#installation-archived
1 messages ยท Page 32 of 1
You can sell it or run other stuff like PiHole
I cant seem to get zigbee2mqtt to work in a clean docker installation. after install of homeassistant, duplicati, mosquitto, Zigbee2mqtt, I get a "10.120.12.200 refused to connect" message
cant figure out what is wrong ๐ฆ
#zigbee-archived can better help - this channel is for installation of HA itself
As always though check the log file
There's a 100% probability that the log tells you the problem
wheres this log file?
In the Z2M config folder
If you need further help see the #zigbee-archived channel
(or the docker installation channel of the Z2M Discord)
right. Thanks. I will go check it right away
Tink but my rpi runs 1.13 so well
I have tried installing Home Assistant multiple times on proxmox using a guide and using the community automatic install script. I have gotten to the web UI but it gets stuck in setup trying to install Downloading docker image ghcr.io/home-assistant/qemux86-64-homeassistant with tag 2023.5.3. and retries the download but it never succeeds.
Nevermind after 10 hours it suddenly worked...
Can i install Homeassistant on the Raspberrypi Image/OS? I wan to install the true VSCODE on the pi, ALSO, VNC Viewer
Sure, via docker
I hope you have an RPI4 with 4GB+ of RAM. I don't know what you mean by 'true VSCode', but there's no need to run it in a GUI on the Pi
I have a pi4 with 8GB RAM. How do i do VNC by docker?
Do i install VNC in the root of the Pi4? So then the Homeasistant is just on top of that?
VNC has nothing to do with HA
you're just talking about running a bunch of things on the Pi, 2 of which are HA and VNC
I don't know whether you think you would use VNC to access the HA interface? You wouldn't
Rarely do servers require a GUI or VNC
@winter glacier what do you mean by "true VSCODE on the pi, ALSO, VNC Viewer"?
Home Assistant is a web-based app, you configure/use it via webbrowser, from wherever.
Are you talking about non-HA usages of the same hardware?
the addon is "studio code server" not vscode
iirc it's a stripped down version idk
That's not true. It just provides a web interface
There's another addon that just provides the server for a native installation elsewhere
I installed the Home Assistant OS on my raspberry pi4, it works great. I want to install VSCode, i had it installed on the Raspberrypi OS, but i switched from that to the Homeassistanmt
Studio Code add-on is VSCode in a container. I believe it is only different in name
Yes, but thoโฆ I donโt like a constant connection like that, because of security
Can i create a Rasbian OS, in a EASY home assistant way to view and use on my tablets?
You are still the host with the addon version
@winter glacier it doesn't matter how you install HA. You'll always be accessing the UI through a web browser. Whether that's from desktops, laptops, phones, or tablets doesn't matter
Containers are apps, but isolated from each other. HA's Add-ons are containers. They are installed on the same host.
Ok, so, am i able to just install it like a regular program on a computer? Or will it be the whole OS? I just installed docker
with all due respect, if you're asking that question, you're not ready for HA in docker.
HAOS is not a general purpose OS, it's really cut down and just for running HA plus the addons (each in a docker container). That being said, there are some apps that you can install as addons, via HA, that may do general "not very HA" things...
Ok ๐ I installed HA on my Mac, and its just like a normal App. Thatโs what Iโm wondering about for a Raspberry pi on either the piOS or maybe Ubuntu OS. Because i want to also have VSCode installed on the pi
what do you mean "installed HA on my mac"? you mean in a virtual machine ?
(vmware, virtualbox, docker (technically not a vm), etc)
do you mean the HA companion app is installed on your mac?
@winter glacier forget everything you know and assume about HA OS and add-ons, and vscode
Yes, the app from the MacOS store. It works really well
That's not HA. That's a client. Little more than a browser in a wrapper, from what I can see...
Hi. First time installer here. I'm wondering how to install mqqt... but in hind sight I guess I installed the wrong way. via this guide, https://www.home-assistant.io/installation/raspberrypi rather than "supervised" which I'm confused about Do I have to wipe my os or can I nuke docker and reinstall with the supervisor... and I guess. I'm relatively a linux noob. I know enough to use basic command lines, but rely on online guides/handholding. FWIW, I'm on an Orange PI 5 with armbian
I don't actually see a guide for supervised, and I just choose the raspberry option because I thought it would be the most comparable.
mqtt is an addon, which is another docker container (beside HA itself). Unless you have docker skills, or time to learn, installing supervisord-based solution (without HAOS) is extra work.
Youโre setting yourself up for failure man
Armbian isnโt supported, only Debian is, and supervised is not meant for noobs so you shouldnโt use it
Yikes
But I will install debian instead.
Guys who donโt know how to receive help for $500 Alex
not doing it IS the solution ๐
There literally is no solution given your inputs
debian support armbian not.
Thank you, I will install debian, I found the github page for supervisor, looks easy enough to follow the guide
It will break
Dude, unless you need your machine for other things, just install HAOS. So much easier.
Itโs meant to be installed by sysadmins only
no HA OS image for orange pi
That leaves docker as next best choice
Yeah
so, docker = container but no supervisor support
So it's not the best choice
I am not following because if you're saying it's the best choice... that's how I thought I installed it according to the guide on the website. But the documentation says its not supported.
I'll just wipe and install debian, and follow the git instructions, it looks easy af.
I'll let you know how it goes
Using the supervised installation method will force you to learn a lot so that's good but given its restrictions it has no benefits over HAOS. Due to its requirements and support you will have all the work but it's not really your own so to speak. Given that your hardware isn't supported by HAOS and you insist on having the supervisor you have little choice. Very few will recommend or can help you with it though. See here for installation methods and how they differ: https://www.home-assistant.io/installation/#compare-installation-methods
I just wish the main page was more helpful. It mentions the super visor. But when you watch yt videos people are often talking about installing plugins and stuff, so I woudln't know that I'm supposed to do something different.
And would actually assume that was a "core" feature, which I guess it is through the HA OS
Can't really control YouTube people
I guess I just don't understand why the supervisor can't be adhoc
it's an ecosystem
But docker im my limited understanding can be used to create an ecosystem.
That's exactly what was done
And it is. With HAOS
So is it just that the HA team doesnt care enough to implement that (and I say care not judgementally) but more like they think it's not so important and people can do otherways
Maybe this helps to understand what HAOS is: #general-archived message
@ivory fjord you can't stop anyone from copying the ecosystem, so in order to support it, rules and requirements were put in place (debian, for example)
The whole "unhealthy"/"unsupported" system exists because of that basically
So it checks that a user is staying within the same requirements as the main project
Okay, well my bad I guess for jumping in without researching which os first
I actually thought about asking here but I got too excited
Otherwise, a plain docker install allows you to build your own ecosystem. Find an mqtt broker image, and run that instead of an add-on for instance.
debian install, easy peasy
Hello, Nginx is constantly stopping and I am getting the following error, can you help?
@weak jacinth I converted your message into a file since it's above 15 lines :+1:
More for #add-ons-archived
hey gents good morning
I'm seriously thinking about getting rid of my HAOS VM instance, and would rather migrate my HA to a docker container.
thing is, no backup restore button or whatsoever is existing in the docker container version
I see I'm supposed to move the content of my config folder to my docker container path, but even after a reboot, nothing seems to be back there
do someone have recommendations for me to achieve this migration successfully ?
