Jun 09 12:13:56 homeassistant systemd[1]: docker.service: Main process exited, code=exited, status=2/INVALIDARGUMENT
Jun 09 12:13:56 homeassistant systemd[1]: docker.service: Failed with result 'exit-code'.
Jun 09 12:13:56 homeassistant systemd[1]: Failed to start Docker Application Container Engine.
Jun 09 12:13:56 homeassistant systemd[1]: Dependency failed for HassOS supervisor.
Jun 09 12:13:56 homeassistant systemd[1]: hassos-supervisor.service: Job hassos-supervisor.service/start failed with result 'dependency'.
Jun 09 12:13:56 homeassistant systemd[1]: Dependency failed for Dropbear SSH daemon.
Jun 09 12:13:56 homeassistant systemd[1]: dropbear.service: Job dropbear.service/start failed with result 'dependency'.
#installation-archived
74689 messages · Page 90 of 75
Make a new VM. Restore a backup
Backup of Home Assistant ?
Because i don't have backups of virtual machine in VMware Workstation
yeah that one I have
How could I get to the CLI of existing VM that is endlessly booting?
How big is your virtual disk image when it isn't running?
You might be able to mount it and poke around in it, but creating a new VM and restoring a backup will take less time.
I'm assuming the your backup isn't only existing on the virtual machine though
it's 30.6 GB
I mounted the a copy of .vmdk file of virtual machine, on a Debian 11 virtual machine. What should I be looking for on the filesystem?
I read the boot logs with journalctl utility
What else I can do?
They are on the VM and on a network share to external hard drive
I would like to prevent Docker failure to reboot the whole Home Assistant OS virtual machine
Or perhaps pass options to GRUB to boot virtual machine in other runlevel than the one with Docker service in it
You mean on adding the virtual machine .vmdk file to another VM, then mounting it and changing something on the HA OS file systems ?
Already done that
root@debian:/mnt/etc/grub.d# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 5G 0 disk
├─sda1 8:1 0 4G 0 part /
├─sda2 8:2 0 1K 0 part
└─sda5 8:5 0 975M 0 part [SWAP]
sdb 8:16 0 32G 0 disk
├─sdb1 8:17 0 32M 0 part
├─sdb2 8:18 0 24M 0 part
├─sdb3 8:19 0 256M 0 part
├─sdb4 8:20 0 24M 0 part
├─sdb5 8:21 0 256M 0 part /mnt
├─sdb6 8:22 0 8M 0 part
├─sdb7 8:23 0 96M 0 part
└─sdb8 8:24 0 31.3G 0 part
/dev/sdb device is the disk of HA OS virtual machine attached to a Debian VM where I can mount its filesystem
I could use journalctl utility and get the last failed endless boot logs
I know why HA OS is rebooting after Docker service fails to start
root@debian:~# mount | grep sdb
/dev/sdb3 on /mnt type squashfs (ro,relatime)
root@debian:~# cat /mnt/etc/systemd/system/docker.service.d/failure.conf
[Unit]
FailureAction=reboot
However I can't change the FailureAction to none as the filesystem is read-only
squashfs
Thanks
I started the restore process
How much time that would take?
There is no progress bar or smth similar
Hi guys I am running HA OS in Rpi4 4GB. It was running all well. But suddenly I dont know if the recent version update caused it or not, bu my memory usage spiked after restarting or running the RPi4 for some time and it shoots to around 60% usage. The log book feels sluggish, and the whole HA stop responding after sometime. But there is no exact interval on when it caused. How can I find the cause for this? Is there any tool that finds out what is using the memory?
I am running HA core on my unraid server. I have installed some docker containers but how do I add them into home assistant now?
What do you mean by that? Add a link in the sidebar, or connect an integration in HA to them?
Hi guys,
I'm stuck on HA preparing page can someone help me out with this? Thanks
Click the blue circle and you'll get more information, then use a code share site to share them if you don't understand them.
Please use a code share site to share code or logs, for example:
- https://dpaste.org/ (select YAML for the language)
- https://www.codepile.net/ (select YAML as 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.
Command '~share' was invoked by: Tinkerer
Thanks for the quick response. I have copied the code and pasted it on the above link clipboard
What hardwarre are you using?
Windows VM Virtual Box - Linux 64 bit
Other than
[06:15:06] INFO: Watchdog restart after closing
s6-svwait: fatal: supervisor died
I don't see obvious issues
No idea, hopefully somebody else will be along who can help
Thanks a lot @humble mirage
morning everyone, can anyone give me a hand with a freshinstall ?
newb here ahah.
So I can't reach my homeassistant via local network "homeassistant.local:8123"
I'm using a rasp 3 b+ connect directly to the router
<@&330946878646517761>
🔥 ?
Whats burning?
🚒
nothing burning just trying to get attention to my question XD
but its's a burning question 👀
...
Pinging mods for that is not the way 🙂
It can be used for requiring moderation, like banning an such 😬
sorry then xp
am i the only 1 having serious problems with installing the VBI image in vritualbox
What problems are you having?
If http://homeassistant.local:8123 doesn't work for your network, you can try http://homeassistant:8123 alternatively. If that doesn't work either. You need to check your router to figure out which IP address your device had assigned. After you found that address you can use that instead. For example, if my router would assign 192.168.1.2 to my device, I could access it using: http://192.168.1.2:8123
i get stuck on a screen that says Booting 'Slot A (OK=0 TRY=1)'
and sometimes it goes to a screen where i can select between slot a and slot b but b says its an invalid filesystem
They left...
🤷♂️
Hello, it is possible to install the HAOS to Rock Pi 4? If yes, which method is the right way?
