomain, you'll technically be connecting to the reverse proxy and it's using the subdomain name to determine what backend service it should connect you to

raven forge
sinful bobcat
#

@raven forge You think there will be any issues with the tunnel?

raven forge
#

Not sure.

sinful bobcat
#

@raven forge I just watched a youtube video explaining the new import tool from prox for ESXi

#

Thanks for sending this

sinful bobcat
#

I dont want to be down for long, and today is my last day off, so if its down I wont be able to fix it for a few days

humble mirage
#

What's the question?

near stump
raven forge
#

For yellow specifics such as which lights blink the #yellow-archived channel might be more appropriate.

languid oyster
#

I'm new to HA, and need help trying to access the webui. I followed the setup instructions to the point it tells me to go to the ip:8123, but that connection times out. I've flashed the imaged twice with the same results. What should I try next? Is there a different image I need other than 12.1 from the PI Imager for a Pi5 8gb? Is this the right place to ask?

humble mirage
#

Is the Pi connected by a cable to your router, with internet access?

timber parcel
#

@languid oyster are you able to connect a screen directly to the pi? It will let you see the status and any error messages

spare bridge
#

Hey i am not new to HA but i have an problem i had never before. i bought an new thin client for my home server and installed Proxmox (8.1.10) on it. I used different ways to set up the VM allays the same result getting this error message:

[supervisor.docker.interface] Updating image ghcr.io/home-assistant/qemux86-64-homeassistant:landingpage to ghcr.io/home-assistant/qemux86-64-homeassistant:2024.3.3
[supervisor.docker.interface] Downloading docker image ghcr.io/home-assistant/qemux86-64-homeassistant with tag 2024.3.3.
[supervisor.docker.interface] Can't install ghcr.io/home-assistant/qemux86-64-homeassistant:2024.3.3: 500 Server Error for http+docker://localhost/v1.43/images/create?tag=2024.3.3&fromImage=ghcr.io%2Fhome-assistant%2Fqemux86-64-homeassistant&platform=linux%2Famd64: Internal Server Error ("Get "https://ghcr.io/v2/": dial tcp: lookup ghcr.io: no such host")

rocky onyx
#

Hi @all, I was trying to install Home Assistant on docker Using plesk. Can anybody help me how to map port there?

manic bane
rocky onyx
manic bane
rocky onyx
rocky onyx
manic bane
manic bane
rocky onyx
rocky onyx
rocky onyx
manic bane
rocky onyx
manic bane
rocky onyx
manic bane
#

Mine is basically the same as this

rocky onyx
rocky onyx
languid oyster
raven forge
#

Please share what you see and also what ha network info, ha supervisor logs and ha core logs says.

languid oyster
#

ha core seems to be crashing. Do I upload the images elsewhere and paste the links here, or how should I share that info?

humble scaffoldBOT
#

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.

#

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

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.

languid oyster
#

Thanks. Do I need to find the log files on the SD card so share them as text?

humble mirage
#

Yes, and yes

#

images of text are horrible

humble scaffoldBOT
#

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

languid oyster
#

I'll turn it off and install linux to a vm and try that. I don't know how how copy text from another machine. I can't access it with ssh or it would be easier.

humble mirage
#

You can install the SSH add-on

#

You can also copy and paste the logs from the UI

humble scaffoldBOT
#
Logs

show your Home Assistant logs

languid oyster
#

I can't access the UI, and have no idea how to install addons without it.

raven forge
#

Pictures of your display/monitor are better than nothing.

#

I explained here how to start a temporary insecure SSH server on HAOS: #installation-archived message
But if your networking is the problem that won't help.

languid oyster
#

Networking is okay. I can access the observer, just not the UI. I can get images easily. I have the pi connected to a capture card, I just need to reboot into windows to use it.

languid oyster
#

I'm not sure what happened, but installing 12.2.rc1 has the UI working.

outer wadi
#

So tired of installing haos now ๐Ÿ˜ญ , after all the issues a week or so ago trying to install it on odroid n2+, i gave up, come back and thought screw it ill install it on an SD and now, it seems to have booted but I cant access it for some reason (IP is correct, verified the mac). Its booted and i have a blue light but just not accessible for some reason.