Yeah you have to restore the data of the homeassistant container manually. Any reason for the switch? How did you create/start the container? Where are you moving the files? Did you stop it before moving the files?
hey Impact ! ๐
well, the reason is I run aLL my other services through docker, and this is the only VM I am running
I see. Makes sense. I do the same actually
also, I plan to eventually lower my power consumption at some point, and i like the fact moving docker containers to a low power consumption rpi instead of my 200W unraid server is pretty easy
200W, constantly, represents 600 euros of electricity yearly in Belgium
so first step is to get everything to run in docker containers on my unraid server, and later on, I might want to only WOL my unraid server when I actually need it
running a script that would wake it when plex, plexamp, or any other power / resource hungry would require it
ok so back to the process
- I simply install the linuxserver.io container using unraids docker GUI
- the folder assigned to home assistant is /mnt/cache/appdata/homeassistant/
- I copied my whole /config folder from my existing HAOS VM to my unraid server
- I copied the content of /config folder to /mnt/cache/appdata/homeassistant
what you say is I should not start HA docker container first, is that right ?
first copy -> then only start the container ?
ok doing this now, the docker cotnainers starts by default when using unraid
Did you copy the hidden . files/directories as well?. I would start the container so it can create folders and such, then stop it and move the data. Maybe clear the folder first.
Make sure to copy the contents of the config folder to /mnt/cache/appdata/homeassistant/, not the folder itself ๐
I have to check if it is possible NOT to start it
Just don't start the container
Tinkerer, as I said, thing is, as soon you install the docker container, it starts
Is it safe to install it, let it start, stop it, copy the content, and restart it ?
That's what I would do so it can create the mount point with the right permissions and so on. I don't think this is needed but clear out the contents of the folder after stopping the container to be safe. Also make sure you set the permisisons of the moved data correctly. The official container uses root but linuxserver's does not afaik. Not sure if it chowns itself
ok first thing first I made sure to include the . files when copying them using the scp command
I installed home assistant
I'm running rm -rf * inside /mnt/cache/appdata/homeassistant/ to make sure it is now empty
Depending on the shell options that will not delete . folders/files: https://unix.stackexchange.com/questions/289391/why-does-my-command-rm-rf-not-delete-all-the-files-in-a-directory
Ah... the joys of shitty GUIs
yay
I agree, but still, it's nice to have on a daily basis
now, HA starts, shitload of errors in teh logs, but when I start its webUI, it keeps asking me for a user / pass
...
It will, you have to log in ๐
no, it asks me to create a user, and a password
it is definitely not reading the db i fed it
ls -lah /mnt/cache/appdata/homeassistant might be interesting
And in those there should be a load of files
Including ones about auth and onboarding
It sounds like those are missing
what filename am I looking for ?
frontend.user_data_818d8cb8f04646b5bb6fdaeb4a1f23f5 ?
I have 3 users, 3 of these files
all located inside .storage ?
Yes, that's what I said before ๐
ok let me post the content of .storage somewhere for you
After a restore, and before you start it up
Try these
ls -lah /mnt/cache/appdata/homeassistant
ls -lah /mnt/cache/appdata/homeassistant/.storage
I'm pretty sure it is due to the fact the docker container starts right after the install
I'm gonna remove it again and use the cli to install it
because the files you mention are inside the .storage folder I copied
auth, etc.
Might also be a bad mount point or something. More information needed
well, to be honest, SOMETHING is happening, as I can see a lot of little errors in the logs which are refering to some of my automations, devices, etc.
so, HA is reading these files
Share them and the output of the commands above. It might also be a good idea to check the container's logs
ok
by the way, it is such a pain we cannot simply recover a backup sing the dockerized version of HA ^^
You technically can, just unzip the backup and put it in the mount point of where you're putting your docker to
oh that's a good tip thanks Calme ๐
Isn't this what you've been doing already?
I'm still interested in the current error logs
I'm reinstalling the container again, will stop it as soon it gets installed, and will untar the full backup file
no that I have another clean install:
root@Tower:/mnt/cache/appdata/homeassistant# ls
automations.yaml configuration.yaml home-assistant.log home-assistant_v2.db scripts.yaml tts/
blueprints/ deps/ home-assistant.log.1 scenes.yaml secrets.yaml
root@Tower:/mnt/cache/appdata/homeassistant# rm -rf *
is it safe to simple run rm -rf in that container folder ?
move the tar file, and untar it right away ?
should I also rm the hidden files ?
Don't see why not but see my hint above rm earlier
rm -rf -- ..?* .[!.]* *
right ?
๐
root@Tower:/mnt/cache/appdata/homeassistant# rm -rf -- ..?* .[!.]* *
root@Tower:/mnt/cache/appdata/homeassistant# ls
root@Tower:/mnt/cache/appdata/homeassistant# ls -lahtr
total 4.0K
drwxrwxrwx 1 nobody users 1.7K May 23 10:32 ../
drwxr-xr-x 1 nobody users 0 May 23 10:39 ./
root@Tower:/mnt/cache/appdata/homeassistant# cd
root@Tower:~# cp 44df32ab.tar /mnt/cache/appdata/homeassistant/
it's totally empty now, I will untar the file
the docker container is stopped
goodness what a rm command
I'd either use the find -delete way, do rm -rf * .* or just rm -rf the whole folder and recreate it. Theoretically you can simply overwrite everything but I don't know if onboarding generates temporary files that would not be overwritten/deleted by a simple cp/mv. You could rsync -delete though ๐
I untar the full backup tar and.....I end uf with a load of other tar files
@inland minnow I converted your message into a file since it's above 15 lines :+1:
I guess I need to untar all the other tars now ?
No. Just homeassistant.tar.gz and maybe the other non addon/core ones
ok cool
now, it untared everything inside a dub folder called data
root@Tower:~/data# ls
automations.yaml deps/ image/ secrets.yaml tts/ zigbee2mqtt/
blueprints/ dwains-dashboard/ nest/ sensors.yaml ui_lovelace_minimalist/
configuration.yaml esphome/ node-red/ switches.yaml utility_meters.yaml
coordinator_backup.json_backup home-assistant_v2.db scenes.yaml templates.yaml www/
custom_components/ home-assistant_v2.db-wal scripts.yaml themes/ zigbee.db
I assume I want to copy all the content of this folder to ../ ?
as it is mapped as the /config folder of my docker container
so long question short, is the content of data the content of my actual /config folder ?
IIrc the data folder holds data of a addon/container's data
I have no data folder inside my HAOS VM instance ^^
[core-ssh ~]$ ls
addons backup config share ssl
[core-ssh ~]$ cd config
[core-ssh config]$ ls
automations.yaml home-assistant.log node-red tts
blueprints home-assistant.log.1 scenes.yaml ui_lovelace_minimalist
configuration.yaml home-assistant.log.fault scripts.yaml utility_meters.yaml
coordinator_backup.json_backup home-assistant_v2.db secrets.yaml www
custom_components home-assistant_v2.db-shm sensors.yaml zigbee.db
deps home-assistant_v2.db-wal switches.yaml zigbee2mqtt
dwains-dashboard image templates.yaml
esphome nest themes
[core-ssh config]$
It's a backup thing
yeah, backups are funny, sometimes
ok so, do I copy the content of /data ../ ?
is there a special cd param to include hidden files and folders ?
or will cp -r * do it ?
sudo cp -pr .??* * /path/to/new/config
Or use rsync. You copy the contents of the data folder of homeassistant.tar.gz to wherever you mapped /config to. In your case /mnt/cache/appdata/homeassistant/
In both cases you'll want to check that all the files were created, with contents
A common fuck-up here is that you don't use sudo and don't get some of the files, or you get them but empty
ok so let me show you the results of ls now
and .storage : https://pastebin.com/Xv76AE4u
which files should I cat to check if they really do have content ?
yeah they have content
I can see my devices in there
and now I see a ton of warnings and issues in the logs : https://pastebin.com/Zih97KGg
but I can login !
huray...
๐
it even speaks with z2m already, wow
now to be honest, it's possible I had these warnings before I moved to the docker container
I will now see if things seems to work
thanks for your help guys !!!!!!!!
stopping the VM now, to see what actually works, and what doesn't
ok, things seem to work quite good except for the fact I can't reach the container from the outtisde anymore
I've recreated the NPM host, a new cert, and for some reason, I end up on the HA loading screen, but never end on the dashboard
"Unable to connect to Home Assistant.