Hi all,
I tried installing ESPHome on HA. I appears it worked just fine for a while, and I even got it going setting up an ESP device. After a while the computer HA runs on crashed (I think, not sure) and since then I can't access HA on the usual url (homassistant.local:8123). The observer url also returns nothing.
HA runs in a virtual maching (Oracle Virtual Box), and I can see that it is doing ok and interact with the Command line as well. I used it to try to restore my backups. Doing that seemed to go fine, but it didn't fix the issue. How do I get in contact with my HA installation again, and does anyone know what happened?
If you use the host IP - rather than homeassistant.local - can you connect?
If I use xxx.xxx.xxx.xxx:8123 I get the error "refused to connect". The IP is the one I see for the machine inside the virtual box, not the one for the physical machine
I have experimented with and without the mask bit as well (the number after the /)
Well, the subnet mask isn't a thing you put in the URL
If you have console access, what does HAOS think its IP is? Does that match what you're trying?
Okay. That may contribute to the fact that it doesn't work, then 😄
If you have console access, what does HAOS think its IP is? Does that match what you're trying?
Heck, from there can you ping your router?
in the HA command line, both ipconfig, ifconfig and ping are unknown commands.
The physical machine it's running on is on the network if that is what you are after. I am controlling it via VN
C
...so how can I see which IP HA thinks it has?
Got it. Sorry for not checking deeply enough. The command dns info does the trick it seems.
I can't ping it from the same home network though (and it also doesn't return any content in the browser on port 8123)
Is the IP it shows on the same subnet as the rest of your computers?
If the subnet is identified as the numbers before the last . in the IP, then no
^^that is what google tells me - im a noob when it comes to network-stuff.
*no as in, no they do not share the same number before the last dot, just to be sure I express myself clearly.
Looks like it is to me. The settings around network looks like this:
Though I could paste in an image here. Anyway under network the "attatched to" option is set to "bridged adapter"
(but with much better spelling, however)
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.
Command '~images' was invoked by: Tinkerer
for VScode minimum 8Gb ram is required?
I was asking in context of x86 hardware for running HA on proxmox or some virtual method
I am searching for these https://shabandigital.com.pk/product-category/all-categories/computers/mini-pc/
Right, but you're asking about the requirements for one add-on
Okay, here's the link @humble mirage : https://i.imgur.com/Bbqzy4v.png
Now you get to work out why you're getting a different IP allocated for HA
No clue, man. I suppose the better approach here is to nuke the VM and start over while keeping a better eye on things so I can actually see if there are any errors or anything.
Anyways, thanks a buch for giving it a try, Tinkerer. It's really appropriated!
Hello, I'm looking to install home assitant for the first time. I have a desktop computer, a tower running truenas, a couple raspberry pis and some other OLD pcs not in use
Is there much of a difference in functionality based on which I use?
No
The hardware you run on is pretty secondary, the main two choices are:
- Do you want add-ons, if so you'll be using Home Assistant OS - which can be run in a VM
- Do you love Docker, if so you'll be using Home Assistant Container
Yes, addon like frigate might require better hardware
otherwise, dashboard is still dashboard
Frigate though benefits more from a Coral device than any CPU choice
I'm not to sure until I have had chance to play with it
It's easy enough to migrate to new hardware anyway
I think it can probably do so much more than I am aware of
If I go for OS and then need to migrate to container or vice versa is that possible?
Yup
They all use the same Home Assistant software, it's just how you install it (and other software) that changes
just three simple steps, Backup, install and restore
availability right now
In that case I think I'll just get my hands dirty with any of them and then come ask some questions once I get a bit more familiar
Hi I am looking for the supervisor option so I can edit a yaml file. started with home assitant yesterday so please be gentle
Found it. thanks a lot
I was looking to create a link in the side bar. I managed to create a new page and then add a web view with the IP address to my docker container. Is there a better way to do it?
I run HAOS with ~20 add-ons. Since a few days the "free disk-sensor" reports about 28 GB less free disk than usual. The nightly snapshots (that are moved to other storage by script after completion) is usually around 14 GB so it could indicate 2 hidden/failed snapshots somewhere. Or something else completely. There are no files in /backup except last night's snapshot.
Since I have no knowledge how to navigate in Linux or in Portainer, I would appreciate some guidance how to find out the files taking up this space. Any tips?
Then it's just a panel_iframe integration, nothing fancy
Command '~panel_iframe' was invoked by: Tinkerer
You should get rid of portainer. Is one of the 20 addons the samba addon?
This is really like any other "where is my disk space going" situation. You would start exploring directories and see which files are big. You can do this using the command line or by using the access you have to the shared folders. Browse the directories, see what is large. You might even be able to run a disk space visualizer from the computer you are connecting from. If you find nothing that explains it, you could try running ha su repair in the console.
can a BT dongle be seen by a VM ha?
(currently my VM HA can't see my laptop's native BT)
should be a USB dongle (on internal USB port, like ngff with the wifi).. you can passthrough that USB device and should work
what hypervisor?
@gaunt ember sorry I dunno what that is
They're asking what VM platform you're using
vb can do passthrough no problem
oh i can't share screenshots 😟
so it has to be usb, it can't do passthrough for the native laptop one
it is probably USB, if it's x86 platform
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.
Command '~images' was invoked by: Tinkerer
if the laptop is fairly new the bt is probably on the m.2 of the wifi
do you happen to know/recommend recommend a dongle that works with HA by any chance?
ooh, now I get it
any bt adapter will do probably
I thought it only worked with specific brands or something ^^
thank you for the help!
As long as you pick something that works with Linux you'll be fine, and most will
Look for CSR, eg CSR 4.0
Yah I see it’s just an iframe. Is there any downside to do it that way or is there another way? I’m not familiar with docker so I’m not really sure what I’m doing haha
It's just a pointer to the web page, that's all
It has nothing to do with Docker, or how you install anything
Did the ha su repair which "completed successfully".
