#installation-archived
1 messages ยท Page 20 of 1
hi, I was setting up duckdns for my ha, I guess I am having trouble figuring out the port forwarding, cause when check on the site canyouseeme.org my port cant be connected.
the error it shows is connection timed out
Does your ISP allow port forwarding
I guess
I have a tp link router.
it has a lot of option https://ibb.co/4NK7DNn
which one should I select
I have no clue, since you just cropped a pic of different protocols
wow, gopher
these are the only protocols available on the router interface
Yes. Whether the router allows it or not is irrelevant if your ISP blocks them
Or has you behind CGNAT
ohk
I have HAOS on my nuc, i opted to update haos from 9.0 to 9.5 an it has gone 25minutes and when i try to connect via browser im unable to connect? Help , what do i doo now?
And, how long should an update take? Should i be worried? ๐ญ
ok, so I made a fresh(first time) install of ha, installed some addons added some integration and devices, everything worked and I made a backup.
[1:18 PM]
then the SD card died (was not a new one) so I got a new one, installed HA and restored the backup
[1:19 PM]
the addons are there but according to the frontend, they are not started (they are) and I cannot install the ssh+ web terminal
[1:20 PM]
the ssh server works so from my main linux machine I can ssh into the system. but I cannot use the web terminal
Try restarting the host not just ha
I have already tried
As in the host itself not just HA. The normal restart button does not restart the host
ok doing that now (ssh;d in and did a reboot
that is a thing I was missing in the front-end a "shutdown/reboot" button
There is one. I just am not sure where it is since I don't use HAOS
> Reboot system
yup found it ๐ well hidden
aand everything is working now ๐
thank you very much
Yay
Any tips on a nuc haos update that wont go through?
Jumbers is the MVP of finding stuff lol
๐งโโ๏ธ
you need people like that
ha ๐จโ๐
I have always done my home automation only with node-red, but making all those dashboards was getting tedious
I have never touched node-red
I have 2 of the Nexus 7's lol
Question: Why isn't there an ISO installer artifact created by the build/release process? It would make it so much easier to install because every hypervisor expects an ISO to install to disk, not a raw disk. Just providing a raw disk image makes installing (especially to hypervisors like Proxmox) quite a lot harder.
You don't need to "install" anything, you just provide the image as a disk. I've never used Proxmox, but in virt-manager (libvirt/QEMU), you just tell it you're importing a disk
Pretty sure Virtualbox does the same thing, but I haven't used it in years
I've used both virt-manager and proxmox, and I found the process quite cumbersome, to be honest. The challenge is actually the work of dd-ing the haos image onto the virtual disk, which is usually not something that is very well explained in the hypervisor docs.
not everyone uses a "file" for the virtual disk.
(and that's talking from the experience of a 20+ years linux sysadmin)
hypervisors in general don't care if its a iso or a ready disk image, they are made to load whatever floats your boat by provisioning. Its the UI and users making this troublesome
Exactly, it is the UI of the hypervisors that is the problem, not the disk image in itself. Most of them don't have a "simple" option of "dd this file to the virtual disk and use that as your boot drive" workflow.
dd this file? there is no need to dd anything
you just point it to the drive image you want to use
it is if you want to e.g. use a zfs zvol as your boot drive for haos
that is not supported
I can't open the ha through duckdns domain, can anyone help pl;s
@low notch Well, for haos it just sees it as a dumb ahci drive, it's the hypervisor's job to manage that part.
no its your job to manage that part, thats why we are discussing this in the first place
else they would release a .img with zfs zvol as format
anyone?
You can't release an .img as a zvol, a zvol is just a block device, as any other, and every hypervisor on the planet supports virtual disks from block devices (in some form or the other).
hypervisors can also just load the premade img and be done, no need for zvol
The short answer is probably that creating and maintaining an ISO-based installer takes time and effort, and nobody has had any interest in doing so. It's an open source project after all, so if nobody is interested in doing it, then it won't get done
@dry compass Are you running the duckdns client on your machine, and is it reporting any issues?
i can access it through https://homeassistant.local:8123/
@odd marlin Of course that is the case, I'm just arguing for HAOS install process being different from anything else in the linux world, where an ISO installer is the defacto standard to release a distro. That's all.
that is very much not the case for embedded devices
I wouldn't call HAOS an embedded device, it's more of an appliance, and most appliance installers are also packaged as ISOs (for anything x86).
At least half of the installs of HA are on Raspberry Pi or similar devices where you can't boot from an ISO. You just dump an image onto an SD card
I'm just arguing if your storage is non-removable that process is more involved, which is made easier with a bootable USB/ISO installer.
But it seems like I'm preaching to a wall here, so I'll go somewhere else. Thanks for be unhelpful and nonforthcoming of new ideas.
who was unhelpful?
I was setting up duckdns this morning done everything but still cant access through the domain
i can access it through https://homeassistant.local:8123/
@dry compass Where are you running the duckdns client?
"where"
addon?
and what does the logs say for the addon?
@dry compass I converted your message into a file since it's above 15 lines :+1:
and have you portforwarded or used a reverse proxy to forward port 8123 from your router to your HAOS machine?
yes
I would not forward port 8123 directly, as that is unencrypted HTTP by default, although I don't know what the duckdns addon does to change things.
Does the duckdns addon actually spin up a reverse proxy? From the logs it looks like that, but I'm not sure. This is probably a question better answered in #add-ons-archived .
Theres a proxmox script out there which downloads the latest HAOS and creates a VM with the requirements all set up for you. One liner lol
@manic bane I'm sure there is, but this is not about "magic" one-liners that hides the complexity. As someone who has run workshops for people getting started with Home Assistant the install process on RPIs is trivial, quite simple actually. But once you want to go a step up and install on a dedicated x86 box because you want some more power the virtualization route is an enticing one. Maybe I should make myself an iPXE boot environment that automatically boots up e.g. alpine inside the VM and lists the drives available and chooses one to dd the .img to and reboots. Maybe that's a better approach. Is there an API somewhere that lists the versions of the various disk images available for each architecture?
haos is for those who have no idea about filesystems and don't care about performance as long as it runs. if you want to differ much from that you may want to chose a different install method (container) which gives you the most freedom
yeah, I've been running Core for several years, but recently transitioned to haos to get the benefits of addons.
If you run container based install everything is just a few extra lines in a docker-compose
the integrated authentication between addons is for sure a benefit
Very easy
Barely
You create a username/password for the addon, then set that in HA, job done
Don't need to worry about anything
If you want to create and support iso creation including an installer for the HAOS image then by all means go ahead with it
HA is open source for a reason
Sure, if I actually had time to devote to that I'd do it. For now I'll have to decide for myself if I'll take the time to write a feature request on GitHub or not. I don't know if you two represent the general consensus of how HA developers feel about this topic or not.
I'm not a HA developer anyway
Just someone who doesn't use HAOS and thinks if you go advanced then go for something like container
I've actually done advanced for quite some time, and now I'm trying to go simpler to reduce the maintenance burden in my spare time. Keeping on top of all the updates to components in my home lab is starting to become a chore, keeping me from exploring the interesting stuff. So the less stuff I need to manually worry about updating the better.
Hence HAOS with addons
From a brief search, I only see 2 feature requests related to ISOs/installers and neither have really gotten any traction
https://community.home-assistant.io/t/iso-image-for-truenass-scale/472048
https://community.home-assistant.io/t/haos-should-be-installable-from-a-usb-drive/341397
I update my docker containers with 1 script, takes me around 5 minutes depending on how many updates/ how fast they come
Then do OS level updates about every fortnight to monthly
How do you handle breaking changes in your docker container's release notes?
that is somewhat the wrong approach, as said haos is for the uninformed who does not care about any of that and just want it to run.
anything outside of that containment amounts in extraordinary work for you, container is the best for sleek performance and customizability
You just blindly upgrade and hope for the best?
breaking changes are the same for every install
@low notch Not talking abotu HA breaking changes here, refering to @manic bane 's one-shot docker upgrade script
I check the release notes first
Then go
I hold off upgrading till I have time to fix anything
The only HA instance that breaks for me is my Beta HA one because I never bother to update things on that until I test stuff out
Since its very much similar to my main instance anything breaking on that is fixed on both before it breaks prod
(And HA is the only one I've had breaking changes for, rest seem to work fine ๐ )
Since I've started using portainer upgrading my docker containers is generally just a few clicks, but when you're running 31 containers across six different hosts it still takes a while to go through all the release notes and figure out the next viable version number (for the ones not running :latest).
All of my stuff runs latest really
Luckily newreleases.io helps a lot out in that department to make it somewhat faster.
I got burned by keycloak when they upgraded major release on :latest with massive breaking changes causing everything to stop working. Been careful since then.
But I guess we're getting a bit off-topic here. I'll get back to my Prometheus deployment, which was on my agenda for today. Trying to migrate off my ELK stack for monitoring.
I just noticed the HA OS update 9.5, checking the releases notes it's only in the beta channel, but I'm not?
@humble surge https://version.home-assistant.io/stable.json
The note needs to be removed probably
Hello Home assistant gods ๐ I have been using home assistant for many month now and i am super addicted and happy. I have thrown to much time and money into this haha ๐ And The new usb sky connect stick i have not been able to test yet ๐
Anyways I have run it on virtual box and i was planning to change PC to run it on. I made copy of virtualbox fully, but i get som EFI problems i think. Does anyone have an ide? Thank you in advanced
Oh i was about to upload img but i see that i cannot do that?
Here you go ๐ again thank you in advanced https://ibb.co/dW3V3cH
How do i browse and clone github files onto certain places in my haos?
Or i cant browse installfiles when i installed the home assistant os?
Hello. I just got Raspberry Pi400 and have flashed Home Assistant to the microSD card. I get HA to boot but upon boot it does not get IP nor does homeassistant.local:8123 nor :4357 work so I can't reach it.
RPi's OS worked just fine with my network.
My router is Unifi Dream Machine and I tried to connect to the RPi by having it directly plugged in to it and via Unifi Switch but no dice, RPi does not get IP for itself when it's booted to Home Assistant CLI.
% ping homeassistant.local
ping: cannot resolve homeassistant.local: Unknown host
Plot thickens. When running login, then nmcli con show and checking again with nmcli I get eth0 unavailable..