Retrying in 57 seconds..."
as I copied the whole previous config, I can confirm I have this in my configuration.yaml file :
http:
# ssl_certificate: /ssl/fullchain.pem
# ssl_key: /ssl/privkey.pem
use_x_forwarded_for: true
trusted_proxies:
- 127.0.0.1
- 192.168.3.242
- 192.168.3.0/24
Check HA's log file
might it be that NPM and HA dockers should not be located on the same IP ?
ok checking the logs now
what am I looking for ?
Any errors
2023-05-23 12:10:23.162 ERROR (MainThread) [homeassistant.components.http.forwarded] Received X-Forwarded-For header from an untrusted proxy 172.17.0.23
hm
There you go
Reboot?
yep
sorry, a restart I meant
it's working now
๐๐ป
I dropped 9 watts since I moved HA to a docker container instead of a VM ๐
this simple little trick made me spare 20,46 euros yearly ^^
it's funny how such small power consumptions, ona. yearly basis, makes a little difference.
ok I have another issue now, I noticed one of my integrations is not working anymore, I delete it, and want to reinstall it and get this error :
Config flow could not be loaded: 500 Internal Server Error Server got itself in trouble
I google for this, and it looks like I simmply have to add my url here :
but ufortunately I keep getting this error when I try to install an integration
logs rells me requierements are missing
homeassistant.requirements.RequirementsNotFound: Requirements for aguaiot not found: ['git+https://github.com/vincentwolsink/py-agua-iot.git#py-agua-iot==0.0.9'].
Unable to install package git+https://github.com/vincentwolsink/py-agua-iot.git#py-agua-iot==0.0.9: ERROR: Error [Errno 2] No such file or directory: 'git' while executing command git version ERROR: Cannot find command 'git' - do you have 'git' installed and in your PATH? WARNING: You are using pip version 22.0.4; however, version 23.1.2 is available. You should consider upgrading via the '/usr/bin/python3 -m pip install --upgrade pip' command.
it looks like this integration tried to use git to install dependencies, and it also looks like the dockerized version of HA has no git command (?)
I created this last issue post
when you try to install it after copying the content in custom_coponents folder, it seems like it tries to pull the integration from github
and because there is no git existing, it fails
I removed the aguaiot folder, added the repository again, re-installed it from HACS, and again :
homeassistant.requirements.RequirementsNotFound: Requirements for aguaiot not found: ['git+https://github.com/vincentwolsink/py-agua-iot.git#py-agua-iot==0.0.11']
No such file or directory: 'git' while executing command git version ERROR: Cannot find command 'git' - do you have 'git' installed and in your PATH?
is there a way to grab a binary of git and to add it to the path ?
git exists in the Docker image
docker run --rm -it --entrypoint bash ghcr.io/home-assistant/home-assistant:stable
Unable to find image 'ghcr.io/home-assistant/home-assistant:stable' locally
...
Status: Downloaded newer image for ghcr.io/home-assistant/home-assistant:stable
7d5861d8bd53:/config# which git
/usr/bin/git
root@Tower:/config/custom_components/aguaiot# which git
root@Tower:/config/custom_components/aguaiot#
not on my instance it seems
I'm using linuxserver.io's version
why would they have removed git ??? this is insane
but ok, I can try to use the official docker image you just pasted
I can see it is pulling WAY more content btw
Tower:/config# which git
/usr/bin/git
OK so linuxserver.io removed git from their container....
and indeed, it works better with GIT
now I can configure my integration as I should
wth...
Well... if I'd known you were using a third party container image.... I'd have noped the fuck out of here already
I'd ask why you'd do that to yourself... but I just have to look at your post history ๐
I was wondering lol
well linuxserver.io is usually a reliable source, nope ?
The one released by the project itself is reliable
If you want support, run the official image
You never know what decisions a third party has made, and what they've broken
Yea they are
How to configure HA with devices on a different subnet?
being a reliable repo, and not removing/adding things that will break HA are not one and the same
we did learn they removed git ๐
indeed. I would say that they aren't malicious, but they also don't necessarily know anything about the HA project and what it needs
yes this is what I was getting at, just said betterer.
What I guess the best, Home Assistant without add-ons is useless
wrong guess
25% of all installs have no add-ons... I'm one of those installs and I can't say I've ever missed them
Mind you... the average install has 12 automations
anyways you get addons with haos, the route everyone should chose when they need addons. (hint: nobody needs them)
without add ons i ca't use sonoff So why is he so good?
yeah, then use haos
Ah... the fucked up Sonoff add-on
but not supervised
yea ewelink smart home on supervised, what do i need to do to be able to control sonoff without add ons?
doesn't work on any my PC's.....
Then run it in a VM
VM on intel atom and 4GB ram? XD
You can run a VM on a Pi if you're feeling lucky
But still, https://github.com/AlexxIT/SonoffLAN
you are fine as long as your host os is not windows
Proxmox is fairly good
but It cant run on this atom with only 1 core it is impossible
what atom is it?
maybe i try other day but i know this can't work good
I don't remember but this is igel d220 someting intele atom z...
I one try vm on my old core 2 quad q9400. It's can't run vm. so how it is possible to run vm on this very slow intel atom
That would run Proxmox just fine
sonoffLAN can work on docker HA without supervised?
Yes
I want see it
Then ... make it so
We'll see if you're still saying that when it throws up Unsupported and stops working
So i just don't make update or find antoher way
it updates itself ๐
ok
And the requirements change, so it can auto-update and suddenly it's now broken
im running VMs on a core2quad with 4gb of RAM that's prolly only slightly better than an atom
Is availabel HAOS on x86 in .iso not .img?
core 2 quad q9400
That'd be fine for Proxmox, not much slower than the laptop I ran Proxmox on for a few years
however, I did realize today that I didn't install the other 4gb cuz it's sitting on my desk
No, it's not an installer, it's a file system image
you sure you enabled virtualization in bios?
but this c2q consume 80W does nothing my intel atom only 5W so that's why i can't use anything else
YES
im running 2 VMs on my c2q with ubuntu as the host and virtualbox
so how to make bootable pendrive useing this i use balenaetcher last time
You can't
you don't make a bootable pendrive, you image the .img to the harddrive
OK thank you so i can just copy this to hard drive?
using balena or something similar
ok thanks for helping
you need to format the drive with the image
yeah you need to flash the img to the drive, copying it won't work
Hey all, this might seem weird i decided to swich over to haos, now i installed it but this machine i am using does not have a network port it only connects via wifi.
once i boot it how do i configure the wifi so it is on my network?
wifi is generally not supported
generally, but how would i go about using it if the pc physicaly has no network port >.<
don't have a USB network card by chance?
nope
usb ports are an issue too because those are on the keyboard that i dont want attached all the time
that's what I was looking for
It's ... not easy to find if you don't know it exists
what is the keyboard connected with?
will take a read through that @humble mirage
im assuming here, that a keyboard with usb ports on it is gonna be connected usb
but I could be wrong
has been known to happen
well I knew it was there, but my wifi didn't catch up yet
nah its got its own weird connector
you mean like... usb-c?
Now I'm curious what device you are running this on
or is it a mac, and using lightning
its got them i just dont have things to connect to it right now
all the old smaller usb ports etc
Looks like it has one USB-C port in the "tablet" but not everyone has adapters lying around.