I can't find anything strange in the samba folders. All the accessible folders, /config and all other samba shares are just a few hundred MBs all together.
In a Windows environment, I'd know what to do and how, but in Linux and containers, I'm totally lost.
Since it took two big leaps of ~14 GB at a time it seems reasonable that there are some incomplete or broken snapshots. Do you know where to look for those?
I suppose they are created in a temp folder somewhere and then moved to /backup?
treafik
hey guys wondering if anyone has an experience with HA Z-Wave configuration in docker ? Having some issues getting the Z-wave portion to work, zigbee was a snap
Looks like i need to run another docker container Zwavejs2Mqtt to get this working
if you used ZHA that's already part of HA. But anything else like Zigbee2MQTT or deCONZ need their own containers
the same applies for Zwavejs
thanks firstly im running HA Core version, What ive done so far is, exposed in the docker compose for HA the USB ports for my Zwave and Zigbee controllers, rebooted HA. I've then installed the ZHA integration- it asked for the Zigbee location which was displayed as an option on the popup, so i think from the Zigbee perspective im good to go. Right?
Where do i go from here regarding Z-Wave. From my research looks like i need to spin up a docker container: zwavejs2mqtt - which is a bridge between zwave controller and HA, if im correct... thanks.
#zwave-archived can help with Z-Wave, but do see the Z-Wave JS docs which cover that
Also, you're running HA Container if you're running HA in Docker
Everybody has the software called Core
Got it missed #zwave-archived. Yes i reviewed the documentation for ZWave JS, which as you said im running the container version puts me into Option 3 create the container Zwavejs2Mqtt - thanks i will work on that installed now.
Hi! I'm trying to run HA as a virtual machine on windows, tried VMware Workstation Player and VirtualBox. For VB it often gets stuck. When trying to restart it has some problems with kernel time synchronization. In VMware it always has the problem with kernel clock, so I was even not able to start it.
Can someone help me if some had this problem before?
That worked! Added my first Zwave device, im off to the races! Thank you!
Hi Guys, i just installed home assistant onto virtualbox, and everything seemed fine, then i downloaded mosquito, and now it won't load the automations page. I can't seem to figure out what is wrong
seems two entirely different things
Agreed, but i have done quite literally nothing else since installing it
what doe's it say when you try loading automations page?
yeah, the two things have nothing to do with each other
nothing, it just loads forever
whereas every other page in HA works just fine?
Ahh
whatever page it is, try clearing your browser cache
it's always cache
nevermind im a fool
i changed my user so it can only authenticate locally
and somehow that blocked access to devices?
i guess turning that off allows me to access devices again
the Allow Network Access only on the user or whatever it's called it means you can't login using a domain name
that shouldn't affect this either
that's very weird
does it happen if you log in on incognito as well?
let me try
it's definitely not reproducible here
okay weird
so if i try to log in
it says i can't log in as a remote user
despite being on the same computer the vm is running on
but yea it happens incognito too
it's supposed to be bridged
was just trying to think of possibilities but really have no idea why it's happening
if you really want a local user maybe make a second one and see if it happens to that one too. Or check again another time if it's still the case
a different browser also or simply ignore haha
Hi, which guide should I follow to install supervised and home assistance in a raspberry pi? (Without installing home assistance os)
Why there’s no point? The home assistant os takes control of everything, I’d like to do more things
Read that link. Doing 'more things' is not supported
If you want that, use the container install
I’m reading the link, thanks, bit can I install addons and similar in the container? (My understanding is that it’s not possible)
No, you install your own software
Hey everyone! tl;dr - First time installing (HA on Proxmox VM), and I'm way past the ~20min ETA that is prompted the first time you visit the local URL. Getting 500 Server Error when trying to download Docker things.
With it being a 500 error, there's nothing I the end-user can do, but maybe bring awareness to an aspect that's experiencing downtime?
To be precise, here is what I'm seeing
22-06-11 03:22:36 ERROR (SyncWorker_3) [supervisor.docker.interface] Can't install ghcr.io/home-assistant/qemux86-64-homeassistant:2022.6.5: 500 Server Error for http+docker://localhost/v1.41/images/create?tag=2022.6.5&fromImage=ghcr.io%2Fhome-assistant%2Fqemux86-64-homeassistant&platform=linux%2Famd64: Internal Server Error ("Get "https://ghcr.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)")
Hi everyone. I'm running a docker with portainer on a Rpi 4 and i'm trying to copy over some configuration files from a previous install. How can i access the home assistant files within the docker container?
I am quite convinced there are "dead" files in my system. My install takes up ~60 GB data including all add-ons. But if I clone the VM and make a complete snapshot restore , it wipes everything and after the restore it only takes up 32 GB. My guess is that there are failed InfluxDB snapshots stored somewhere since they consume about 12-14 GB each and nothing else comes even close to that. Any tips where an incomplete snapshot would be stored and how I could check if that is the case?
I did a docker install earlier today and had this as well. My solution was a pretty simple one. I just rebooted the system. I wasted quite a bit of time first looking through firewall settings on my router ...
I want to restore a .tar backup that I created from the webinterface of a Docker Install. It is going to another Docker install. I have googled this, and the common denominator seems to be to 'just untar it on the new system'. Don't I have to put it in a specific folder?
It replaces the content of your config directory
so i have to copy it into that directory and untar it there, right? or one folder above it?