OK, minor update. Seems like I'm getting stuck on "Error: Canโt update config on eth0: Connection โSupervisor eth0โ is not available on device eth0 because device has no carrier" but I have no solution on how to continue from that. Any help would be appreciated.
That sounds like a network interface issue, not software. Cable not plugged in, bad cable, etc
Checked and cable is OK, worked fine on RPi's OS and is fully plugged in.
Link LEDs on?
Not with HA and just changed my cable and port from switch
Yeah, just restarted RPi and same result. This is with different cable and port from switch.
I also flashed HA 9.5., from 9.4., to my SD and that didn't solve the issue either
To be clear; I'd want to run HA on a separate installation, i.e. on it's own SD card, instead of piling it on top of RPI's OS. Hence the separate SD card for HA and RPi OS.
Anything interesting in dmesg? I'd say you can try enabling auto-negotiate through nmcli but I'd be guessing at that point
dmesg|grep eth0 only gives "link is down" and then numbers and "renamed from..."
nmcli device show : eth0 GENERAL.STATE 20 (Unavailable)
nmcli connection up ifname eth0 same error?
Error: Canโt update config on eth0: Connection โSupervisor eth0โ is not available on device eth0 because device has no carrier
OCR โค๏ธ
Last chance, trying to boot HA with image flashed to USB flashdrive. Probably does not work but have no other ideas anymore, unless manually editing wi-fi credentials to the configs but that kind of defeats the purpose, as wi-fi is, notoriously, unreliable
Couldn't catch the error but it was along the lines of kernel failing to synchronize time within the 1m 30 s. Makes sense though as there's no connection
But yeah, no, trying the USB flashdrive did not help. ๐ฆ
Is there a way to automate home assistant system updates?
Last update: went as far as to disconnect my router and switch and plugged RPi straight to my incoming ethernet; could still not get the connection to work so it's not my router nor the switch causing the issue but rather, to make a Mac analogy: some kext is not loading and there's no easy way to fix that.
A start job is running for wait Until Kernel Tine Synchronized (52s / 1min 30s)
(FAILED] Failed to start Wait Until Kernel Time Synchronized.
See systemctl status systend-tine-walt-sync.service' for details.
OK Reached target Systen Tine Synchronized.
OK Started Discard unused blocks once a weck.
OK Reached target Tiner Units.
Starting Hass0S AppArmor..
OK Finished HassOS AppArnor.
Starting Docker Application Container Engine..
That's the last logs that I could capture. I'll resume this at another time but leaving those in case they help in any way.
I found someone else having the same issue and posted to their thread, in case someone else runs into this issue as well. https://community.home-assistant.io/t/network-problem-on-raspberry-pi-4/524828
hm the joys of toy computers
Hi, I'm trying to setup Home Assistant on my raspberry pi 3 and I've already burned the HA image and on the sd card and booted the raspberry pi but I can't access it through the browser (which the last instalation step) but if I ping the ip adress of the pi ,which I got from connecting a screen to the pi and entering the command line, it says its connected.
Any idea on how to fix it?
How are you navigating to it in the browser
i think i should posted here not general?.... but is there a spot on a new install of HA to put the backup file.. im fiding the 20 min wait is stuck least 2 hours... i not sure if HA is down where it gets the initial setup but can i copy the backup file to a location on a newly installed HA so it can read it and install from that? as i went from SD card to SSD. using a sabrent adapter in the blue usb port on a raspbery pie 4 and i using a 240gb kingston ssd
The recommended way to run on a Pi with an SSD is to keep the OS installed on an SD card and then use the "Move datadisk" option to move all of your data to the SSD
https://www.home-assistant.io/faq/usb_boot/
https://www.home-assistant.io/common-tasks/os/#using-external-data-disk
ok ill check it out... as i was having ha corruption error and the date and time was set 2 days earlier.. so figured id move from a sd card to ssd as i been told before u wanna run ha off ssd not sd card... but ill check out link
If your SD card has gone bad, you'd want to start with a new SD card as well. Once you move the data disk though, the SD card becomes mostly read-only, except for OS updates
isnt that even worse? now you need both to work..
the sd card i guessing gone flaky as it doesnt always boot... and it cant sync the time.. and you get the error of ha critical error.. and i dont have a spare 32gb sd card.. i do have a spare 8gb sd card.. but how do you do a full restore cuz wont it try to first install on a sd card then you gotta move to ssd card... so i guessing i need another 32gb sd card first then right
Anyone ever see a problem where after a backup restore your addons appear to not be running from the frontend?
As in they are all grayed out and if you go into one you can click start even though it's already started
They are actually running, it's just the frontend. Nbd, but thought I'd ask. Maybe something I could report
Restart the host. Settings > Hardware >
> Reboot system
its probably a problem with cached stuff in the browser
Clearing the cache or just straight opening it in a fresh incognito still has the same result
Yeah that would probably work. Are you saying it's a known issue though?
Every time I've seen somebody mention addons not working after a restore, a reboot of the system fixes it
its a good idea either way to make sure your stuff actualy works
I checked a few logs
This is just my test box. I'm just noticing now because before my addons were completely broken after restore for disk space reasons. Making a fix to my provisioning so it resizes the VM image first
Had some worrying issues before where it looked like my mariadb backups were all corrupt
so i re formated my sd card re installed HA on it and its been more then an hour and still stuck on preparing ha that it install under 20 min... i had a monitor hooked up the rpi4 only thing that failed was ther kerneal sync clock whatever that is.. and its sitting at the login screen... and like the usb drive i was trying to install my pfsense says my rpi4 is only using like 300 bits/s to the pi so its not installing or anything.. whats going on? is the pi broken? and i dont have a keyboard on the pi... but ya browser still stuck on it should only take up to 20 min
apparently your pi has trouble reaching the internet
hmm its dns is setup to 1.1.1.1 it was working but i figure might try installing on a different pi and then just shove the card in... maybe ill plug a computer on the cable of the pi and see but internet should be working
but ill trouble shoot some more
you can try a usb ethernet adapter if your ethernet port is suspected defect
I am installing the core on a pi, right when running hass i got ImportError: cannot import name 'contextfilter' from 'jinja2' (/srv/homeassistant/lib/python3.7/site-packages/jinja2/__init__.py)
Core is really not a recommended install method and is meant for experts. You're better off with HAOS or Container
container4lyfe
What would be a good path for the config?
/opt/homeassistant or something is a good place for those kinds of things if you like to follow Linux filesystem conventions, but it really doesn't matter
Okay, thank you!
i couldnt install HA on my 1 raspberry pi wanted to re install it.. couple hours didnt work.. now i trying to install it on a working one.. and its also been stuck for an hour where it says takes 20 min.. at webpage that says it will only take up to 20 min.. is it needing internet to setup or it is just unpacking files? if its needing internet.. is there a way to test if the server it gets its files is down?
god i just updated docker from 17.x.x to 20.x.x lol
why in the creation of earth did i put a 16 gig sd into my pi ๐
the god damn thing errored after 20 minutes of pulling because i have no storage lmfao
im planning to switch my home assistant install from my current docker install to a dedicated vm running home assistant OS. anything i should know before i start?
was just going to do a backup from the home assistant settings, will that be enough?
Yes just restore the backup and you're good to go
Sometimes if you have a large backup then the web uploader can fail, but you can always go through the initial setup, install the Samba add-on, upload the backup, and then restore that way
what vm software are you using and which OS?
proxmox
thanks
I only ask bc i used Virtual box under windows and that was hell (incredibly slow) so i didnt want you to have to go through that ๐
Switched to ubuntu server and it works incredibly well
thanks for the warning
i am having a tone of issues with my docker VM (not just HA but that is the main thing) so i want to get it on its own VM so it is more robust
I heard ppl love using proxmox hope that works well for ya
and i figured if im doing that i might as well use HAOS so i can use add ons
i keep my plex stuff in a separate VM for the same reason
is there anything you loose going from docker to VM?
Im not sure if you look on the HA website youre gaining not losing
what issues?
also i'm guessing you know this already, but add-ons are just pieces of software you could install standalone and connect to HA
You give up the flexibility of installing whatever you want for the convenience of addons which are essentially Docker containers with some additional magic to integrate more deeply with HA
basicaly the whole system crashes. it is something on that VM because i left the VM off for a week and had no crashes where it was crashing at least once a day befor
and by the whole system i mean i had to go hold the power button on the proxmox box
ok thats fine, i can just install external things on another VM if i need to
That's basically my approach to HA. I run HAOS because it just works, does it own thing, makes all its own backups for me, and if I want to do unsupported things I just install them on other machines
that's... interesting. what OS?
the VM? Ubuntu 20.04.1 LTS
my plex VM that dosnt seam to have issues is also Ubuntu but 20.04.0 LTS
honestly i just picked at random, never really worked out the pros and cons of each
I've been running Ubuntu Server for years on a bunch of personal machines and a few thousand at work, and it has always been rock solid
oh, my proxmox is also quite out of date
gonna need to fix that before i can install the VM
I currently run HA in python env. however i'd like to find a way to keep it HA automatically updated. are there any options for that?
Core (python venvs) is not a recommended install method. You should really consider Container or HAOS. Regardless though, automatic updates of HA is not a good idea because of breaking changes
Hello, it's me again. So to summarise; I was troubleshooting yesterday for my RPi400 as it couldn't get eth0 enabled no matter what I tried, common error being "Connection 'Supervisor eth0 is not available' on device eth0 because device has no carrier". I've now tried to connect the RPi to my Mac with ethernet to see if I could get them to connect but alas, no dice.
With all of my testing now done, excluding buying an USB to ethernet adapter, would it be correct to assume that the RPi itself has a faulty ethernet port?
It does though work just fine on RPi OS.
OK, not ruling out that though as I tend to have bad luck with HW. ๐
That was my thought as well.
But an annoying problem to solve though. Seems like a few people have this issue as well but there's no common nominator as there's not enough data. My current suspicion is that my Unifi Dream Machine having some issues. I've also flashed HA to the SD now a few times and also tried from flashdrive in order to rule out broken bootable media or corrupted image.
Is the image for the 400 ? Not sure if that works
I tried with Raspberry Pi imager and chose image for RPi 400.
Just checked and image for RPi4 and RPi400 is the same, according to the RPi imager.