WiFi might work fine depending on location and your router/AP's shittyness
i love this project already
its mini usb dont think usb-c existed when this thing was produced
its been fine when running with debian and supervised setup but was having other issues there i dont think i set it up correctly so trying haos
i feel like there are no drivers for the wifi card
it has intel wifi
should be ok
please tell me you plan to put a zigbee coordinator in that single usb 3 port 
nah no zigbee for now
aw man
will replace this with something else, just messing around for now
see now this is what I'd run my dashboard on
this pc?
yes
yeah that will happen at some point too
also side note is there any way to rotate this screen my neck is getting sore.
something like xrandr blah blah
i get error wlan0 does not exist with that
if i type
"network scan" it should show me the available network adapters right?
Does ha network info or ip a list the interface? Did you try the NetworkManager method tinkered linked to?
Also see this section of it: https://github.com/home-assistant/operating-system/blob/dev/Documentation/network.md#enabling-wi-fi
weird, I thought I replied, yes you need to run network info to list the adapters
interfaces:[]
ran nmcli radio
and get
WIFI-HW Wifi WWAN-hw wwan
missing enabled missing enabled
hence i think no drivers
or something like that
just "nmcli"
yes
well that should be a list of all the network adapters in your system
inclusive of the containers
nmcli device show
try that
veth(some characters) <- About 6 of these
those would be the docker interfaces almost certainly
you don't see any adapters with a different name?
see docker0, hassio, veth and loop back nothing wifi ish
nmcli radio shows missing under WIFI-HW
well if you know the mac address of your wifi adapter you could see if any of them match
i think its drivers, since none of these are the wifi from what i can tell. And the nmcli radio shows missing which either suggests no wifi card installed or driver issue
might be easier to just go back to running an os and going with docker
i want supervised tho because i am lazy to do all the docker stuff ๐
Anybody running the Plex Media Server addon on an Odroid M1? Thinking about this hardware for my RV HA instance. Media delivery will be one of the main functions this will be doing. Already check out Samba NAS and looks like that will work nicely for adding extra storage (I think).
around 24 hours ago, my cpu hard spiked https://imgur.com/a/WXH9mu5
how can i see what process is causing this
top, glances, docker stats
hey guys, most times when i go to refresh all yaml config i get this error in the corner of the screen and it doesn't refresh all yaml
Failed to call service homeassistant/reload_all. TypeError: 'NoneType' object is not iterable
does anyone know how to fix it?
Morning all, I seem to be caught at the first hurdle. I cannot get my pc (bmax b1) to boot into home assistant, I've followed the guide online and when I boot the PC it goes straight into the bios. Can anyone please help me? I'm pulling my hair out
Does the computer support UEFI?
is this the PC? http://bmaxit.com/2019/minipc_1102/18.html
Did you disable secure boot?
you're using m.2?
hey, i have a problem, i cannot backup my system, i had an automation which made a backup every week, now I have seen that for 3 months no backup was made. I tried to make a backup myself. There was an error, I should check the supervisor log. I have the log but do not see where the problem is, another backup I can not do because my homeassistant is now in the "freeze"
there is the log https://1drv.ms/u/s!AghsoZDPPMkmgs1SvIkygk86SUTq_A?e=NqfUUz
its to long for discord :D
Hi. I'm having some problems with my MariaDB MySQL database after a HA update.
After the update HA is trying to change the charset and collate and the update is timing out after several hours.
My database is just over 100 GB..! Yes, it is huge, but I have more than 500GB free space on the drive holding the DB files so that should not be the reason for the failure and it has succeeded with several DB upgrades as I have run the same DB with HA for several years.
Does anybody have any suggestions that I can try to save the database and keep some or all of the data?
why don't you use influx?
I have started using influx recently, bit I want to keep historical data for many of my entities and for several I want to keep using MySQL and Influx in parallell to have the Recorder data in the native gui
I want to restore a backup (200MB). The moment I give the command, the Pi loses its connection to the network and nothing else happens. If I unplug the Pi after half an hour and plug it back in, it just boots up with the current configuration.
Anyone have an idea how I can solve this?
I don't feel like completely wiping my system, reinstalling it and then hoping the backup will work.
I have now fixed it. I gave it to a professional ๐. He fixed it within seconds
Hello! I've been running HA core for some years now, and about a year ago I could no longer restart HA from HA itself. It just quits when I press restart. I haven't looked that much into it until now, and I can't find any reason it's not restarting by itself. Any ideas? Is this a common issue from some update that was (~a year ago)?
I think that's a known issue, your service daemon should ensure it restarts
If I was using a daemon then it would just be the service launcher that was wrongly configured, that would make sense.
However, I'm not using a daemon - I never did. I like it running in an interactive console.
Well. Not sure tbh. I think it's more of a habit from before it was so easy to get logs directly from HA.
It's always been trivial to get the logs
journalctl
That's how I've done it for ... 5 years
7... actually
well that is a big revelation for me
Somehow, in all my years, I've never known of that command and for ungodly reasons never bothered to look up the very obvious existence of it
It's part of systemd
If you've somehow got a distro that doesn't use it there's going to be something similar
(and that's excluding the log file that's written by HA itself)
My distro uses it, I'm just dumb not to look it up.
Any day you learn something new is a day not wasted ๐
hmm yes I've never made a service for a pyenv before. Nice to see the example.
If you're using pyenv then it's not meaningfully different, hang on...
I think your link was sufficient
[Unit]
Description=Home Assistant
After=network-online.target
[Service]
Type=simple
User=homeassistant
ExecStart=/data/homeassistant/.pyenv/versions/venv-3.10.9/bin/hass -c "/data/homeassistant/.homeassistant" --log-rotate-days 1
[Install]
WantedBy=multi-user.target
``` from my own install
Hmm. So it's just executing the binary without sourcing the pyenv's bin/activate? I'm somehow used to always sourcing it first, but I haven't actually looked at the activation script. There are no important path dependencies etc?
By calling hass from the venv you're effectively sourcing it
The one on the forum does exactly the same, it's just in a different location
Oh i see now
you update HA manually with pip too right? Or is there some clever way I've overlooked here too
What is the reason for not utilizing pyenv to set the global version of Python?
Hi there, did someone managed to install HA OS on an OrangePI 5 and if yes, is there a guide?
I flash HDD disk with HAOS and it work!! but pity i can't get accses via SSH to install something(not using SSH& Terminal add-ons)
I also have problem with my openvpn server accses i can connect using LTE to my main server but I can't connect to another in my home Lan, maybe it's time to use wireguard
Ok, so the option using docker is still the best solution to run HA on an OrangePI?
"using docker...to run HAOS" doesn't really make sense
my understanding is that HAOS isn't supported on the OrangePi, and using a container install is the recommended way to install HA on one
โฆI donโt get it. U say it doesnโt make sense to use docker and then a container is the only way to get it installed, what am I missing here?
Sry my mistake, I meant HA.
Then yes, Docker (aka the Container install method) is good
Donโt know if this is the right place for my question.
I have installed HA on a HP mini PC. On my iPhone app i have local and non-local access. For local access on my Macbook (homeassistant.local:8123) i forgot my username and password. Via terminal i do have SSH access.
Any ideas how i can recover my username / password?
3slow5me
TCPoPigeon
I've said it would probably be faster for me to handwrite the binary packets as they come in
I actually remember that tho LOL
im gig e ftth here
Is this the correct channel for Recorder / MySQL questions
recorder is an #integrations-archived but there is really no reason to choke HA with a 100Gb DB
What even is in that DB
@quasi wedge I converted your message into a file since it's above 15 lines :+1:
#voice-assistants-archived can help with Assist
the thing is i've seen that error when i've tried to do other things in HA, so i'm not sure if it's an assist related problem or something in my install/config/whatever
That is a good point and I intent to trim the database considerably, but I would still like to get it working first.
I have found a few integrations with many entities updateing far more often than what is necessary to record, that have now been added to the list of ignored entities in the recorder.
How big is it now? Just throw it away, influx has everything you need
And please stop cross posting
Problem is that I recently added Influx, and I need some of the older data in the mysql recorder database for long term statistics and tracking. Cross posted because you directed me to #integrations-archived a few messages above. Sorry if that was incorrect.