I see 12.2.rc1 is out and sairon has made some changes, i doubt they're going to fix the mmc issue on n2+ but as sd card doesnt seem to want to work, i might as well try.

At some point i might actually get HAOS to work, otherwise i might just return to installing it on my nas and docker.. at least that functions even if it isnt as fast as id like ๐Ÿ˜ญ

#

๐ŸŽ‰ interestingly the sdcard version just appeared, i guess it was slow.. ill finish this off, give 12.2.rc1 a quick go on the mmc - expecting it to not work still - then probably end up sticking with the sdcard for now

outer wadi
#

Yeah, 12.2.rc1 has the same issue i was having before. Dunno of n2+ has changed, or the mmc has changed, or whats going on. When i looked before it looked like it was trying to boot from the wrong device id, but i dont know enough about it.

Hopefully the SD finishes setting up soon, seems to be taking an age.

tender wren
#

My guess would be a faulty eMMC

outer wadi
#

boots perfectly fine with ubuntu

#

im also not the only person having this issue but it does seem rare, maybe a new revision or something

#

It was noticed on logging that a successful boot has:

Card did not respond to voltage select! : -110
switch to partitions #0, OK
mmc1(part 0) is current device
Scanning mmc 1:1...

A failed boot has:

Card did not respond to voltage select! : -110
unable to select a mode : -5
MMC Device 2 not found
no mmc device at slot 2

Not sure why its looing at mmc device 2, when seemingly mmc device 1 is where the mmc is.

#

Between those two boots, it was noted that the working one was using mmc with revision 0.4, but the failed had mmc revision 0.5, mine is also revision 0.5 (and doesnt boot). That could just be coincedence, or there could be something in that. Would need someone with way more linux or haos knowledge than I have.

However, I dont think its something I can solve myself. I looked for a way to force a specific boot device but I couldnt see one. I suspect if it just boots the right device itll work fine

dire cape
#

Hi, I am trying to install Home Assitant Core on a linux sbc (running Linux 4 Tegra, Ubuntu 20.04), get to ""pip3 install homeassistant==2024.3.3"" and get the error ERROR: No matching distribution found for homeassistant==2024.3.3, latest version showing in the list is 2021.12.10, anoyone able to guide on where i went wrong?

humble mirage
#

Your version of Python is too old

dire cape
#

running pyhton 3.13

humble mirage
#

Why are you running a non-release version?

#

Alpha versions aren't supported

dire cape
#

ah hang on, looks like the venv is using 3.8.10 for some reason

humble mirage
#

You have to use 3.12.x right now, until 3.13 is stable and HA adopts it

#

You can save a lot of pain by just using Docker ๐Ÿ˜‰

#

(the docs say 3.11, but they're outdated)

dire cape
#

let me roll it back quickly then, i am not familar with docker, hence not taking that route

humble mirage
#

Docker makes life easier, it's worth learning

#

I say that as somebody who ran a pure Core install since 2016 and just recently switched to Docker

#

Makes venvs so easy to work with

dire cape
#

thanks for the quick response, installing now after mucking with python venvs for far too long till i got 3.12 venv running properly

humble mirage
#

Just keep an eye on the release notes once 3.13 goes stable since 3.12 support will be dropped a few months later

grand pivot
regal mural
#

Hey, does any one know how I should go about if I want to change the hostname (boot url/default url) before installation. I want it to boot to the address I specified, rather than the default. I am confused if its core or HAOS functionality. So any help will be useful. Thanks!

ember basin
#

Hello, by any chance someone has an up-to-date script for installing ha core on termux?

humble mirage
#

That's unsupported

#

There's probably some guides on the forum, though whether they cover Python 3.12... shrug

ember basin
#

do you have by any chance a script for ubuntu/debian? I might just try to use a proot distro

humble mirage
#
  1. Instally Python using pyenv
  2. Follow the HA docs for a Core install
  3. Reconsider your choices and move to Docker
ember basin
ember basin
#

Almost got it running, altough I can't fix a specific error: It fails to build maturin, any idea about why it happens? Give me 10s and I will share some logs

#

I have rust installed