Raspberry has their own tool to flash an OS to SD card, which I used. https://www.raspberrypi.com/news/raspberry-pi-imager-imaging-utility/
Anyways, you want to get rid of that unifi ๐ฉ
Maybe so but that's for another time. ๐ But yeah, I'm getting stuck to the Home Assistant Command Line and can't get IP for the RPi400, no matter how I try. Basically same as this thread, which I also replied to. https://community.home-assistant.io/t/network-problem-on-raspberry-pi-4/524828
Every day thereโs about 20-50 people coming on this discord with either: IP not working or DNS not working. Every damn time.
And it is always HaOS
Iโm not sure but yeah its totally possible that theyโre doing something nasty, just go with a container install. Way less headache
That's unfortunate as I would rather have RPi running it due to low power consumption. And now that I finally, after 1,5 years of waiting, managed to grab one..
If I can't get this to work I might have to consider that. More to learn then as Docker is something I've yet to use..
You would be up and running a long time ago if you focused on docker alone
Heh, probably true but wanted to rule out hw failure and to learn what's wrong and how to fix it, as RPi would still be my preferred option.
You bought a keyboard with a computer attached my dude
Yeah, so?
You waited 1,5 years for that?
Not that specifially but overall, for an RPi
To cut it short: I've been interested to run HA and would have liked to run it on separate device, where RPi comes into play. RPi's have been under the rock since the pandemic and managed to find one now, though not exactly my preferred option but good enough, hence RPi400.
no problems with a container install. i promise
Unifi stuff works
You could've gotten a thin client and used that for HA, power usage is basically nothing between them
Yes, I'm still on my return window if need be. I was more so focused on RPi as I've been interested on it for a while now
It is very much a toy/ education tool not super great for production
thanks Jumpers, yeah i know its not a recommended way. i installed it like that a long time ago and just kept going since then
Can I only flash the OS on Pi 4 an 3?
what else do you want?
last time i tried rpi1 setup it took almost 3 hours
that thing a potato lmfao
Hi, I recently setup duck dns but I ran into a problem, the ip in the duckdns website changes to my pc ip for no reason whereas it whould be my internet ip. Is there a solution to it?
please do not cross post
I was deleting it from that section, cause this is a installation query I believe
someone of the 11k people will eventually respond when they have time and a similar setup
i cant seem to install HA is the file server down or something? i did a clean install on the sd card.. and i left it over night and its still stuck on 20 min.. is it a glitch in the os its not setting up... or does it connect to a website to setup files and if so is it down?
i cant seem to install HA is the file server down or something?
have you tried restarting it
ya i have both just sit on the prepairing home assistant this can take up to 20 min
ive tried different browsers too nothing
i just posted on the community forum maybe someone seen same issue.. try to expand my help but i dont understand
cuz internet works at the network cable.
@spare heron I converted your message into a file since it's above 15 lines :+1:
the whole pi died after that ๐
Oof
from where are you downloading it
oh wait, that doesn't seem like a download issue at all
no where i mean is the os downloading when it sits at the 20 min or is maybe something corrupt in the os
i have tried both etcher and raspi installer to install HA
then i boot up and i goto the website and it then says the takes up to 20 min
if i have a screen plugged into the pi i get to the login screen
and thats it it never sets up
so i wanted to know if its not connecting to a website to download or maybe something corrupt in the os .. this happening on 2 pis now i tried
there seems to be issue in the dockerhub in some of my cloud VM, the images are not pulling for dockerhub
Why do you have ha on a cloud VM
check the logs?
how you check the logs at console?
on the ha cli try "su logs"
it says "/bin/ash: su: not found"
im guessing because ha hasnt set itself up yet?
That sounds like you're at a shell rather than the ha> prompt. In that case it would be ha su logs
lot of errors can ill see if i can screen caputre with my cell phone.. on the HA community page i posted what i got when i typed "ha supervisor update" gbcr.io/home-assistant/aarch64-hassio-supervisor:2023.01.1: 500 server error for http+docker//localhost/v1.41/images/create?tag=2023.01.1&fromimage=gbcr.io%2fhome-assistant%2faarch64-hassio-supervsior&platform=linux%2Farm64: internal server error (โget โhttps://gbcr.io/v2/โ: x509: cerficate has expired or is not yet valid: current time 2022-07-14100:45:002 is before 2022-07-i9t00:00:002โ) ill try to get a pic with my cell
Your system clock is wrong
ah wont let me post pic... i know a week ago HA reported a critical error.. and the link they offered on ha website said try these things didnt fix it.. now i did 2 new installs.. of ha neither has gotten past the 20 min preparing... so reason i figured maybe corrupt thing or it cant get to website it needs files for that maybe its down
Seems like something is preventing it from reaching an NTP server to update its clock
here is the link to the HA community i asked in there.. the pics i couldnt post here are on the bottom of the page
how do you force HA to use 192.168.0.1 as the ntp server then
that way it forces ha to continue the setup
or how do i ping the ntp server to see its working
from the console
but i dunno why does time block HA from being able to installed
The underlying NTP settings (as with the rest of the OS) aren't meant to be changed. You can see the NTP status with timedatectl timesync-status
Because SSL requires a working and correct clock to function. Without the right time set, it can't download anything over HTTPS
here i thought orginally when you download HA for raspberry pi the file is downloaded and when you burn it to the SD card the os there and it just sets up locally that it doesnt need internet.. least thats orginally how i thought it worked... so do i type timedatectl timesync-status at the console?
and can i ping the npt from console to see if its reaching
so i tried the timedetectl and timedetectl timesync-status and since i cant past image easilly its on the bottom of my link https://community.home-assistant.io/t/i-cant-seem-to-install-ha-is-the-file-server-down-or-something/526773/16
i tried to nano /etc/systemd/timesyncd.conf but there is no nano as i was gonna get it to point to 192.168.0.1
i did find a google . says make a file in the boot location on the sd card a timesyncd.conf i did [Time] NTP=192.168.0.1 so hopefully that gets me started
Like I said, the OS isn't meant to be modified. It seems like you have some sort of networking/firewall issue that needs to be resolved
it started 1 week ago
like i can ping time.cloudflare.com now.. but earlier i couldnt...
least on the computer havent re tried on the pi
guess that didnt work... i did cat timesysncd.conf on the pi and it is [Time] FallbackNTP=time.cloudflare.com #speed up-boot as first attemp is done before network is up connectionretrysec=10
so i dont get what to check look for or test
so how do i force to sync.. what other options are there
Hello !I am currently struggling to run the docker ha on my raspberry pi B. I get the message "connection refused" either on my windows browser (http://RASPBERRY_IP:8123) and from my ssh connection to the pi (wget -p localhost:8123). I have a pi-hole running on my PI which is plugged to my internet box. What can I do ?
I can access the pihole web interface properly from my windows computer
Also it seems that I don't have any .yaml configuration file in my docker HA directory
I used this command :
docker run -d
--name homeassistant
--privileged
--restart=unless-stopped
-e TZ=Europe/London
-v /home/pi/ha:/config
--network=host
ghcr.io/home-assistant/home-assistant:stable
Iโve been running HA through Docker for many years now, is there any benefit to โupgradeโ to HA OS on say a standalone RPi?
It's not going to be an upgrade if you've been running it through Docker and you're comfortable w/ that. All you'd get extra are add-ons, but that's identical to you creating other Docker containers yourself. If anything, it may be a downgrade depending on the resources of the RPi.
What's in the log
Great! Iโd love to have the add-ons inside of HA but then it doesnโt sound worth it since ill have more or less the same thing as now. Iโm also not going to get rid of my docker system since it runs a bunch of other stuff as well
All the add-ons are are Docker containers that are managed for you... less flexible (but "easier to maintain") than DIY Docker
Again, depending on the RPi, you may run yourself into resource issues w/ add-ons
RPis are fine for starter systems or things that aren't too active, but not great for anything else
I was looking into HA yellow but it doesnโt seem like id get anything more than i already have
I don't find any errors in the system log nor the boot log. If you meant the HA logs I don't find them...
Just like docker logs homeassistant
s6-rc: info: service s6rc-oneshot-runner: starting
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
[17:19:01] INFO: Home Assistant Core finish process exit code 256
[17:19:01] INFO: Home Assistant Core finish process received signal 4
Normally if I ping localhost:8123 from my pi it should respond ?
pi@raspberrypi:/var/log $ wget -p localhost:8123
--2023-01-29 17:25:25-- http://localhost:8123/
Resolving localhost (localhost)... ::1, 127.0.0.1
Connecting to localhost (localhost)|::1|:8123... failed: Connection refused.
Connecting to localhost (localhost)|127.0.0.1|:8123... failed: Connection refused.
On port 80 (my pi-hole webserver I guess) it says "connected"
Hello everyone !
I have a quick question. I just did a fresh install of Home Assistant OS 2023.1.7 on a NUC. And all the add-ons (SCS, Node-RED, File Editor) keeps restarting randomly every few seconds (30-60 seconds)
Can't find any resources online about this issue. Any recommendations?
After installation, how can I connect my Pi4, to internet using WiFi instead of ethernet?
WiFi really isn't recommended, but if you're on HAOS you can set an IP via the network settings page
Open your Home Assistant instance and manage your systems network configuration
Which means I was to use URL during installation?
START UP YOUR RASPBERRY PI
Insert the boot media (SD card) you just created.
Attach an Ethernet cable for network.(missing this at my place how can I go about it??)
Attach the power cable.
In the browser of your Desktop system, within a few minutes you will be able to reach your new Home Assistant on
oh, if you don't have a way to connect to ethernet at all, you'll have to connect a monitor and keyboard and change the settings via the CLI
If you're at the ha> prompt, then network update will give you the relevant arguments
does anyone know how i can install an ssl certificate that i generate from my internal certificate authority
Just upload it to the /config folder and configure the HTTP integration to use it
https://www.home-assistant.io/integrations/http/
Some troubleshooting, cough 11 hours later but including life and other stuff, I think I found the issue with RPi 400 and connection. It seems like there's a bug and either the ethernet driver is missing or not loaded upon first boot for the onboard ethernet port. Once I changed to USB to ethernet adapter I got network online. I do have some modifications done to the system now so I'll have to double check if the issue comes back with vanilla HA OS 9.5.