Problem is that this is not a real issue
Your issue is that your DB carried crap for way too long
The default is 10 days
I am not sure where the problem is. Yes the database is large, but that should not be an issue in and of itself as long as there is sufficient memory and storage for the database, was my belief anyways. I am not here to report a problem with the HA code, but to see if there is someone experienced with this sort of issue who could give me some hints about what to look at to revive the database without dropping everything.
That is the issue though. The database is too large for HA to deal with it
Hints would be trimming it
The database migration code clearly doesn't expect monster databases
You could open an issue about it and hope the dev gets to it
Thanks. The largest table, by far, is the states table. Would it cause any issues to simply delete all rows belonging to certain entity id's from this table?
I presume you would just loose the states for those entities
hey guys, is there a way to tell if my HA has become corrupt? i'm getting some really, really stuff happening that i can't explain?
That's a pretty vague question, but ... the log is always a place to start
logs
you'd have to explain at least some of the behavior, our mind reading spells ran out
ok so just as i say that one of the issues just magically fixed itself
go figure
but the other thing that's been happening is i've been getting this error a lot:
'NoneType' object has no attribute 'items'
i'm getting it when i try to reload all my yaml config in dev settings (despite it saying there's no config issues), when i try to run an assist command it just fails with that error in the logs, and there was one more time i've seen it but i've forgotten where now
it just seems like a weird thing to keep showing up on all the various things that i would have thought are unrelated (assist, configs etc)?
Home assistant, pilight (and ha integration) and sonoff zigbee stick on the same raspberry pi? Home assistant core is the right installation method?
Container (Docker) > Core
Says the guy using Core
I just installed virtual box and haโฆ how do I install the extensions so I can get access to the usb dongles? Or is there something else needed??
its in the settings of virtualbox, on the vm itself. you enable the dongle for the vm, exposing it
Never used docker before.. But I should definitely learn how to use it soon
I donโt see the usb stuffโฆ everything is greyed out
never mindโฆ got itโฆ thanks mucho
Question: Did I make a mistake installing HA Core on Linux? Should I have installed the Docker setup?
I have a lightly used linux laptop that I have for sundries on my network. When looking at the installation options, I've programmed a bit in Python, and wasn't spooked by the "stuff" needed to install it and get it running. I successfully have it installed and have rudimentary automations going, etc. However, I didn't catch that if you install Core, you don't have some functionality. I'm interested, for example to get Node Red up and running, but see now, that you can't just grab the "add-on" and run it - which appears to be standard in the other installation options.
Are there other items like that, which I should I think about before going further down the line with Core? Should I just go ahead and do a Docker install on that laptop instead?
If you wanted add-ons then you wanted Home Assistant OS
Docker doesn't bring add-ons, only HAOS does
Nobody needs add-ons though, they're just one way of installing software
Gotcha - I see - so ultimately, any items like that can still be added on, even with a Core installation.
I just might have to do a little extra install work
and maintain everything
OK - So I have my regular IDE (VS Code) setup now to SSH into the Linux Machine, with the .yaml files all in a workspace. Skipping NodeRed for now.
is there a way to find out why a brand new installation of home assistant on windows virtualbox keeps restarting
no addons installed
is it running out of memory perhaps? I have assigned 2 cores and 2048MB of ram to the machine
Is it the whole VM restarting (aka rebooting) or just the Core container restarting?
I think its the core container, its on the CLI and then it goes all black, and restarts, sometimes during restart process it hangs on "Booting Slot A.."
I only noticed this as of 2023.5.4 btw
That's the whole VM rebooting
I only started playing around with HA a few days ago, it was on 2023.5.3, as far as I could tell it was working fine, then yesterday got prompted to update to 2023.5.4, started seeing these issues, decided to completely erase the entire VM and reinstall from scratch but now it seems to install straight to 2023.5.4 when installing from scratch and it keeps having these issues
Presumably 2023.5.3 and 2023.5.4
yes, sorry
The whole numbers matter ๐
I've not heard anybody else report those problems on a clean install, not yet anyway
is there a way to force it to install to 2023.5.3 on a clean install
No, but you could try a downgrade
Of course I suspect the problem isn't in the Core version
alright, will try a clean install again and see if/when I can reproduce it, then will try a downgrade, thank you
I found I can consistently reproduce it when installing the Advanced SSH & Web Terminal on version 2023.5.4, a few seconds after hitting install it will reboot and fail to install, but sometimes it also just reboots it on its own, will try to give it more RAM and see if it still happens
ahh, cant downgrade, keeps restarting
im running my VM with 2048 ram as well and I have that addon installed and I never used more than say 40% RAM so far
yeah raising the ram didnt fix it, not sure why it keeps crashing
what CPU is your VM host?
2 cpus
yay was able to finally downgrade to 2023.5.3 and now I can install Advanced SSH & Web Terminal without a crash, first good sign, will keep a lookout for other random restarts
yeah much more stable after the downgrade, no crashes so far
I am struggling pretty bad on my first install
23-05-26 18:52:21 ERROR (SyncWorker_1) [supervisor.docker.interface] Can't install ghcr.io/home-assistant/generic-x86-64-homeassistant:2023.5.4: 500 Server Error for http+docker://localhost/v1.42/images/create?tag=2023.5.4&fromImage=ghcr.io%2Fhome-assistant%2Fgeneric-x86-64-homeassistant&platform=linux%2Famd64: Internal Server Error ("Get "https://ghcr.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)")
23-05-26 18:52:21 WARNING (MainThread) [supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30sec
23-05-26 18:52:51 INFO (SyncWorker_2) [supervisor.docker.interface] Updating image ghcr.io/home-assistant/generic-x86-64-homeassistant:landingpage to ghcr.io/home-assistant/generic-x86-64-homeassistant:2023.5.4
23-05-26 18:52:51 INFO (SyncWorker_2) [supervisor.docker.interface] Downloading docker image ghcr.io/home-assistant/generic-x86-64-homeassistant with tag 2023.5.4.
23-05-26 18:58:26 ERROR (SyncWorker_2) [supervisor.docker.interface] Can't install ghcr.io/home-assistant/generic-x86-64-homeassistant:2023.5.4: 500 Server Error for http+docker://localhost/v1.42/images/create?tag=2023.5.4&fromImage=ghcr.io%2Fhome-assistant%2Fgeneric-x86-64-homeassistant&platform=linux%2Famd64: Internal Server Error ("Get "https://ghcr.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)")
23-05-26 18:58:26 WARNING (MainThread) [supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30sec
23-05-26 18:58:56 INFO (SyncWorker_0) [supervisor.docker.interface] Updating image ghcr.io/home-assistant/generic-x86-64-homeassistant:landingpage to ghcr.io/home-assistant/generic-x86-64-homeassistant:2023.5.4
23-05-26 18:58:56 INFO (SyncWorker_0) [supervisor.docker.interface] Downloading docker image ghcr.io/home-assistant/generic-x86-64-homeassistant with tag 2023.5.4.
Sorry, I dont know how to format the text for assistance. I do industrial automation programming but linux syntax and getting help on discord is a weakspot
Looks like dns Issue
Three ``` for code usually but it's a log so it's fine for the length but pastebin etc is fine to share a link too
Has anyone successfully set a static IP for Home Assistant OS in a VM? I'm trying to run haOS in a VM on Proxmox but the network does not run a DHCP server, so the Proxmox scripts are no good, and I get an error whenever I try to set the Ip I get a supervisor timeout error manually.
I am trying to install HAOS. I set up the installer using balena etcher and successfully set up the boot drive. I disabled secure boot and UEFI is enabled. I'm not sure where to specify the EFI entry in the bios or if that's even necessary.