You should look at whether the tar file has a directory at its root, but you want the content in your config dir
Hey there! I'm currently having some trouble with my Home Assistant setup, and i'm thinking about doing a fresh install. I'm currently using the VMware Image from Hass.io. It works, but after 1-2 days the GUI stops responding, and i have to restart the VM to get it working again. So what is the more stable option now? Should i do a fresh install using the Hass.io Image, or should i just install an Ubuntu/Debian/whatever and install Home Assistant afterwards? What would you prefer? 🙂
I'm not sure where to ask this question, but is anyone else running their install in Kubernetes/Docker and have node red working?
You probably wanted #node-red-archived
i wonder about possible advantages running Home Assistant Operating System
until now i just use the docker version which runs fantastic
@rich plover if it's fantastic then keep it as is. HAOS is docker, but managed for you
HA Core in docker via Docker-compose...
When trying to install HACS my system is telling me I do not have permission to write to my directory? Even sudo won't fix it.... I have also already chmodded the folder to 775
any ideas?
HA Core in docker via Docker-compose...
that just sounds like a container installation
but worded strangely...
in any case, your config directory is just a directory on your server, so review the owner/group/permissions
yeah....I thought I'd just describe my circumstances exactly lest I miss relevant information
cause on an old laptop that I used as a PoC the last couple of months prior to this new permanent NUC I think I installed it without docker compose. And didn't have any permission issues
but yeah...you're right of course. I just took a digital crowbar to it. gave the folder 777 chmod temporarily as I installed it
Noticed my installation was no longer online, this is what I'm greeted with after power cycling. I'm surprised I was able to get it working in the first place, this is have no clue what to do
https://cdn.discordapp.com/attachments/397426163649216512/985648411250401310/20220612_155143.jpg
"Good" news, your SD card is dead/dying
You may be able to image the SD card and recover your config, since you have no backups that's your only hope at this point
Meh I'm ok with starting fresh. I'm not super knowledgeable when it comes to stuff like this. Between trying to get teleport to work again on my router, getting syncthings to install on truenas and home assistant setup properly, I've managed to get truenas running and had home assist accessible but not really useful
Hello everyone. HomeAssistant doesn't see usb passthrough device. When running lsusb on the host the Aeotec Z-Stick 7 is listed as Bus 003 Device 004: ID 10c4:ea60 Silicon Labs CP210x UART Bridge and dmesg |grep tty* shows the most recent device as usb 3-2: cp210x converter now attached to ttyUSB0. I've plugged and unplugged multiple times to confirm it is indeed ttyUSB0. The path I'm using is /dev/usb/ttyUSB0 but HomeAssistant doesn't see the device.
OS/docker: UNRAID 6.10.2/Docker 20.10.14
Hardware: AsRock Rack X399D8A-2T, AMD Threadripper 2950X
Docker command:/path/to/device = /dev/usb/ttyUSB0
Logs: https://hastebin.skyra.pw/qofibufuru.yaml (Container logs, container terminal, unraid terminal)
I've tried the install using the Home-Assistant-Core and homeassistant images from the Community App store.
I've run lsusb in the container's terminal and can see that there is a device listed with and ID that matches the Z-Stick 10c4:ea60 but it's still not showing up. Am I missing something and just being a noob? Thank you in advance!
are there different pros and cons of installing mosquitto on the OS directly vs in a docker container (when HA and everything else is also in a docker container)?
It all works the same regardless
If you're already running other things in Docker then Docker makes sense
Hi, i'm running HA, zwave2mqtt, zigbee2mqtt and nodered on a ubuntu server with docker now.
I want to seperate my installations.
So i have new laptop witch i have installed proxmox on, and set up a vm with HASOS. I'm now a little bit confused. How do i transfer my zigbee/zwave/nodered containers to this? I really don't want to start from scratch again
You can't really migrate things other than HA from Container to HAOS
ok, so maybe i should install ubuntu on this laptop and just move my containers.
I'm missing out on the cool things in haos tho 😦
"Cool" things?
You mean having a point-n-click thing for installing software... I'll pass
🙂
Hi all,
A friend of mine has the following issue when installing the generic x86-64 image (https://www.home-assistant.io/installation/generic-x86-64/)
For some reason, he cannot reach Github through the Home Assistant terminal add-on using this image. When he pings github.com he gets a message ping: bad address 'github.com'. However, he can ping other domains from within HA Terminal like google.com or nu.nl for example. Is this something that sounds familiar to one of you?
@twin stag You can disable link preview in Discord by adding < before and > after the URL, like this: <http://home-assistant.io/>
Command '~links' was invoked by: Tinkerer
It sounds like a DNS problem
Yes, you should say so, but the strange thing is that he can ping github.com from his laptop...
Please edit your post and fix the link 😉
Sure, but the laptop isn't HA
HAOS uses some fallback DNS servers AFAIK
Yeah, I thought, maybe it was his firewall, but that couldn't be. I never used this image, so I do not know where to search tbh.
it's supposed to. Never seem to actually help 😅
What do you mean?
HAOS has fallback DNS servers in case the one taken from the DHCP don't resolve
but they also usually don't work automatically
one option set manual DNS server under Settings --> Network --> Ipv4
to like 9.9.9.9 or 1.1.1.1 or for example
Didn't work unfortunately. tried the Google DNS addresses, but to no avail
of course not
would have been too easy 😅
the other option that usually does fix it is through nmcli, which needs access to the host so keyboard/monitor
I tried to use login, but it asks me for a user and I have no clue on his system. Tried root, but does not work
you are not in the host console
Indeed
I know nmcli exists in the advanced addon but I don't know if it can do the same as through host
nmcli is not known as a command, so I guess not. I did not install this setup, so I am a bit in the dark here.
if it doesn't you are in the official ssh addon
[core-shh] or whatever it starts with
there's also the other one, that one has nmcli
Indeed. let me install the other one then.