I'll try to test this further tomorrow and post to github about the issue, if I can verify it. But in short: if someone has an issue with RPi400 (possibly 4 as well) not connecting to internet at all during boot to Home Assistant Command Line (CLI) but it works in Raspbian: more than likely your install might not have loaded ethernet drivers or they're missing, try using USB to ethernet adapter, instead of your network having issues (at least in my case). That worked for me.
Hello... I am trying to install HA fresh on a RPI4. I used the HA OS image and followed instruction but the install is stuck on preparing Home Assistant, should take 20 minutes screen. It's been stuck there since yesterday. Apparently it has something to do the time on the RPI. So my question is, how can I adjust the time on the RPI ? I don't think SSH is enabled by default. And even iit was, which user would I use as a login since I can't get to the UI to create one... Thanks !
HI,
After updating to HAOS 9.5 the website:8123 no longer works, the observer goes to port 4357 and shows:
Supervisor: Connected
Supported: Supported
Healthy: Healthy
it run on RPi4...
A login via ssh is possible, but what can I do there? apparently there are not m,al log files under /var/log/ ...???
What can I do to start the environment again, even a reboot from the RPi didn't help.
I had similar today, it had been up and running for a while after upgrading HAOS. I managed to get it working again by running ha core check from the terminal. Some bits were still a bit wonky after that, but the web interface was back up and running, which allowed me to do a restart from the Developer menu. Everything seems to be working correctly now.
How can I do that?
when you ssh in do you get the "ha" prompt?
ok, i fou dit
[core-ssh ~]$ ha core check
took a while to run on mine
Wasn't due to HA, it was an error on my part, I had created a yaml file that was empty, he probably didn't like that.
So now I know what to do with the ssh login... I'll have to read up a bit.
I don't have any .yaml file in my "ha" folder, is it normal ?
I juste have a "core" file
I used docker
docker run -d
--name homeassistant
--privileged
--restart=unless-stopped
-e TZ=Europe/London
-v /home/pi/ha:/config
--network=host
ghcr.io/home-assistant/home-assistant:stable
And it is a file
That's so weird
I tried this on my WSL2, and it creates more files and folders
How do i create the ssl certificate becuase when i sign the cert request it will be in cer format
and the key will be .key
so if i change somethning in config.yaml do i have to do anything? or does it find it automatically?
You'll need to reload that integration or restart HA. But do a config check before you do either of those.
- HAOS
ha core check - Container uses
dockercommands - Core requires you to activate the venv first
switch:
- platform: remote_rpi_gpio
host: 192.168.40.51
ports:
11: Doorbell Button
12: Voip
would that work for a remote gpio raspberry pi?
I finally figured it out! Got hass-os working on macos with qemu, so much faster than the s**** virtualbox.
MARIADB: ERROR: Got unexpected response from the API: There is already a MySQL service in use from core_mariadb
I am trying to solve the problem above, I have found a few references in Github and Discord to this issue...but I can never find the resolution.
Does anyone here know how to fix it?
(Note: HAOS running on Synology Virtual Machine Manager)
Sounds like you wanted #add-ons-archived for getting help with getting the MariaDB add-on running
trying to setup a VM with home assistant to port from a container and it has been stuck on the restore for over half an hour, any ideas?
It's often recommended that you do the standard onboarding and then restore, that way is more likely to work
thanks will do
ok how do i restore once i have gone through the standard setup?
thanks
Hello everyone,
I have home assistant installed on my SD card for my Odroid N2+ .
With the latest OS update to 9.5 the SD card failed and now im looking into installing HA on a SSD.
I have a USB ready SATA SSD but the Odroid wont seem to boot directly from the SSD.
What are possible solutions to this, if any.
I've looked online but couldnt find anything helpful related to this issue
Has anyone set up a vlan where devices with access to WiFi are isolated on a separate network with no access to the internet and only home assistant can control them?
yes
Yep
so my homeassistant keeps crashing, third time today
and the only thing I can do is restart the pi, I already restored a backup on a new sd card but it keeps crashing
it is a very new install with only 4 addons and HACS addes
I am quite good with linux (been using it for more then 20 years) but without access to any of the systems logs (usually in /var/log/*) its quite hard to find the problem
the power supply is a standard original raspberry pi psu and its running on a pi4
anybody any idea where to begin?
according to the about:
Home Assistant 2023.1.7
Supervisor 2023.01.1
Operating System 9.5
Frontend 20230110.0 - latest
wow why is there so much dns lookups
By design, apparently
52 in one minute?
Jan 30 17:07:15 homeassistant hassio_dns[452]: [INFO] 172.30.32.1:41654 - 23401 "PTR IN 164.178.168.192.in-addr.arpa. udp 46 false 512" NXDOMAIN qr,rd,ra 46 0.032689607s
There's a few forum threads, and issues, about it
status: wontfix
Upgrade your DNS server with go-faster stripes
hmmz I am using the dns server of my internet provider
With your Linux experience, HA Container might be a more desirable option for you
reading about it, is the addon store still available with that?
no. But if you're using a container for HA, you can just do the same for whatever services you want
Hello, i'm having an issue with my install showing as unhealthy. It seems to be related to "Home Assistant WebSocket API". What is this? I'm not clear what it is used for and how i can resolve the issue. It's blocking any updates due to the error.
dunno how to clear your block, but here's docs on what the websocket api is for
tl;dr -- it's an API endpoint so stuff can talk to your home assistant install using websockets for a constant data stream instead of one-shot POST calls
What is your install type
HAOS on rpi 4 with SSD
Traceback (most recent call last):
File "/usr/local/lib/python3.10/site-packages/aiohttp/web_protocol.py", line 334, in data_received
messages, upgraded, tail = self._request_parser.feed_data(data)
File "aiohttp/_http_parser.pyx", line 551, in aiohttp._http_parser.HttpParser.feed_data
aiohttp.http_exceptions.BadStatusLine: 400, message="Bad status line 'Invalid method encountered'"
2023-01-26 23:53:28.075 ERROR (MainThread) [frontend.js.latest.202301040] :0:0 ResizeObserver loop completed with undelivered notifications.
2023-01-26 23:54:38.772 ERROR (MainThread) [homeassistant.components.websocket_api.http.connection] [547807341024] Error updating Home Assistant Core 'HomeAssistantCore.update' blocked from execution, system is not healthy - setup
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/components/hassio/update.py", line 303, in async_install
await async_update_core(self.hass, version=version, backup=backup)
File "/usr/src/homeassistant/homeassistant/components/hassio/handler.py", line 51, in _wrapper
raise HassioAPIError(data["message"])
homeassistant.components.hassio.handler.HassioAPIError: 'HomeAssistantCore.update' blocked from execution, system is not healthy - setup
Does the secrets.yaml support jinja?
You can put Jinja in them, but it's just a text substitution
So the part that matters is that where you use the secret supports templates
Great! Could you share any resources on how to go about it?
So the wifi devices have zero access to the outside internet?
Are you able to access Home Assistant from the web, outside your home?
Hello ๐ Just getting started with Home Assistant and Iโm currently stuck at the preparing Home Assistant screen for about 30 minutes now. Any suggestions?
Click the blue dot
Installation
Yes, the iot vlan has its own ssid and no access to the internet. Ha has access to the iot vlan. And ha is reachable from the internet via a proxy in DMZ.
onemarcfifty has a lot of good videos if you want to learn more about setting up the vlans with wifi
https://www.youtube.com/@OneMarcFifty/videos
Hello everyone, I'm trying to install Home Assistant OS on a raspberry pi 4. I am not able to get the web page to open. Looking at the display output i see that it give the message [WARN] Home Assistant CLI not starting! Jump into emergency console... Also the kernel network time sync seems to fail. Could this be causing issues? Do you guys have any ideas?
Can anyone help I can't input to domains field at addon Let's Encrypt
ok moved
Weird error. Fresh install using the guide but get this error: Server Error ("Get "https://ghcr.io/v2/": x509: certificate has expired or is not yet valid: current time 2022-07-13T12:18:40Z is before 2022-07-19T00:00:00Z")
Seems like it's not getting the correct network time
Im stuck at the same issue ๐ฆ
Connect to the pi with a keyboard and monitor and boot the Pi
I did. get to the cli when doing that, but cant seem to be able to set the date with any ha command
I actually solved it now. Don't know why it worked but installing raspberry OS first and then installing home assistant seemed to have done the trick
Hey, I was recently setting up duckdns. But there's something which is blocking me to work it out. My ISP provides 2 public ip's, I want duckdns to access my ip which shows on my router, is it possible to do so
I have configured my router to show ip on ha
When you say want duckdns to access my ip are you using the DuckDNS add-on, or integration?
(or something else to update the IP)
Add on
Then #add-ons-archived may know what's possible there, but the other option is to disable that and manually update it
My ip changes everyday so manually wont be possible
But ill go in the add on and ask if something helps
#/settings/websocket
Hey all I've spent days now and cannot seem to find a way to enable websocket support using the NGINX add-on. Have a Shelly HT Plus and the device connects to HA, but with no entities. Tried with wss and ws, with IP, and hostname. The only thing that seems to connect is ws with hostname. ( NGINX is being used for SSL and that works fine. Based on all my reading web socket support needs to be added to nginx. I can't seem to figure out how to do that using the add-on version though. I think perhaps a custom file of some sort...
I do get this when adding shelly device. which seems to confirm websocket issue. 2023/01/31 14:28:30 [error] 134#134: *296 connect() failed (111: Connection refused) while connecting to upstream, client: 192.0.1.46, server: ha.sXXXX.com, request: "GET /api/websocket HTTP/1.1", upstream: "http://172.30.32.1:8123/api/websocket", host: "ha.sXXXX.com" #add-ons-archived
Anyone else using portainer to manage their container install?
I tried stopping my stack, and redeploying with the pull latest image option selected. But it still just stays on 2023.1.7, not 2023.2
Ahh looks like I do have the stable tag
So im guessing 2023.2 isnt marked as stable yet
it's in beta
tbh homeassistant is the only image I know not using :latest for the latest stable
Also portainer has a habit of breaking stuff...
hey guys so im setting up home assistant on on my windows 10 pc using virtual box every time i start it gets hung up on this screen and i cannot connect to http://homeassistant.local:8123/
Hung up on what screen
I am running odroid with home assistant OS, how to install from commandline anything in the operating system (no access to apt-get)? I have problem with date/timeand i would like to have ntp installed so time is synchronized over internet
HAOS already uses NTP. If the clock isn't syncing, you likely have some sort of networking issue
thats strange because it has access to the intern
no problem with getting the sensor data from tuya etc
no firewall that is brutal and cuts out everything
And it's an embedded os. There's no package manager
It uses time.cloudflare.com if that helps you track down the issue
There's a Chrony addon you could use
but chrony is a server from what i read so if my local time is bad than it will not help right?