Anyone set up HA on a Libre Computer Le Potato Mini Single Board Computer? https://www.amazon.com/dp/B0BQG668P6?psc=1&ref=ppx_yo2ov_dt_b_product_details Currently a newb when it comes to pis and HA, so I could use some help!
Awesome, thanks. Is that under the network update interface --ipv4-nameserver?
nmcli is archaic at best
I have no idea, feel dumb
It's OK, to feel dumb is normal. When you think you know everything that's when you're nuts
trying to import the qcow into proxmox, it won't boot from it. depending how I create the VM etc, it either dumps me at an efi prompt, or just refuses to boot.
I think I'm just "off" today. even work was blah.
thanks. I'm looking at wundertech.net's guide currently, and I think I see what I did wrong.
sometimes I literally hate working on my own projects. I just want to be braindead when I'm not working lol.
Yea
this is why adults are allowed alcohol
I don't drink, and if I did I'd be afraid of liver damage.
my bloodwork isn't good. high liver, low kidney, could be some kind of damage.
seems like if I eat the wrong thing(s) in the wrong quantity, I have this brain fog that makes me feel like a pleb.
Depends on genes and eating right and water intake and all that. You can drink yourself to death before you turn 30
my doctor told me I was doing everything wrong, and I fixed that quick
picked a new doctor and all was good
well I'm 41 so that's not good
I also drink over a gallon of water before noon, so there's that too
there is a such thing as "too much of a good thing is a bad thing" but idk your situation.
You can drink too much water too
I don't eat meat or fried foods (much) and my metabolism is faster than the drop in the economy
oh wow.
I just made a sous vide steak for the first time and it was absolutely delicious.
going to make more tomorrow.
I also have a deep fryer in my kitchen.
Ok, after google session, cli. I have it installed on a wyse if that matters
Deep frying is not an approved way to install HA
Instructions unclear. raspberry pi stuck in fryer.
@brittle sail I think I was chatting with you earlier about the pi, etc... rethought it and am going with proxmox install after-all. I do plan on doing more with proxmox in the future so I figure, why not?
Iโm sure you could make a deep fryer smartโฆ you shouldnโt.. but you could
Well actually, it may be nice to add a smart plug or whatever it's called between the fryer and the wall. We always unplug it anyway, but if it's there and that's what we plug into, then we know when we leave the house we definitely shut the fryer. ๐
though I would worry about amperage.
https://www.wundertech.net/how-to-set-up-home-assistant-on-proxmox/
Going through this, the manual install section.
Yea thatโs a popular choice
Most plugs do 10-15 amps. A lot of them canโt actually handle 15 or they end up welding the relay contacts together
yeah I don't ever want to approach the advertised max. I got 15 amp plugs off amazon. "Sonoff" brand.
Fun stuff for another day. Custom firmware and integration for home assistant. Easy way to make your own sensors and devices
oOoOoo, interesting.
yeah for me, one thing at a time. I almost never have time for projects, and when I do, I gotta go slow so I don't burn out again.
Itโs also commonly used to flash fully local open source firmware onto devices with Chinese firmware on them like smart plugs
... yeah, I was thinking about that...
maybe after this test I will return these and get the non-lite versions.
I would personally
However itโs a bitch to take them apart and solder on to little pads. They make pogo pin things to simplify it
oh I have to solder to reflash?
You can watch a YouTube video and see if you want to fuck with it
Or buy the pogo pin things that makes it way easier
hm, I haven't soldered in years. Threw out the iron last year before I moved out of state.
oo, maybe I will.
Watch a vid then decide
Also iirc the lite donโt do power monitoring
but now it's in my mind.
But the s31 do
oh yeah, saw something about that.
Good for washing machines, 3d printers, all sorts of things
hmm
as stated earlier today, I'm all about saving power currently.
(as I sit behind my 4 monitors with my server rack next to me :D)
I live in Texas Iโm about not melting
you're lucky, last time I read instructions I ended up with my penis caught in the ceiling fan
The old skin flute ๐ช crapshoot
this ... is going to consume my life...
texas is cute and all, but come to the actual tropics for real heat
Itโs like 80F overnight all summer itโs rough
it's 84F right now at 11:15pm
When I was a teenager, I went on tour to Brazil. Literally sat there dripping sweat in the dead of night.
and it' aint' summer yet
and I'm eating soup rn
so like I said @brittle sail come get some REAL heat
No.jif
when I was in the rainforest the guide was like... you're the only one not sweating
80F at night is like air-conditioning to me
I literally set my AC at 78F
I like it cool
there is a difference between cool and unobtanium
if you want your AC to crap out, you can set it lower here
but it will die
and on that note, I'm goind to bed
Later dude
How do I fix a CLI dns issue
I did this:
dns restart```
did not fix it, still have that internal server error 500
A little more context would be helpful
ya, probably, my bad
@raven forge I am doing my first install ever on a wyse and it is spitting out some errors on my desktop:
23-05-27 10:42:54 WARNING (MainThread) [supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30sec
23-05-27 10:43:24 INFO (SyncWorker_2) [supervisor.docker.interface] Updating image ghcr.io/home-assistant/generic-x86-64-homeassistant:landingpage to ghcr.io/home-assistant/generic-x86-64-homeassistant:2023.5.4
23-05-27 10:43:24 INFO (SyncWorker_2) [supervisor.docker.interface] Downloading docker image ghcr.io/home-assistant/generic-x86-64-homeassistant with tag 2023.5.4.```
most of my programming experience is not in structured text so the syntax is a learning curve
no. hardwired to router
in my router or in ha?
on that box
Whatever dhcp dns is giving out isnโt working
There should be a guide in the docs to change via nmcli is it not working?
I just started messing with nmcli maybe 2 minutes ago, I need to find that doc
Iโm not sure if that is possible before it finishes installing but I think it is
or try nmcli --help
Tbh itโs kinda confusing
Searching here brings up conversations had in the past usually the same thing you want to do
yeah but nmcli config changes would need a full reboot
so prolly not gonna work during install actually
so might need to change them on the router instead
Probably makes sense to check ip a, ha network info, ping and nslookup
This is an nmcli list of code, but below you say nmcli probably won't work during install, should I be trying anything in this guide?
well if you can just change it on the router and try restarting the install I think it would be easier
but, you can try othewise, it can't hurt to experiment
Netgear R6900v2 is my router, dns is set to 8.8.8.8 and 1.1.1.1
Before you change anything, make sure the host cannot resolve hostnames
how do I check that?
Dns or dhcp dns?
#installation-archived message
nslookup google.com, ping google.com
If you set dns in the router thatโs just giving the router itself specific dns servers. Dhcp dns Is what it hands out to clients when they get an ip
I need to find the answer to this, let me dig in this clunky ui
I can just do ha ping google.com? TIL
Or you can check with cat /etc/resolv.conf/ha network info to see what name servers are configured to use. nslookup should also tell you though
under LAN Setup I have "Use router as dhcp server" checked with a start and end address
that's just DHCP
Thatโs the pool of addresses it picks from
are you running stock firmware or DD-wrt
stock and recently updated firmware on the router
usa
do I have to be in nmcli to ping or anything? because ha doesnt like ping
No. Just enter login int the ha > prompt, then run the commands. It should look like #
ya, missed that part. I left that part and didnt go back in
ping says alive, nslookup gave info also
Weird
I'm glad its not something easy and me being dumb, wouldnt be the first time tho
So much for unnecessary to confirm ๐
and y'all believe it
At least thatโs the command shared above
just try changing the DNS server, it takes like 1 second
You can run curl -v ghcr.io or and do a nslookup for it as well to confirm the theory. I'm not convinced it's a DNS issue
plz hold
Famous last words
lmfao
Disable ipv6
Is this a known issue?