Yes, got it!
I use the other one myself too, let's try this
Unfortunately this did not help either. The thing is that pinging other addresses like google.com is working. It seems that pinging github just does not work...
Yes, I did
login did not elevate my rights though. It keeps asking for a password. Even with the other addon
Protection mode is disabled
that's expected though
access to the host isn't possible with an addon
they are containers, they don't have access
either keyboard/monitor or enabling debug ssh on port 22222 gives that kind of access
is HAOS directly connected to the router or branched out to some switch?
Yeah, I am not close to him right now, so I cannot look straight into his server.
But thanks for your input. I will have to investigate this further. Giving access on port 22222 is something that I'm not sure about if I can do this remotely. I guess not....
not really. Would need a usb stick that he has to plug in the device with the ssh keys
you can make the keys and he just plugs it in 😄
on supervisor does he get 505 errors ?
I tried to retrieve the address of github using the dig -x command with a github ip-address and it does return a github address. Not sure what that means exactly, but I wanted to make sure that there is some connection.
I'll check
You said originally he can't ping github on the addon but that's probably irrelevant or? As in the real issue he can't update or install anything new I assume because of it
That is exactly the problem. He cannot install HACS
That was initially his question to me and then I found out that I couldn't pin github
That does work indeed, but because of the fact that github cannot be reached, he cannot install HACS. I can ping Github on my install without any issues.
and this also doesn't work instead? https://raw.githubusercontent.com/hacs/get/main/get
Nope. Can't ping that either. Also get ping: bad address
He's not blocked on github. He can ping github on all his other machines
I have to go to dinner now. Sorry, have to go for a while.
Thanks for your support!
We get at least two days of sun a year here I'll have you know
You're from Finland?
Hi folks, I'm using balena to run home-assistant on my raspberry pi 4. I maintain this on balenaHub . I was wondering if it's possible to make this part of official installation guide? I'm happy to do the work but I need some direction 🙂
Is there an apt-repo or way to install Home Assistant directly on Ubuntu 22.04?
Docker is fine.
Thanks!
So I was looking to migrate a local raspberry pi install to a docker image install on a raspberry pi. However i cant see an option under backups to import the back up export from my original install into the new docker install. Any ideas?
Is it the same Pi?
If so, just point the /config folder at the existing folder
If not, just tar up your config folder, copy it to the new Pi, and untar it wherever you want it
AFAIK backups only exist in Core and Container to allow people to more easily migrate to HAOS
I'm looking for some help with a docker install. I'm trying to get HA setup using network_mode:host (so that I can add the homekit integration) but it stops working. I'm also running a pihole container on the same server
Are you using Docker Compose, or command line?
docker-compose
Please use a code share site to share code or logs, for example:
- https://dpaste.org/ (select YAML for the language)
- https://www.codepile.net/ (select YAML as 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.
Command '~share' was invoked by: Tinkerer
Share your compose file then ☝️
You're not using host mode networking
# network_mode: host
ports:
- "8123:8123"
networks:
- proxy-network
``` ❌
yea, thats what im saying, it works in this config, but if I swap to use network mode host it doesnt work
network_mode: host
# ports:
# - "8123:8123"
# networks:
# - proxy-network
What is "doesn't work"?
then i cant access the dashboard
Without the networks section either?
am I still supposed to access on port 8123?
I am unable to check for updates and receive the message "You do not have any integrations that provide updates.". I have a x86_64 machine running latest Home Assistant OS. How can I get the updates working?
I'm working through an issue, hopefully someone here can help. I have my HA (running as a container) working currently, but I think something isn't quite right with the configuration. In the configuration.yaml file, I have it set like this:
http:
ssl_certificate: /etc/letsencrypt/fullchain1.pem
ssl_key: /etc/letsencrypt/privkey1.pem
base_url: https://my.custom.domain
use_x_forwarded_for: true
trusted_proxies:
- 192.168.1.5
- 172.17.0.4
The documents say I should remove base_url and use external_url and internal_url instead, but when I fill those in and comment out base_url, it goes to safemode when starting up. Any ideas?
they don't go there. They either go under homeassistant: or to be specified in the UI
And base_url: was removed a long time ago
It's not going into safe mode as a result of you removing it. That would be due to something else
Thank you. The issue was I was putting internal_url and external_url under http: which I can see now is in the wrong spot. I actually added the domain in the UI as @grand pivot suggested, and then just removed the base_url part in http in the yaml and it's working fine now (and tts works, which was my original issue actually). Thank you.
I have tested a bit more since my last question(s) in <#installation-archived message> and <#installation-archived message>
Then I accidentally restarted HA while a backup/snapshot was in the making today, and now another 10 GB of storage is lost in space.
So I am quite certain that there are broken/incomplete snapshots taking up space somewhere.
Does anyone now where those backup files are stored during the backup (before completion) and how to delete them?
Hello!
I messed up my HomeAssistant install and (shame on me) I don't have a backup.
I have a "Home Assistant Operating System" Setup (prebuild imaged flashed to my RPi4)
Online I found the solution to "copy the /config folder to a new SD-Card"
But when I mount my current SD I don't see a /config folder anywhere. I only have a /docker /supervisor /logs and /lost+found folder.
So I guess the trick with copying /config won't work for me.
Is there anything I could do to transfer my old setup to a new SD?
On the file system the HA config is under /mnt/data/supervisor/homeassistant
Maybe it's under the /supervisor folder you found
There is an homeassistant folder!