Oh I didn't realize Chrony had an option to update the system clock. I see it in the docs though
is there a way to call an entity in an add-on config?
What do you mean?
Few add-ons interact with HA's entities (I'd say none, but I'm sure there's going to be one that uses HA's API or MQTT)
I want to call an entity to tell my ip in duckdns because I have 2 public ip's
#add-ons-archived can tell you if that DuckDNS add-on supports it - but the docs for it will tell you
I'd expect the answer is no
https://github.com/home-assistant/addons/blob/master/duckdns/DOCS.md#option-ipv4-optional is probably what you need
๐
this shows the ip which I dont want to show
i want to show ip which is shown in my router
have configured the entity for it but how do I call it in the config is what I dont know
You need something that shows you that IP in a URL
Most hacky solution would be to scrape it from the router (https://www.home-assistant.io/integrations/scrape), write it to a file under /config/www/ (https://www.home-assistant.io/integrations/file#notifications), and then use that path (https://www.home-assistant.io/integrations/http#hosting-files) in the add-on config
I think they are behind CGNAT so as I pointed out to them in add ons they're not going to have it work
Running into some weird things when trying to install home assistant OS on a raspberry pie 4. It gets stuck in the install with the following error. Server Error ("Get "https://ghcr.io/v2/": x509: certificate has expired or is not yet valid: current time 2022-07-13T11:39:14Z is before 2022-07-19T00:00:00Z") seems like that it doesn't pick up the correct time on install, and I can't find any options for that when flashing the sd card?
hi im from Iran i can't update my home assistant server because internet problem with my government i have a VPN on my phone but i can't share it with my server anyone knows a way i share my VPN with my server (i use napsternetv with a config file to access free internet)
internet sharing usually works around vpns running on mobile clients, which leaves you with the current situation you are in
you'd need a vpn client on haos or on your router
Is there any way you can set system time from the console?
you can try timedatectl
it fails during headless setup, and I try to figure out how I can set the time for it to actually see the certs as valid
it not being able to fetch the current time is a sign of no internet connection, manually adjusting the time will not get you further in that case
Its connected with ethernet to a managed ubiquiti switch and since I can visit homeassistant.local the connection seems fine
a working homeassistant.local is no confirmation of a working internet connection
Ah, you're one of those.
What? They're somebody who applies logic?
Can't read Supervisor data: System is not ready with state: setup
Error during setup of component hassio
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/setup.py", line 253, in _async_setup_component
result = await task
File "/usr/src/homeassistant/homeassistant/components/hassio/init.py", line 587, in async_setup
await repairs.setup()
File "/usr/src/homeassistant/homeassistant/components/hassio/repairs.py", line 149, in setup
await self.update()
File "/usr/src/homeassistant/homeassistant/components/hassio/repairs.py", line 157, in update
data = await self._client.get_resolution_info()
File "/usr/src/homeassistant/homeassistant/components/hassio/handler.py", line 51, in _wrapper
raise HassioAPIError(data["message"])
homeassistant.components.hassio.handler.HassioAPIError: System is not ready with state: setup
No, the type of (usually) knowledgable person who sits in different irc/discord/ channels, enjoying the fact that they know more than the person asking for help
You know, if you don't want help there's easier ways of going around it
Being an ass to people trying to help you is a great approach to it though
Neither is being an ass to people asking for help....
I think you're confused
a working homeassistant.local is no confirmation of a working internet connection
is not being an ass
It's a statement of fact
you probably need to check your dns settings, your homeassistant device does not appear to reach the internet as said
what does it/the logs say?
This page is no longer available. It has either expired, been removed by its creator, or removed by one of the Pastebin staff.
Please use a code share site to share code or logs, for example:
- http://pastie.org/ (select YAML for the language)
- https://dpaste.org/ (select YAML for the language)
- https://paste.debian.net/ (you guessed it, select YAML as the language)
Please don't use Pastebin, since it can randomly add spaces to the main view. Please also don't share text as images since it makes it harder for people to help you. Remember that others may have colour blindness, impaired vision, etc.
Hi, I'm having issues installing this specific orjson-3.8.1.tar.gz, when trying to do home assistant core install, homeassistant==2023.1.7
ERROR: Could not build wheels for maturin, which is required to install pyproject.toml-based projects
Any reason you're using a Core install and not Container (Docker)?
This is an advanced installation process, and some steps might differ on your system. Considering the nature of this installation type, we assume you can handle subtle differences between this document and the system configuration you are using. When in doubt, please consider one of the other installation methods, as they might be a better fit instead.
(I'm going to guess that the full error tells you that you're missing something, probably cargo)
Hi, I encountered this issue:
Home Assistant observer
Supervisor: Connected
Supported: Unsupported
Healthy: Unhealthy
Do you have an idea how to debug and fix this?
Is that a Home Assistant OS install, or a Supervised install?
Sorry if I understand wrong but I see this: Home Assistant OS 9.5
Home Assistant OS then
That eliminates most of the common errors, unless you installed the Portainer add-on
I see a sticker with Click to enable ping the original author
@paper patio When using Discord's Reply feature it defaults to pinging the person you reply to, which can get frustrating for the target. Use Shift + click on the Reply option, or click @ ON to @ OFF to stop this - on the right side of the compose bar.
You have to change this every time (thank the Discord devs for that).
And ... if you pay attention... it tells you to Click to disable
Well failed though...

I dont know this function... I clicked it multiple times. Dont know if it is enabled or disabled now.
Anyways, I dont know what "eliminates most of the common errors, unless you installed the Portainer add-on"
No
Anyone have good install instructions to install home assistant core on Mac m1?
Mac instructs on website are not directed for macOS.
https://www.home-assistant.io/installation/macos#install-home-assistant-core
Core isn't a recommended install method. You should probably use a VM
With Core it assumes you can translate the instructions to your OS
(The instructions are also the same, even on Windows, which is ... amusing)
Okay I donโt want to go down the virtualbox route. What is the KVM way, that didnโt make sense to me
On a Mac M1 you'll need something like UTM instead
So, any other suggestions maybe? Since I dont have Portainer installed?
Yeah... stop tagging me in replies ๐
Okay, I tried that now. So to get Bluetooth to work have to do proxy or usb plug?
Haha ok if that helps...
onboard bluetooth on a mac mini is likely pci so you might not be able to pass that through. also macos will not let you take it over I assume, you are better off with a usb bluetooth dongle
you should probably go the linux vm + container install, I doubt the aarch64 image will boot in utm
The HA UI should tell you why it's unsupported/unhealthy - no idea whether the Observer does
what about my issue that HASSIO decided to stop working?
is there anything i can do to troubleshoot it somehow
You mean the logs? Any specific system log I should check? F.e. Home Assistant Core?
Nah, in the UI, but somebody who uses HAOS will have to guide you if you can't find it
Ah I thought you meant the logs in HA UI (in path /config/logs). Ok maybe someone who uses HAOS can help me?
Btw, do you know if there is a difference between HA Observer and HA Supervisor?
In the UI I found some system information which differs a bit from the HA Observer:
Home Assistant Supervisor
Host Operating System Home Assistant OS 9.5
Update Channel stable
Supervisor Version supervisor-2023.01.1
Agent Version 1.4.1
Docker Version 20.10.22
Healthy true
Supported true
@paper patio the observer can tell you if the supervisor is running
you use :4357 instead of :8123
@frosty path your logs show a lot of problems with custom integrations. You should start by disabling those just to filter them out
Yes I know but I wasn't sure about the information there. It says
Supervisor: Connected
Supported: Unsupported
Healthy: Unhealthy
While in the HA Supervisor in the UI it says:
Healthy true
Supported true
rename the custom_components directory, no_custom_components for example
and then restart HA
@paper patio reboot the system, then check logs again (yes, config/home-assistant.log file)
hey everyone. i just tried too install home assistant on a raspberry pi 4 i have it booting the os on a usb stick. it has worked on othe projects when i turn it on the green ACT led on the pi starts blinking once per second wich means that the usb stick is working properly. it didnt show up in my network scanner even thogh my ethernet cable green light was flashing the pi got hot so i thoght it was doing its own thing so i waited A DAY and it still has not showed up in my mobile network scanner and i cant connect homeassistant.local:8123 i tried to refflash the image but it still has the same issue. anyone got any idea on why this is happening. i really want too get this too work so i can control my new homemade blinds (:
it pings every ip in the network. and yes it is plugged in via ethernet directly too the router
why not use your router to see connected devices instead of the "scanner"
Booting from USB isn't recommended for the pi
https://www.home-assistant.io/faq/usb_boot/
oh, I just noticed that. I am blind. ๐ ty jumbers.
well the thing is i am 15 and its my dads router. i dont know if he would want me snooping in there and i read that a network scanner should work fine. but shouldnt it technically work? all of my other projects have worked
with a usb stick
using a router to view the network is a lot more normal than pinging your entire network
If your device doesn't support mDNS, you won't be able to use the magical homeassistant.local url. Are you on windows?
yes im on windows
okay, that isn't your issue then
because i am pretty sure green flashing light means usb i ok
It's going to be hard to run a server in your home without having router access to at least observe what's happening. Maybe that would be ok if you ask. A USB stick is going to be very slow, and likely become very hot
And USB sticks tend to have a shorter life than SD cards
And yeah it will be a "works for now" kind of experience
and my ethernet cable flashesh from time too time. but if i look at my pc internet cable it flashesh alot more. my ethernet cable flashesh once in maybe 8 seconds
but if i flash somthing else like raspberry pi os it works
if i wanted too try wireless connection too see if that works better how would i do that
That is highly unsuggested as well. Do you not have an SD card for your pi?
thats the thing i only have one sd card and that is in my MagicMirror project
do you really think that is the problem? my usb stick is pretty fast
Alright, well the only other thing I can suggest is hooking a monitor up to the pi to see what is going on. You need to replace the USB drive or else it's sort a waste of everyone's time to try to continue to debug.
yea ok sorry. i will try too debug with the monitor soon
no problem at all, really. You are way more likely to get help from others when you're following the suggested or known paths and I don't want you to make it harder on yourself.