Ipv6 config is often broken in home networks, and Alpine requires it to be working if enabled
my ISP decides to break ipv6 on its network every few weeks so I have it disabled all the way down
and I'm not running my 6to4 tunnel so there's that too
I work in industrial automation so plz hold is probably my most used teams message. Whether looking up drawings or changing the logic, nothing is that quick lol
I think it's thinking
did you disable ipv6?
Ope
Hands up, that's on me
I did not
Do I do that in the ha list or stay logged in
To check the IPv6 situation the ip a and ha network info command outputs I asked for earlier would be helpful. You can also try curl -6 -v ghcr.io
I think github isn't even available via IPv6: https://www.nslookup.io/domains/ghcr.io/dns-records/aaaa/
Well, everything. You can
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.
Yeah you don't have a "real" IPv6 address. I'm not sure what
it's thinking
is supposed to mean. I can guess that it times out, but I'd rather not
Curl took a while
Makes sense that it doesn't work as you only have a link-local IPv6 address but even if you had any, ghcr.io has no AAAA record so it's only available via IPv4 anyway
Let me try and disable the IPv6
I'm assuming the address can stay blank if it's disabled?
usually it will still show a local ip even tho it's disabled
I can't load your images because I'm on very limited bandwidth
https://imgur.com/a/7Sbls0b
Disabled in ha, no address tho
I saw that you can't open the images, after I sent the image
Reading is hard
well now that's disabled, did you try changing DNS servers
I did to 9.9.9.9 and 1.1.1.1
and restarted the install fresh ?
ya, the supervisor was struggling to start
I see this in the thing
23-05-27 12:07:42 INFO (MainThread) [supervisor.resolution.checks.base] Run check for IssueType.DNS_SERVER_IPV6_ERROR/ContextType.DNS_SERVER
23-05-27 12:07:42 INFO (MainThread) [supervisor.resolution.module] Create new issue IssueType.DNS_SERVER_IPV6_ERROR - ContextType.DNS_SERVER / dns://10.0.0.38
23-05-27 12:07:42 INFO (MainThread) [supervisor.resolution.checks.base] Run check for IssueType.DNS_SERVER_FAILED/ContextType.DNS_SERVER
23-05-27 12:07:42 INFO (MainThread) [supervisor.host.services] Updating service information
23-05-27 12:07:42 INFO (MainThread) [supervisor.host.network] Updating local network information
23-05-27 12:07:42 INFO (MainThread) [supervisor.host.sound] Updating PulseAudio information
23-05-27 12:07:42 INFO (MainThread) [supervisor.host.manager] Host information reload completed
23-05-27 12:08:02 INFO (MainThread) [supervisor.resolution.module] Create new issue IssueType.DNS_SERVER_FAILED - ContextType.DNS_SERVER / dns://10.0.0.38
23-05-27 12:08:02 INFO (MainThread) [supervisor.resolution.checks.base] Run check for IssueType.MULTIPLE_DATA_DISKS/ContextType.SYSTEM```
wait, does that show my dns server is set to my ip address..?
You did that earlier, no? <#installation-archived message>
curious, if you put ghcr.io/home-assistant/generic-x86-64-homeassistant:landingpage into your browser on your PC on the same network does it load?
ya it loads
Here's a direct link to avoid the imgur garbage if you still need it: https://i.imgur.com/lW8cu7A.jpeg
I feel as though I misunderstood the command I did. do I need to run that command with 8.8.8.8?
or 9.9.9.9 now
It's common to use the gateway ip for this but this should work as well
10.0.0.38 is the ip of the unit
ya, gotta change that
Ok, restarting install after fixing that
What prompted you to change this?
to 10.0.0.38. 10.0.0.1 should be fine (and usually the default) but 10.0.0.38 is the device itself and it probably doesn't run a DNS server
It didn't work before I did that either
Supervisor struggling to start. System is not ready with state: setup
Sorry for the delay, had to run a quick errand. So this is early on in the install stuff:
23-05-25 02:40:05 INFO (MainThread) [supervisor.dbus.timedate] Load dbus interface org.freedesktop.timedate1
23-05-25 02:40:05 INFO (MainThread) [supervisor.host.services] Updating service information
23-05-25 02:40:05 INFO (MainThread) [supervisor.host.sound] Updating PulseAudio information
23-05-25 02:40:05 INFO (MainThread) [supervisor.host.apparmor] Loading AppArmor Profiles: {'hassio-supervisor'}
23-05-25 02:40:10 WARNING (MainThread) [supervisor.utils.whoami] Can't fetch Whoami data: Cannot connect to host services.home-assistant.io:443 ssl:default [Try again]
23-05-25 02:40:10 WARNING (MainThread) [supervisor.core] Can't adjust Time/Date settings: Can't fetch Whoami data: Cannot connect to host services.home-assistant.io:443 ssl:default [Try again]
23-05-25 02:40:10 INFO (MainThread) [supervisor.docker.monitor] Started docker events monitor```
Then I see this fatal error
23-05-25 02:40:34 INFO (MainThread) [supervisor.resolution.module] Create new issue IssueType.FATAL_ERROR - ContextType.STORE / core
23-05-25 02:40:34 WARNING (MainThread) [supervisor.jobs] 'GitRepo.clone' blocked from execution, no supervisor internet connection
23-05-25 02:40:34 ERROR (MainThread) [supervisor.store] Can't add repository https://github.com/esphome/home-assistant-addon due to 'GitRepo.clone' blocked from execution, no supervisor internet connection
23-05-25 02:40:34 INFO (MainThread) [supervisor.resolution.module] Create new suggestion SuggestionType.EXECUTE_RELOAD - ContextType.STORE / 5c53de3b
23-05-25 02:40:34 INFO (MainThread) [supervisor.resolution.module] Create new issue IssueType.FATAL_ERROR - ContextType.STORE / 5c53de3b```
also ipv6 disabled on router
@raven forge @frank perch here is what I am getting after a reboot, I am learning a lot but making no progress lol
Sorry if this is obvious (I did search!) but is there an easy way to see if my installation is Supported? It used to be in one of the settings pages but those have changed a lot and I can no longer find it. I did find it in the observer; is that now the primary way?
(If it matters, I'm running supervised on an RPi4 running Debian 11.)
I'm curious why you didn't just use HAOS
question, is this the same install or did you try one fresh already since the changes
Is that to me or someone else?
Clearly you...
I mean, you've got hardware that's supported by HAOS... there's no reason to use Supervised unless you want to ignore the requirements and run other software...
๐
I'm used to it; I get asked that every time I ask a question about my installation. But since there are many conversations happening and I am not following the rest and it wasn't a reply, I wanted to make sure.
are you here because the logs say your install is fucky wucky?
I do run other software on it; nothing complex enough that I can't maintain Supported status.
I've been running it for years at this point. I can't remember if it was easier to do a USB SSD installation on supervised than HAOS when I first did it or not, but regardless, I do run other software.
And for what it's worth, it is listed as Supervised in observer.
like reflash and start over that way? I have not done that yet if you all wanted me to
I was just wondering where the UI part went.
yeah I would start fresh knowing what you know now
and just do what you need to, I think it's borked
they changed it a year ago or so- supervisor tab is gone it's just integrated in different settings menus
you're the 1% of the 1% using supervised without issues lol
Do you know which one? I think I've walked through every configuration menu at this point looking for where that particularly piece of it went.
what piece? "whether your installation is supported"?
lol. Well I put a lot of work into it and I keep it as close to base Debian as I can and still run my other stuff.
you deduce that yourself
if you are running supervised on debian and following the upgrade stuff necessary then it's "supported" in the sense that you're technically doing the right things to make it not break
Indeed. It used to have a thing that said Supported or not and linked you to things like the cgroup setting that were interfering withtstatus. Maybe that part is just gone. Observer still has this:
Supervisor: Connected
Supported: Supported
Healthy: Healthy ```
(Can't seem to use image.)
yeah
last time I did it I did it through ubuntu, do you see any reason why that wouldnt work or should I get an adapter for usb?