In the supervisor folder there is also a backup folder
I will copy things to my PC and then try a new install and copy files over 👍
And set up backups
Oh, and many of the files and folders are "hidden" by default, don't miss those
Yeah first thing when I get all running again is a backup 😅
is it possible to access the camera connector on HASSOS?
if no, will hassos be happy on rpi3?
due to shortage I am forced to make this choice
OS is on rp4 8gb
I am trying to use my new 64mp cam
I'd be surprised if HAOS supports it
https://www.home-assistant.io/integrations/rpi_camera/ is pretty clear that it only works on Core installs
hi so i want to install HAOS on a proxmox vm, is there somewhere i can download an iso or other image file that is not for a spesific system?
on the install page it only has options for VirtualBox, KVM and Vmware
perfect thanks
I'm pretty sure the forum has some Proxmox guides to help you
The scripts from tteck are also pretty descent and maintained
Those are the ones I was thinking of
it does but it was "curl this script directly into the bash on your proxmox box" and i couldnet find the downlad location in the script so i wasnt comfortable using it.
though should always look at what scripts are doing first
I haven't been hacked yet here 😄
oh it is almost cirtainly safe i just prefer to do things manualy for stuff like this (also my proxmox is slightly out of date so i didnt want to risk the script breaking things)
im still on 6.2 (updating it has been on my todo list for a while)
just found that, i was looking for the more general image file extentions (eg .iso)
there isn't one
yep i know that now, but when i was looking through the script for the download path on the weekend i didnt
All possible builds are listed here https://github.com/home-assistant/operating-system/releases
Found answer to my own question, saving the answer so I can find it next time:
There were ~40 GB of failed snapshots (backups) in /mnt/data/supervisor/tmp/.
With those removed, everything is back to normal.
made the following compose file https://paste.debian.net/1244095
produces the next:
in "./docker-compose.yml", line 3, column 5
expected <block end>, but found '<block mapping
in "./docker-compose.yml", line 20, column 8```
any ideas
Your indentation is inconsistent
pardon ?
volumes:
ports:
is that all it takes to brake it 😱
It's YAML, all the usual YAML rules apply
it's fixed omg
Hi, is it possible to install Home Assistant Supervised on portainer in a pi4?
No
Supervised is almost always the wrong choice anyway
You either want HAOS or Container. If you're using Supervised so you can run "other software" outside of add-ons then you're doing it wrong
I really dont know anything about this, but I want to run Home Assistant on portainer in a pi4
and want to be able to use the add-on store
Those two things exclude the other
If you want to use Portainer then you use the Container install method and get no add-ons
If you want add-ons then you install HAOS and there's no Portainer
but if I install the haos version, I really cant do anything more with the pi4, right?
Right, but that's the same with Supervised
Supervised just trades a managed OS for Debian Bullseye that you manage
so its not possible to use home assistant with add-ons store and other things on the pi4?
Not in a supported manner, no
You can ignore this and run Supervised, but then things will break at some point because you ignored the requirements
This is, clearly, not advised
This may be a dumb question, but since a docker container can run any OS then it should be able to run HAOS. Then it should be possible to run Home Assistant on HAOS in a docker container?
The OS is there to run docker and facilitate everything running in docker containers, and you want to put all of that in a docker container?
I was writing a whole paragraph there 🙈
And then you could put whatever you have running all that in another container 😆
all the containers have some minimal base OS, so I guess
Oh I didn't realize that was why the OS is there. That's good to know - thank you.
If you want to run HA in a container, run the HA container. Then you have all the rest of available docker images as your "add-ons"
There's no need to make it overly complicated by running HA OS
so if I have a k8s cluster or mini-kube I can run ha in 1 container (pod) and deploy add-ons in other pods ?
no HA add-ons...just the usual software available as images and run as containers out there in the world
got it so if I understand correctly instead of using the add-on feature in the UI I would deploy images of software like MTTQ and just configure HA pod to talk to them
right
amazing, thank you
Docker containers don't really run an OS though, they're a partition of the "host" OS
that's a good point
anyone had experience with hass not working on a VM when set to use multiple cores? using virtualbox and it only finishes boot/setup process when it's set to 1 core
Did you find a solution to the "You do not have any integrations that provide updates." issue?
I have no Idea where to ask help for this so I guess we'll start here.
I have a new HA VM in unraid, I have a Sonoff zigbee USB dongle and I can't seem to get Zigbee2MQTT to work. I get errors in the log
Sounds like #zigbee-archived , where you should expand on 'errors'
Is there a way to stop HA to keep rebooting? Updated to latest HA os, and there must be an error happening, as it loop reboots. But can't read whats happening as it's too fast
Runnig on intel nuc8
Seems to fail at start job is running for Wait Until Kernel Time Synchronized
Is there a way to see if there is a running backup job from HA? Even something I could use as a condition in a script?
I have a snapshot (backup) scheduled that runs every night. Nowadays it takes about 6 hours to complete since InfluxDB is quite large.
This causes problems if I restart HA during the backup since it leaves a broken backup file that takes up a lot of space.
There isn't a blog post or forum thread for every HAOS update? When updating core I tend to wait a few days and read comments but I see little discussion about HAOS updates. And if I recall correctly there were major issues for some people a release or two back.
Hey there! I followed the HASS Core Pi install guide and everything went well.
However, I'm trying to get HASS to start on startup / power on the Pi 4 but can't find any documentation / pointers on that. I would normally setup a systemd file but given that Core runs inside a Python Env (I don't know much at all about how those work), I assume that's not an option.
Any pointers or help would be appreciated!
Docker FTW
https://community.home-assistant.io/t/autostart-using-systemd/199497 has you covered though
Oh so it is Systemd, even with the PythonEnv. Thank you!
Also, been trying to avoid docker mostly cause I have a bunch of other stuff running on the Pi already and most of it doesn't exist in Docker format so been trying to just run everything as standalones for now (hence the no OS option too).