Hi, I'm about to submit a bug of eth0 on RPi 400 not having drivers or not loading on cli. I'm not 100% sure on which repo to submit the bug though. Is it an issue with Supervisor, cli or what?
Operating System
hi guys, i used to have home assistant on docker and now i decided to install the supervised version to be able to use repositories addons and all that cool stuff, the problem is that im getting pretty high cpu usage and the server is not reachable.
ha banner returns all the correct information and the supervisor works
Hopefully you're using Home Assistant OS, and not HA Supervised?
i am using supervised
ok but i also have other stuff on that machine, i currently have debian
Other stuff... that means you ignored the requirements already
This won't end well
Thanks, bug submitted. Hopefully this is fixed soonish as I'm now using 100 Mbps adapter and would prefer to get the full 1000 Mbps from RPi. ๐ https://github.com/home-assistant/operating-system/issues/2338
do you think it is better to just stick with the docker image in this situation ?
container, yes
alright, thank you guys
well i just wanted to try addons and repositories
Running Supervised is like using a blow torch to inspect the inside of a petrol tanker
addons are just containers with more blindspots
If by repositories you mean custom components (eg HACS) then you get those already
If you want to run other software outside of add-ons it's the only supported way
(I mean, there is ||Core|| but unless you're a linux admin just don't)
im pretty good with linux
Docker
"Pretty good" is still a reason to run Docker
Can make Linux dance to my tune ... still use Docker
ok i connected a monitor and i took a picture but i dont seem to have permisions too upload pictures here. so when i boot up i get somthing saying " FAILED Failied too start network manager wait online but then it seems too fix itself by saying reached target network is online then it waits for somthing called kernel time syncronized. and that also seems too fix itself. then i get to a screen saying: "welcome too the home assistant command line. System information. ipv4 adresses for eth0: (blank) like its actually not anything there. "ipv4 adresses for wlan0: same here. home assistant URL "http://homeassistant.local:8123" i dont get why that is and also is there a way too send pictures in this channel. because i cant seem too do it on my phone atleast and thats where i have the pictures.
if you want more context scroll up too my other post
there
@spice cape 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.
You did plug the network cable in?
Yes ๐คฃ
I'm pretty sure anyway
Maybe the Ethernet cable is faulty
But it is glowing green and orange
I could try too change the cable sson
Hmm yes I will try that but shouldn't the cable need too work in order for the initial setup too work?
Maybe it shuts off the Ethernet after the setup or something
That seems to be the same issue I was having. If you have one you could try to use USB-A to Ethernet adapter in the meantime and see if that would fix the issue, it did for me. I've made a bug report of this to Home Assistant github. https://github.com/home-assistant/operating-system/issues/2338
One method to narrow down this issue is to boot the Raspberry to Raspbian and see if the ethernet port works there. If it does then it could be the same issue I was having, as per my previous reply.
ok i w8ill try too flash rasbian
When i unplugg and replug Ethernet while in home assistant command line i get absolutely no error message or anything saying that the Ethernet is disconnected nor when i plug it back in. Just too add. I will try flash raspbian
Here is an error
[failed]
Seems too be important
As usual it starts but I can't access the website and no ip is being displayed
Just checked; that's the same issue I was having so it seems to be related. If you didn't try to boot into Raspbian yet do that as well so that you can narrow down is the issue with your Raspberry or Home Assistant. If your ethernet works in Raspbian it's a software issue and most likely the same that I'm having. If it does not work in Raspbian nor in Home Assistant then it's either a bad cable (change that first and then try again in Raspbian. If it still doesn't work try different port from router or switch) or a fault in RPi itself.
Yes the cable works no problem how did you fix yours? I have a pi 3. Could that work fine with home assistant
I'd also recommend having more than one SD card in hand for testing, makes it a bit faster to switch between two systems. You can use USB flashdrive too but don't rely on them. Not sure do they wear out faster. Overall, I'd recommend having at least two SD cards, one for Raspbian and one for Home Assistant, and one USB flashdrive if you have to do some testing with commands. That way you can always have at least one clean installation.
OK, that's great so that means it's not hardware but rather a software issue. Unfortunately I didn't, as of yet. What I did in the meantime is to get USB to Ethernet adapter and use that, though I'm now limited to 10/100 instead of 10/100/1000 that RPi would provide.
Ok i think i have a adapter lying around that i could test otherwise I will test with pi 3
One method that you could try, though I've yet to test this and most likely won't as I have now a working connection via adapter, is to connect the RPi to Wi-Fi during initial setup and then check if the ethernet port could be re-enabled on HA. I've yet to go that far though, as I had no time to continue with setup.
How hard is it too do wireless setup
What do i have too change
do i have too reflash my os and then plugg in the adapter or could i just do it now and restart
So you have Home Assistant on the card now? I'd turn off RPi first then plug in the adapter and see if the connection works
Just to be sure
Ok then it could be that your adapter is not supported either..
Well something changed now when it is done booting i get a new system information ip but still no ip in it. It says ipv4 adresses for enp1s0u1u1: and then nothing here
You could try these steps to set up your RPi via Wifi https://community.home-assistant.io/t/guide-connecting-pi-with-home-assistant-os-to-wifi-or-other-networking-changes/98768
Do note that you're on your own with that as I've not done it nor can help you with troubleshooting
I was about to do that and start from step 2. as that was the part I found useful for my situation and it might've worked for me. But alas, I did not test it.
If you had any setup on your old Home Assistant setup and can loose it; flash your card and then try the wi-fi but again, only if you're comfortable with it.
Do note that it is recommended to not use Wi-Fi in the long term and I don't know if you can force the eth0 to come online from completed installation.
Got it so same as me. However if you do get the connection to work please post about it to the github issue I mentioned and to this post so that others who have the same issue have something to go by. https://community.home-assistant.io/t/network-problem-on-raspberry-pi-4/524828 and https://github.com/home-assistant/operating-system/issues/2338
The more information that can be provided the better as then there's more of a chance to get the issue fixed.
If you're willing to you can also try to use Docker and run Home Assistant from there but that's, again, something that I'm not familiar with it. You might be able to run Home Assistant through Docker from Raspbian though.. not sure.
Best of luck with the setup and please do post back if you can get your installation setup completely. I'd tinker and see if I could get eth0 working later, as in if setup could be completed and eth0 enabled once the first setup is complete, but don't really have time for that now.
I GOT IT TOO WORK. I just switched my USB to a ad card i managed too find flashed the os on that and then plugged the Ethernet cable into a Ethernet extender. (The thing where 1 Ethernet cable splits too 10 ) because before i was just plugging it straight too the router and now it works great!
o.O it sort of sounds like you were trying to use a modem/router combo that was in bridge mode. It sounds like you're now actually using a device that is routing.
yea as i said im 15 i have no idea about any of that. i just saw a hole and shoved it in
okay, use protection
Hi, i just followed the install steps for QNAP (Docker, ARM, homeassistant/armhf-homeassistant) from the getting started page.
after the container has finished creating and trying to start it, i always get an "exec /init: exec format error".
i dont know what that means and how to fix this. Any help would be appreciated ๐
It usually means you have the wrong image for the CPU in your computer
@humble mirage i have a ARM Cortex-A57 wich is a ARMv8-A Processor and i used the homeassistant/armhf-homeassistant with should support arm cpus
This ain't IRC, you don't have to @ me ๐
Norrmally you just pull ghcr.io/home-assistant/home-assistant:stable and let the system handle it
Did you try that?
yea some other error ill come back when i replicated it
Hello! I went to check on myHome Assistant instance (Raspberry Pi 4, installed from the image) and found that it wasn't working today.
Plugging in the HDMI, I saw a black screen with a blinking underscore.
I restarted the Pi, watched it go through the boot sequence. It got to the screen with "Waiting for Home Assistant CLI to be ready" text, and then the text flickered a few times before returning to the same blank-screen-blinky-underscore as before.
Any ideas on how to get it back up and running?
The most likely problems are power or failing SD card, and I'd start with power
Oh, fun! Okay, that sounds fixable.
Do you have anything connected to the USB ports? Are you powering it with a Pi power supply, or something else?
The SD card should be fairly new, and I think I remember disabling some logs to keep it happier.
New can still fail fast ๐
Caught me. I'm using a 4A phone power supply.
Yeah... that'll do it
I'm not sure I even have a dedicated 5V power supply.. but maybe I can find one if I go looking.
I assume a high-wattage, PD-enabled USB-C charger isn't going to work well, either?
@sterile palm I converted your message into a file since it's above 15 lines :+1:
@sterile palm When using Discord's Reply feature it defaults to pinging the person you reply to, which can get frustrating for the target. Use Shift + click on the Reply option, or click @ ON to @ OFF to stop this - on the right side of the compose bar.
You have to change this every time (thank the Discord devs for that).
Well... that looks like a really shit NAS for Docker
It... does not.. But also, aren't USB power supplies not supposed to supply more than 3A? And only 2A without negotiation?
They won't supply over 1A without negotiation (from memory)
I did not actually meant to ping you, lmao
It was plain text and it got converted!
I tried to hard to make it plain text just to get burned
that sure is a pretty background
Heh, sorry. I'm the opposite -- I need the pings to keep track of anything. ADHD drops conversations on the floor without it. ๐ญ
Yeah, it'd be nice of the Discord devs could allow the recipient to choose...
Hmmm... You can turn off pings, in different ways, but I'm not sure which setting would apply to replies specifically...
what does shit nas mean ?
no way to debug this ?
Yeah, the downside is that as a mod it's handy to be pinged for things like spam runs... so while you can turn them off server wide, you can't pick and choose
Yeah, that's why for servers I mod, I usually have people ping roles when they need help, or to post in a particular channel, instead of just pinging a mod by name directly. DMs work too.
But it's annoying that we have to sort that out for ourselves!
i just tryed to increase my memory allocation up to 6gb but its no difference
Okay, proper power supply on order, trying out a different (dumber?) phone power supply to see if that makes a difference. I really hope it's not the SD card, that would be... annoying.