I know the Repairs part has the problems now, and that's clear.
Was just curious if that status was still available in the main UI, and it sounds like no.
sorry, what do you mean you did it thru ubuntu
settings -> about?
you're running HAOS in virtualbox?
I walked through those and didn't see it, though maybe I missed? Text on that page is just Home Assistant 2023.5.4 Supervisor 2023.04.1 Frontend 20230503.3 - latest, and the links all go to external sites.
For sure not important; was mostly curious and wanted to double-check my cgroup fix worked. But not a big deal at all if it's not there anymore.
settings > system > repairs >
> system information ?
FWIW, the cgroup repair instructions don't work on RPi4, at least on a USB SSD install, because it doesn't use grub. I'll provide feedback on the docs using the link there.
Yes!! Thank you. I forgot to click through the dot menu on that page.
Wouldn't hurt, but since it seems like y'all are steering people away from supervised, I could see burying it a little more.
But it's at least visible in the UI, which is nice.
Thank you both for the help!
(FWIW, it was totally stable those 6 months it was unsupported w/cgroup v2, but I don't pretend to know what that actually affects and I don't need v2 for anything I'm doing, so I'd rather keep it in the Supported bucket.)
"y'all" ? what did I do lol
You were here! ๐
I don't steer anyone ๐ no one even listens to me lol
lol
No, sorry. I booted to Ubuntu live on a USB drive, installed the etcher there, and ran the etcher on the wyse that way. It's how whatever YouTube video I found first did it
ahhh I understand now
no, that should work just fine
just I think because the install hung and rebooted so many times prolly best to start fresh and do what you learned
Doing errands now, so install later. But from fresh install, it should (theoretically) just run?
My worry is it is going to try and do the install and hang up because of dns stuff again and I get stuck right where I am now
Idk if you can set stuff before the install or what
@stuck knot I converted your message into a file since it's above 15 lines :+1:
Ah removing the --init fixed the issue. Interesting..
is it possible to run HAOS on a rock 5b SBC?
Ok, downloaded image on one USB, Ubuntu on other. Booted to Ubuntu, wiped the wyse partitions, flashing the image from first USB onto wyse.
Fingers crossed
here we go
Waiting for supervisor to startup
System information
Error returned from Supervisor: System is not ready with state: setup```
Want me to try something like
banner --log-level=debug
Just waited and did banner, wait, banner and it started. Let me check the install status
@frank perch
23-05-27 20:25:42 INFO (MainThread) [supervisor.resolution.fixup] System autofix complete
23-05-27 20:25:42 INFO (MainThread) [supervisor.host.services] Updating service information
23-05-27 20:25:42 INFO (MainThread) [supervisor.host.network] Updating local network information
23-05-27 20:25:42 INFO (MainThread) [supervisor.host.sound] Updating PulseAudio information
23-05-27 20:25:42 INFO (MainThread) [supervisor.host.manager] Host information reload completed
23-05-27 20:31:41 ERROR (SyncWorker_2) [supervisor.docker.interface] Can't install ghcr.io/home-assistant/generic-x86-64-homeassistant:2023.5.4: 500 Server Error for http+docker://localhost/v1.42/images/create?tag=2023.5.4&fromImage=ghcr.io%2Fhome-assistant%2Fgeneric-x86-64-homeassistant&platform=linux%2Famd64: Internal Server Error ("Get "https://ghcr.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)")
23-05-27 20:31:41 WARNING (MainThread) [supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30sec
23-05-27 20:32:11 INFO (SyncWorker_2) [supervisor.docker.interface] Updating image ghcr.io/home-assistant/generic-x86-64-homeassistant:landingpage to ghcr.io/home-assistant/generic-x86-64-homeassistant:2023.5.4
23-05-27 20:32:11 INFO (SyncWorker_2) [supervisor.docker.interface] Downloading docker image ghcr.io/home-assistant/generic-x86-64-homeassistant with tag 2023.5.4.```
ha network info shows you have internet ?
supervisor_internet: true
should be right at the bottom
Iโd be screaming at this point
I've been screaming all day
Must be nice
Are you hiding from the feds or something?
Why are you in such a godless place with no fiber or even broadband
Youโre making T1 lines from 1995 look good
Ya it did. I just disabled IPv6 and set the DNS servers and rebooted
I'm so frustrated it's not funny. Started drinking early
Get drunk enough and buy a mini pc off eBay to replace it
Thin clients are better than a pi but still pretty weak. Fine for base home assistant os and a few addons of course
32gb and 8gb ram, what more does it need
I have a feeling no matter what I install it on its going to do this
Ok it's back up, yes it says true. Checking install
Yea if itโs dns related you need to fix that obviously
gfd
23-05-27 20:42:44 INFO (MainThread) [supervisor.plugins.manager] audio does not have the latest version 2023.04.1, updating
23-05-27 20:42:44 INFO (SyncWorker_0) [supervisor.docker.interface] Updating image ghcr.io/home-assistant/amd64-hassio-audio:2023.04.0 to ghcr.io/home-assistant/amd64-hassio-audio:2023.04.1
23-05-27 20:42:44 INFO (SyncWorker_0) [supervisor.docker.interface] Downloading docker image ghcr.io/home-assistant/amd64-hassio-audio with tag 2023.04.1.
23-05-27 20:49:00 ERROR (SyncWorker_0) [supervisor.docker.interface] Can't install ghcr.io/home-assistant/amd64-hassio-audio:2023.04.1: 500 Server Error for http+docker://localhost/v1.42/images/create?tag=2023.04.1&fromImage=ghcr.io%2Fhome-assistant%2Famd64-hassio-audio&platform=linux%2Famd64: Internal Server Error ("Get "https://ghcr.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)")
23-05-27 20:49:00 ERROR (MainThread) [supervisor.plugins.audio] Audio update failed
23-05-27 20:49:00 ERROR (MainThread) [supervisor.plugins.manager] Can't update audio to 2023.04.1, the Supervisor healthy could be compromised!
23-05-27 20:49:00 INFO (MainThread) [supervisor.resolution.module] Create new suggestion SuggestionType.EXECUTE_UPDATE - ContextType.PLUGIN / audio```
ha core check
For fun try 10.0 install
yeah I was about to say try an older version
I doubt thatโs it but likeโฆ
the ipv4-nameservers is set to the ip address of my router, does that sound right???
Not if you want it to use an authoritative one like cloudflare that your router isnโt handing out
yeah if it's DHCP that's typical
ok, so I didnt touch that
all I've done was add the dns servers that are in my router
Above he never confirmed it was dhcp dns instead of dns just for the router iirc
fuck you may be right
Could be handing out some isp bullshit
how do I confirm that
Idk if he can reach ghcr.io from a client that gets the same nameservers from the router then Iโm tempted to blame haos
Assuming itโs not the ipv6 issue discussed earlier
Iโd be tempted to plug that bitch in at a friends house just to test lol
ha dns options --fallback=false
and for shits and giggles let's see what ha resolution info says
well I have a netgear as well but I'm not on stock FW
https://community.home-assistant.io/t/installing-home-assistant-supervised-on-debian-11/200253/914 guys on debian were getting around it by manually editing the hosts file
idk if haos likes that tho
me? an x10 with DD-WRT
Is it just a search for router and flash it?
i wouldnt flash your router while drinking dude
well you might have to load a prep firmware, then flash it
but yeah don't do that
it's another rabbit hole in itself
I realized it as I hit send
I don't really want to get a different router
no
you don't have to get another router
you can flash yours
but again, not while drinking
It hasn't been heavy enough lol
this isnt even a confirmed issue just an educated guess based on information in posts above + your struggles
I wish there was a way of confirming it