Where it's an option, Docker generally is going to make your life easier
You can absolutely mix and match stuff in and out of Docker, but it solves so many dependency issues
hi, im getting "eof" on black background when trying to open web interface after last update? how to fix it?
Did you install the SSH add-on?
yes but i only used it in browser, and cant connect now
devices added work through homekit though
You've tried the browser in Incognito mode to avoid any cache issues?
Is there a way to enter the HAOS cli before the system completely boots?
Before it boots, no, because it's not booted
Any sort of recovery mode?
Wym
HAOS has a dual partition setup, like modern Android phones
https://www.home-assistant.io/blog/2022/05/16/home-assistant-os-release-8/ shows the dual boot slots
Tried both, no luck. System dies in a reported bug. Machine does not connect to network and perma reboots
Thats what im having. Others reported being able to access cli, that's why I had hope
finally moving to a mini pc... yeah. prepping myself, and reading https://www.home-assistant.io/installation/generic-x86-64#write-the-image-to-your-boot-media how do I do that when booted from a Ubuntu flashdrive?
Uses Disks are how it's called in ubuntu
I did find https://community.home-assistant.io/t/installing-home-assistant-os-on-a-mini-pc/314722/34 posted, but that used the Ubuntu imager, and a re-downloaded file, and not as the instructions tell us the Url for Balena
Balena is one option yes
although someone was reporting balena wasn't working because some dependency was missing if remember correctly
but how to open Balena in the first place, when I bootup from Ubuntu>
you first need to download it
it doesn't just exist in Ubuntu
alternatively download /or plug in a USB with the HAOS image and use Disk to "restore" to the drive that HAOS image (same thing as flashing)
Disks / Disks Utility I cannot remember the name
where are you confusing yourslef?
a ok, so thats what that community post does: Boot Ubuntu, switch flashdrive with HAOS image, and Restore
I assume you have a GUI Ubuntu booted?
yep
if you have an adapter open it up and flash the drive directly
found the step between Booting Ubuntu and firing up Balena missing in the documentation instructions
It says Copying a live operating system (e.g. Ubuntu) onto a USB device. Then, insert this USB device into your x86-64 hardware and start the Ubuntu.
yes thats fine
Having never used Ubuntu, that triggered my proactive planning
Why not boot it up in a VM and test it out?
Ehhh. Im a simple Mac user....
Are you suggesting I find a way to run a VM on my Mac, dl Ubuntu and test run the sequence?
Why not?
It's not hard to do, and then you're moving beyond guess what's going on stage of planning
I guess, it just never crossed my mind
ok Ill go and have a look how to do that, thx
The bug though says it get's stuck on waiting for a network which yes shouldn't happen given the timeout is 90 seconds. But yours gets stuck there too or it keeps rebooting?
not the same if it keeps rebooting
the command?
Seems to fail at start job is running for Wait Until Kernel Time Synchronized
Had posted this above
but when that fails it should move on not reboot
and if it doesn't fail it would get stuck there while time keeps counting
it's a fresh installation?
No, but it's the third time it happens. If the machine reboots, i get stucl in that. No access to network and need to connect monitor and keyboard. Dresh install solves till next reboot
Yet on the fresh install the network works just fine on DHCP?
Yep
This is definitely worth to open an issue. It's not the same as the other one. First yours reboots instead and the fact it happens after a single reboot is just weird
Got it
what hardware is it?
Nuc8
Are boot logs stored somewhere? Im backup reflashing, buti will gather info next time
pls 😩
@polar bronze I've experienced the opposite once. It wouldn't boot without 2 cpus more than one cpu.
Or have a look at running Proxmox at your new Mini PC. For me as and old Windows user, it was as easy to setup as setting up HA on the Intel NUC I used before. Click 'n run basically. Great tutorial on YouTube https://www.juanmtech.com/install-proxmox-and-virtualize-home-assistant/. And it is lovely to have VM snapshots before upgrades. In my experience it brings very little extra overhead. My +7 years old laptop runs Proxmox at 5% CPU with HA and Unifi Controller.
Time to try for a new install.
Hello ! My home assistant os Bootloop after today's update, should i ask for help here ? or in general ? #330990055533576204 is deprecated
Here
Cool !
So i have the os on a nuc, and since the update done from the webui it automaticaly shutdown right after starting the docker service, like a docker container were asking for shutdown as soon as it boot.
it look like there is 2 .... version ? A and B in grub, but both do the same
any clue ?
i usealy run core on a venv but i am trying to make it more "brain dead" for the familly
Hmm another user today reported the same
x86 to ?
My VM did that too, but given I just have it to help people who're afraid of clicking around the UI it's no loss to me
I have mine in a VM too 🙈
But if it ~~helps ~~ makes me better in your eyes I have also one in container 😛
@wide wind did yours also started rebooting after updating
he mentioned though it works fine on first install then it will keep rebooting if it reboots even once
it is a very recent install, and even before the update i didn't reboot it a lot ... if at all almost
you updated the OS ? I only see an update a week ago not today
Oh 2022.6.6 you mean Core
no i mean os
Oh right
as it happens there was a new Core released today so was confusing
OS was 8.2 a week ago
a week Oo
that doesn't feel right
i had nothing yesterday, and something this morning ... unless hass OS delay update to avoid download rush ?
according to github yes https://github.com/home-assistant/operating-system/releases
we don't delay no but unless it didn't make it to the file it checks
perhaps your instance was mysteriously unable to contact the update server
the device is on the lan so no firewall... maybe a bad rule in the ads blocking dns 🤷♂️
that's odd but not "important" XD
but regardless of that it boots and at what stage does it then reboot?
or never boots entirely?
are you the one that had issues with NTP?
that would also point to some network issue
nop rob :)
the device never boot fully, it start booting, i can see the service starting (it look like systemd, don't know if it's that or something else) it wait for network , network is found, it start docker, and then initiate a shutdown
so i can't get a shell
also the shutdown is way to fast to read anything
sorry if by booting you mean reaching kernel, yes it does, but if you mean reaching userland no it doesn't
i can't tell the nuc generation, pretty old
If tou look under, it should say the model number. But yeah, we are at nuc12 now, so mine is not new either
Nuked,reinstalled and restored backup
you have them offsite i guess ?
i am still setting it up, so right now there isn't backup outside of the nuc
so ... some live os, copy the backup from ... don't know where on the ssd to a usb key, reflash, and restore ?