... lol, I'm clueless... There's very clearly a yellow lightning bolt in the corner of the screen telling me I'm underpowering the Pi. xD
Time to go power supply hunting...
you can hold down shift while you hit the reply icon
Oooooooh... PoE Pi hat. ๐
https://www.microcenter.com/product/638336/raspberry-pi-poe-hat
is there a way to run esphome without hassio? I am just running core in a container.
ESPHome is a system to control your ESP8266/ESP32, with a native integration for Home Assistant.
You can find their documentation here, and get help with ESPHome itself here in the #diy-archived channel (#integrations-archived for the Home Assistant side integration with it). They also have their own Discord server too.
You can get help there โ๏ธ
cool, thanks
thats dude, the error is gone. now stuck on creating config im just gonna wait see what happens
Sigh. Well, the PoE hat works, but solving the power issues did not solve my problem. Time to check the SD card, I guess?
... I should also try it with the dedicated Pi power adapter I picked up, just in case.
Anybody have issues upgrading to 2023.2? I started it over an hour ago and itโs still spinning
hello. i'm having trouble getting ConBee II set up on a new Home Assistant install. i have tried Core and Docker install methods and both end up with the same error. logs only show:
2023-02-01 17:51:33.944 ERROR (MainThread) [aiohttp.server] Error handling request
the UI starts by saying "Loading next step for Zigbee Home Automation" and then after a moment says "Error"
i searched GitHub but didn't seem like i found anything to indicate a current know issue with ZHA or the ConBee. any ideas on what to try?
I did get it working on an Intel Mac with qemu. I believe qemu works on Mac m1 as well. UTM is based on qemu too
The hard part was getting a suitable ovmf for the uefi boot
Yeah I got it to work with UTM as well, but also have core working also.
With core, I can use the built in Bluetooth
Interesting. UTM wouldn't work for me, so I had to manually piece together a qemu command
This guide works: https://siytek.com/home-assistant-supervised-on-mac/
Wow. Soo long and it's setting up ha supervised. I ran HAOS directly on qemu
Anyone have issues updating their docker containers? I'm not 100% sure which version I was on before, but the log says updating the database schema from 30 to 33, and now I get the onboarding screen (but my automations are trying to run and are throwing errors for unknown entities)
this has happened twice to me recently, the last time I took it as an opportunity to reconfigure everything, but not this time
aannnd now it's throwing the error: RuntimeError: No Home Assistant provider found
Supervised is a trap
What is throwing the error?
Also how do you update your containers
I tried to go through the onboarding (to see if I could get into a halfway usable instance)
anyways, I update by pulling the latest version through portainer, then recreated the container so it would use that newly pulled image
I (thankfully) always rsync the files elsewhere first, so I was able to go back to the last image I had pulled on 2023-01-12 (I'm guessing 2023.1.2), and everything is fine with that
Alright, not sure what I did differently, but repeating what I believe is the same steps got me to 2023.2.0
It could well have been portainer stuffing it up
Possibly, I've had other odd issues happen with portainer before
I am running HA OS in a VM on ESXI. I've mounted my NFS share for storage and passed through to VM but HAOS is not recognizing the extra storage space.
I found some stuff online suggesting it's not actually possible to do this - is that true? If so, is it possible to mount an NFS share directly in HAOS?
Basically I'm trying to give Frigate plenty of storage to record onto without killing my little local 32gb HAOS drive.
(also not sure if this is the most appropriate channel, apologizes if not)
No supported way to mount a network share to HAOS
If you want that recommend swapping to a container based install
Ah, crap.
I actually just got the HAOS up and running because the docker container I was using didn't support add-ons ๐
I guess I can just expand the HAOS drive space
What is the most expansive installation method that supports the most things? Container?
container is the best in terms of flexibility and compatibility
Gotcha, I appreciate that info. I may try and spin up another VM and go that route just to see. Thanks ๐
Well poop, I can't expand the disk size either since it was imported from HA directly and not created by ESXI. Well I sure did waste a bunch of time doing this haha!
Would I be able to just direct attach a USB drive? Or does HAOS not recognize any other drive but itself?
How do people with HAOS run frigate and store recordings? Or something similar
Hmm. So people who run HAOS either don't use the provided VMDK or they just don't use HAOS to do frigate, etc.
Is it possible to somehow edit the VMDK?
Add-ons aren't magic
Add-ons are just software in Docker with sprinkles for the Supervisor
If an add-on exists then there's a 99.99999999999999999999% probability that the same software exists as a Docker image
Fair. I'm just curious how people run frigate as an add-on within HAOS. I suppose it's just a proof of concept-type thing then and not actually production enviro?
The Frigate devs discourage it ... so that should tell you something
Ahh, fair.
Oh, interesting. Well then I guess I could spin up frigate in docker on a separate VM and add that in with the proxy addon, however if I'm doing that I might as well just spin up HA in docker as well if it gives more control/compatibility
So, I have HA already running in a docker container but on unRaid. Would this limit my options as well, should I run it in a VM on ESXI to get a more "real" docker experience?
I don't know enough about unRAID, but if you're not limited to some crappy UI then you should be fine
Yeah, I can access terminal and such for each container with unraid
Hmm, might go back that route. Thank you ๐
I appreciate all the info
IMO if you can create and manage containers with Docker Compose you'll have a better time than if it locks you into something like Portainer
rebooting didnt fix it. In the logs I see quite some warnings and also some errors. Not sure what exactly I should be looking for.
Yeah, it's either template-based (shitty UI) or Portainer to manage with unRaid I believe
Portainer ... has issues
Some deployment approaches are overwriting the PATH environment value to an old value. If you see errors like ./run: line 7: exec: nginx: not found, then you need to stop overriding your PATH. This happens with both Proxmox LXC and Portainer (#3223 the database will need to be deleted to roll-back successfully.
from https://github.com/blakeblackshear/frigate/releases
Similar fuckery has caused issues for other images, including HA
oof, that's rough
As long as you know it may shoot you in the foot though at least you're prepared
True... fair point.
Perhaps I'll try and do this all with docker compose on a fresh VM instead of trying to workaround unraid.
I do love HA's backups. It was super easy to migrate from one install to another.
Same goes for any backup though 
If I am going the way of docker, I want supervised - yes?
It appears to have the most functionality
Container is preferred; Supervised is not really worth the bother unless you absolutely need it
supervised is a trap. don't go that route.
you won't have addons, but you can use other containers alongside HA with docker-compose anyway
Supervised is what you use if you hate your future self
Hahahaha, fair enough. Thank you
The end-goal is frigate, mosquitto, and zigbee2mqtt but can run those alongside with docker compose I'm sure
it was once a swiss army knife but long ago that turned into a trap
It's still a swiss army knife, but it's rusty, and some blades will self-deploy when in your pocket
I appreciate the advice
So I'm pretty new to docker, still learning - I'm seeing that for linux, there's docker engine or docker desktop which runs containers in a VM but is more standard across platforms. I am guessing I should probably go with the docker desktop version, correct?
Sorry, I realize that's not a question directly related to HA but it's sort of indirect considering my end-goal.
You don't want desktop, no
Docker Desktop is for, well, desktops, not servers
Install Debian, install Docker + Compose
Ahh, thank you! Much appreciated.
I just have a look into docker-compose.yaml at the main page and why there is set an option privileged: true ?? There is described how to add an USB devices with devices option so don't get it for what is that privileged? Looks really badly for every one with a container experience. It's seriously security issue
The point is that it makes it easier for those that aren't great with Linux
I've lost track of how many people use Docker but can't find their device path
Clearly if you feel it's a risk then ... don't use it
Hi,
I use the Linux-Docker variant with external mariadb. DB-size is approx 40gb. Log shows "Database is about to upgrade from schema version: 30 to: 33". This is already running for some time. How long might this take?
What version of MariaDB?
It is later than recommended in the release blog post.
Server version: 10.6.11-MariaDB-0ubuntu0.22.04.1 - Ubuntu 22.04
I believe the blog says 10.6.9
Hopefully this is a good place to dump this, i have an issue, i have an ubuntu server running vritualbox and a HA OS VM in it. If i shutdown the VM and try to start it up again, i get a VBoxManage: error: The virtual machine 'home_assistant' has terminated unexpectedly during startup because of signal 6 VBoxManage: error: Details: code NS_ERROR_FAILURE (0x80004005), component MachineWrap, interface IMachine I then reboot the host and everything is fine.... anyone seen anything like this?
The google search for this bring up a lot of things which don't really makes sense for this particular case
@obtuse talon if you're running HA OS you can get the console by connecting with keyboard and monitor
but i am locked in
I heard something about thread/matter is not supported in a docker install only with vm or baremetal, anyone know? I was planning to go to docker only install
yes, the matter addon is currently made only for haos afaik
you can try building it yourself? ๐คทโโ๏ธ
damn :/ would hate to get a vm just because of that
just... wait?
what devices have you that you need it immediately?
not immediately, just wanna be ready, got the skyconnect
you gain pretty much nothing by doing that now
yeah dno if it's worth to do all docker install then if it's not supported in the future
who said it won't be supported in future?
dno about the future, did see the home assistant live release and they said only supported in haos
currently, yes.
as I said
but you gain nothing from it currently anyway, so ๐คทโโ๏ธ
its FOMO but there is nothing to miss out of yet
yeah ain't gonna setup at the moment then, just don't wanna do all docker-compose and then switch to proxmox when/if it's needed ๐
Nobody needs Proxmox ๐
what's wrong with Proxmox ๐
it's... austrian
nah, we are proud container install users and we need our daily fix to tell you
xD haha
Proxmox is what I used in the past I used LXC before docker
All HA + other stuff purely on docker?
docker, if not possible or not needed - bare metal. windows does not dockerize so thats the last thing I use vms for pretty much
docker without a VM would be best, thinking about running light desktop + HA on a n6005
desktop
why?
just for wathing movies, shows etc not managing, was thinking portainer
bad thinking but you'll learn ๐
people here seem to be the most peaceful with running ha on a dedicated machine
Router on its own hardware, HA on its own hardware, that's the life for me. Everything else goes on other machines, mostly in VMs
Hi, I have been running HASSOS 9.3 in a QEMU VM on Unraid for a few months now. A few days ago, I was unable to access homeassistant. Looking at the VM, it was stuck in the EFI shell. FS0: (where BOOT and cmdline.txt should be) was completely empty, so today I copied those two from the original qcow2 boot partition to my VM (mounted the original disk as well and simply cp'd from FS1 to FS0). It now boots, but I am presented with the initial setup screen. Running backups using the cli tool also shows there are no backups :/. I have a backup of the qcow from before I tried to fix the boot partition
I'm not sure what to do
SQLite. I would guess 30 minutes. MariaDB likely 4 hours. Ymmv based on disk speed. 1 hour per 10GB of state data for a moderate disk. 10m per 10GB for a fast ssd and fast cpu. These numbers are guesses based on prior migrations
Thanks for all inputs, gonna think about it and maybe do all docker ๐
Maybe something like this https://pastebin.com/WJLSNhyJ
thanks: now I ran into this: The recorder backlog queue reached the maximum size of 65000 events; usually, the system is CPU bound, I/O bound, or the database is corrupt due to a disk problem; The recorder will stop recording events to avoid running out of memory
I have a relatively old SSD + i5 processor
How can I retry the migration or fix this?
wooha - hd full
free up disk space and as long as you don't have any actual corruption, restart HA, and it should try again.
Make sure you always have at least 1.25x the size of the database in free space available for temp tables and schema changes
yes, freed 18 gb and restarted. but 1.25x is hard as the ssd is only 128gb. any way to shrink the db beforehand? use older docker container and then reduce "purge_keep_days" ?
you can call the recorder.purge service manually to delete old data
tried that right now - no result so far. maybe it is not possible during a running migration
I want to use my raspberry pi 4 4gb as homeassistant host and want to connect it to my tv so i can see my dashboard. Can i use the normal homeassistant imager or use pi OS and install it on that
HAOS does not output a visual interface. Just a command line.
You could install Debian and run Home Assistant in a docker container. But you may also be better off using some thing like the Cast #integrations-archived to display your dashboard on the TV
Its an old school tv so that doesnt work, Im really new to this, is it hard to do docker containers etc?
Docker containers aren't hard. If you're new to this I'd install HAOS and look at getting something you can connect to the TV to see a dashboard on. Maybe a Pi Zero W, maybe a Chromecast ๐
Whatโs a good way to display a hass dashboard on a chrome cast btw?
Does it need to be manually casted by a pc or something?
Owh okay, but the thing is i want it on one pi. Not multiple
HA can cast natively.
I need to play with that
Home Assistant is a server. You access all dashboards from other devices using a browser. Your phone, another laptop, pc, or anything else in your home that has a browser.
Just use the cast integration?
Thanks
The HA instance on my Pi4B is crashing and throwing "read only filesystem" errors with an SD card and 2 different SATA SSDs connected via USB3.0.
Is my Pi cursed? Did I just get really unlucky with storage media?
(more info in thread)
Rip sd card
Yeah okay, but i want to use an old tv i dont use for showing temperature and ring camera etc but i dont want to use multiple pi's. I just want to host HA and project it on tv
You'd still need some other device connected to the tv to cast a dashboard to it. A chromecast, etc
and even then you can't interact with it
Owh, that too bad
Then im gonna experiment and export it later to my nas if i upgrade my ram
I will install Kubuntu, set HA in docker and Chrome browser, also in docker. Do it in compose and get what you want. It's not easy for beginners but 100% possibile.
Hi! I've had to re-flash my SD card after I accidentally locked myself out of HA (turned off IPv4, don't ask), but I'm 20+ minutes into having plugged the freshly installed version back into my Raspberry Pi 4 and I'm still unable to access HA via homeassistant.local:8123 or the ip address directly. Was pointed in this direction for help!
The specific error in the browser is "This site canโt be reached. homeassistant.local refused to connect."
I've tried multiple browsers, and multiple devices.
I've also tried flashing the SD card a second time.
Without a monitor hooked up, it's hard to know what the problem is
Yeah I'm getting that feeling too. Will have to go buy a micro hdmi or w/e tomorrow morning. Any other tips I can try before then?
Serial tty should also work
isnt there another port you can reach to see if it's having issues
like 4000 something
4357
At this point :4357 is mainly telling you if the supervisor is running or not, which can be helpful
Like, "oh my problem is that bad" helpful
you can try changing DNS on your router, and then reboot the pi
After updating to Core 2023.2.0 some Aqara humidity sensors do not work with the statistic diagram card. After restoring Core 2023.1.7 everything works fine again.
Any ideas?
Thanks a lot!
I've seen a couple posts I think mentioning something wrong with humidity statistics. Suspect this might be the fix coming in .1 : https://github.com/home-assistant/core/pull/87202
Oh looks like 2023.2.1 is already released just now.
Weird, I'm on 2023.2.0 and humidity, voltage, etc all load fine
I upgraded to 2023.2 together with mariadb and all my history is gone, no energy nothing shows up in logbook and history
Is it still running a migration or something?
You'll have seen a persistent notification when the migration started, if you left it it'll have cleared when the migration finished
I donโt see anything on the current log
Mariadb log also does not show any problems but still history does not load from anywhere
is there another version of HAOS with different network drivers, my nic is not supported
Did you upgrade the Mariadb add on first?
(If you did it at literally the same time I guess Ha could have got confused if Maria restarted while it was trying to migrate)
File a bug in the os repository with details of the nic and how to enable them on Linux. They might add them unless itโs something really esoteric.
I upgraded to 2023.2 it gave me a message about a performance bug in mariadb and that I should upgrade it, I did, and not all the history is gone
if I look into the statics_run table, I only have entries after the upgrade time
how can i restore my backup? ๐
ah i found the option, fingers crossed
What nic are you using
Never mind, started working after bios update
so just finished installing on baremetal and pc has internal video card INTEL and I added an Nvidia dual port video card but not getting video output from that card, only the internal nic.. how can I find out if I can utilize the card for Hw acceleration for transcoding
oof, i restored backup, upgraded again, now none of the addons work, they don't register as running
||[2023-02-03T09:06:31.850Z] info Wrote default config file to ~/.config/code-server/config.yaml
[2023-02-03T09:06:32.976Z] info code-server 4.9.1 f7989a4dfcf21085e52157a01924d79d708bcc05
[2023-02-03T09:06:32.979Z] info Using user-data-dir /data/vscode
[2023-02-03T09:06:33.270Z] info Using config file ~/.config/code-server/config.yaml
[2023-02-03T09:06:33.270Z] info HTTP server listening on http://0.0.0.0:1337/ [2023-02-03T09:06:33.270Z] info - Authentication is disabled [2023-02-03T09:06:33.271Z] info - Not serving HTTPS||
You don't get a video output from a nic, I think you mean GPU
Also what did you install exactly? HAOS? If so what are you wanting gpu for
Restart your host
@manic bane on it
Thank you that fixed it
I have the same problem, but no available backup. On 2023.2.1 the error is still there and my Energy-Board is gone too.
@dreamy turtle I converted your message into a file since it's above 15 lines :+1:
Hey ๐ i am not able to install universal-silabs-flasher on my HA on Raspberry Pi, for flashing my skyconnect stick
hey everyone, hass fails detect the modbus converter as audio, can I change the group to dialout?
ls /dev/ttyUSB0 -la
crw-rw---- 1 root audio 188, 0 Feb 3 14:28 /dev/ttyUSB0
Sorry I meant gpu, I installed Haos on bare metal looking to get better performance from frigate
Hi
one of this hardware would be OK for HAss ?
TINY Lenovo M53 Tiny: Celeron J1800 2.41GHz 4G 16GB SSD NO-Optical
TINY Lenovo M600 Tiny: Celeron N3000 1.04GHz 4G 16GB SSD NO-Optical
with 256 Gb SSD
Take a look at these
https://www.servethehome.com/introducing-project-tinyminimicro-home-lab-revolution/
thanks !
If in doubt, search for the CPUMark rating - the Pi4 scores around 900
yes, that's because I have one unit of each in a closet ...
was an option to give them a usage
If you already have them sitting around, there's no harm in giving them a try. Worst that'll happen is it'll run slowly. It sounded like you were asking for advice on buying something
i'm not getting HOAS booting on an Futro-S920. Deleting my attempt and trying Proxmox => is booting. What can i try else? (sudo efibootmgr -c... allready done)
Check out https://community.home-assistant.io/t/no-bootable-medium-haos-fujitsu-s920/516008/6, ensure secure boot is off and that the efibootmgr command is correct
Hello,
I have an issue with moving the data disk to an usb SATA SSD.
I tried both the UI and the CLI way of moving data disk to /dev/sda.
The data disk remains on the eMMC storage however
https://snipboard.io/P1wFl5.jpg
Any idea how to solve this issue?
thanks for the reply, but as told above (today at 12:03) secure boot is off and efibootmgr is correct
ty
A button in the UI would be for #frontend-archived
A physical button ... that'd be more #hardware-archived probably
@dreamy turtle if you have proxmox running, why not run haos in a vm?
Run this command, in a shell on your proxmox, "one click" HAOS install
bash -c "$(wget -qLO - https://github.com/tteck/Proxmox/raw/main/vm/haos-vm-v5.sh)"
because of lack of space - only got 8GB on the SSD and that is nearly not enough for ha, mqtt, influxDB, hacs, ... / so with proxmox added, the ssd is definitly to small.
Yeah bro. Get a bigger one ๐ pretty cheap
Why is it that when I get an update via docker, when HA restarts, it fails to start up logging and a whole heap of stuff, but if I do a restart via the UI, it all restarts successfully
How did you install HA?
just in a docker-compose.yaml
What does it mean for logging not to start up?
ie the problem seems to trigger when my cron does a: docker compose pull && docker compose up -d
Upgrading indiscriminately like that is a bad idea
you know the history logger stuff? I believe it stores it all in a mariadb database - but I have a depends_on in the hass service on a healthy mariadb - so the DB should be always up.
at least as far as HA is concerned
I have no such problem
strange - it happens to me every time an auto-update happens....
as I said though, restarting HA via the UI causes everything to launch properly
Automatically updating isn't a good idea
well, never updating is also a bad idea.....
and random updates that happen are also a bad idea.
so, I dunno - we get updates to the docker image for a reason - or they're just being done randomly, and that'd be a bad idea....
so which bad idea is the best?
Almost every monthly release comes with breaking changes which will eventually bite you if you're doing automatic updates
that's fine - every normal docker image update causes problems anyway
Hi all, I keep getting these errors when installing:
https://imgur.com/Wf8b09z