If you boot live, the folder is easy to find
does backup include everything ? (config done on other docker app)
Yeah, got it back to how it was before
ok
Would be cool to find clue tho ... avoiding this happening to other
Tried seeing the error, but even if I record the screen is so fast its blurred, and reboots
yeah, that's maddening in that video - it blurs at the critical time
let's try a slomo then XD
i have the screen at like ... 30% because i have furniture over ...
and that damn kvm never got html5 update
i can see a faild to start for the docker engine
do you know if there is any flag to add to the kernel line in order to stop the automagic reboot ?
that way i can try to look into it
I thought it would roll back after some small number of failed boots
3
but I thought generic now uses GRUB and always shows you an option to choose the slot no?
similar thing here, but no clear solution: https://community.home-assistant.io/t/hassio-wont-start-failed-to-start-docker-application-container-engine/95407/25
yeah, that's what I was searching for
I guess that's not very helpful
sudo nautilus certainly doesn't sound like something that would exist
also 2 years old
Yeah so now all UEFI platforms use GRUB instead
so it should be showing the two boot slots on startup
the real issue isn't docker engine or whatever failing but the fact it's rebooting itself
or in one's case shutting down even
In my case, I did see the two boot options, but both were doing the same thing. Loop rebooting
and thats not due to docker not working? i mean what would haos do if docker doesnt work. nothing?
As far as I remember it keeps trying. But if anything restarts the service not reboot
so worst case it hangs indefinitely trying to start it
I'll go fire another VM and see
I'm definitely not going to reboot my production now 😂
Maybe it'll work better the second, third, fourth, or 10000th time...
But if its TasksMax=infinity does it actually fail fail to then reboot?
i have the 2, the 2 does the same
it's been rebooting for 9h
sooooo
10000th
It never stops, I remember leaving it overnight, and in the morning was still looping
no reason not to XD
you could try adding systemd.journald.forward_to_console=1 to the kernel cmdline, maybe you can get a glance at the actual error docker is having
or boot with systemd.mask=docker.service and then wait some time and it should drop you into an emergency shell
ls
then you can look at the journal for docker with journalctl -u docker and that should hopefully turn up whats going wrong
my position to use the computer is crazy bad, so i'll try to keep it short but i will try to boot with the mask option
kepping you posted
moving my install from rpi to a intel 3040. When waiting for the 20 minute setup it locks due to low disc space. I'm using a 500gb ssd, but when balena flashes the disc it makes 8 minor partitions, any ideas on how to distribute more?
it's supposed to expand on initial boot
ah, then it makes sense on my last try. I made a regular partition. Ill try again
Ah no just flash the drive and boot it
it should do what it needs to do itself
you made a 9th partition huh?
Yes I did, but reflashed just so booting as we speak
Ah yeah that would explain it
i cheated but i have this
I have the system.journal from a live os if it help (sorry i slipped)
seems the file content is corrupted for some reason
is the filesystem full or damaged?
or maybe docker crashed while writing it or something. i beleive its regenrated on startup, you could just try removing it
i could only find a key.json in the "hass-overlay" partition, and it look empty
there is a lot of free space, like 200GB at least but i don't know which one is important
and i can garenti they wasn't any power outage, this i in a half high baie with a server grade UPS
the file is actually full of \00\00\00\00\00...... a very long line of it
nothing more
file have been removed, and the device boot
we've got CLI !
since i have all the log from journald, i may be able to locate what went wrong in them, but given the number of boot, that would be some digging XD
And we are live on Proxmox.
can't find anything to odd durring upgrade but that don't mean i know what i'm looking for ...
I will try upgrading again from 8.1 to 8.2
upgrade went without an itch
i need to go to bed or i'll be dead tomorrow, but please @ me with what you think i should do next after work.
This look to not be a one off, i am technical enough to manage my way around with you help but hassio isn't made with people like me in mind...
For what's worth I tried the qco2 image on my proxmox. Bunch of restarts. Went back to 8.1 and to 8.2 again and nothing broke
given that i didn't find a thing in the update log, my guess would be something else broke the file before, and it just didn't manage to reboot
also auto-reboot if docker don't start isn't a good idea IMO, a shell is beter than a full reinstall
read ya tomorow
getting back to this, I have Virtual Box and Ubuntu installed, did make it to the end of https://community.home-assistant.io/t/installing-home-assistant-os-on-a-mini-pc/314722/43 but then upon restart it fails, because it cant find a boot device.. I did selkect the HAOS boot partition before clicking restart obviously...
ofc, the link is for a real pc, not a virtual box, so that will make a difference.
Got it working after some other issues I managed to resolve. Thanks for the help! Sometimes it just helps to get some conformation in order to get going
And yes... As everyone has stated before. rpi is way slower than a computer
Hehe
also depends on the pi 😛
PI zero can go make dinner come back and will still be installing
I've used a rpi 4 4 gb 🙂 Not that big of a deal, but it loads way faster, for instance grafana
But honestly, I'm so amazed by the developers. It's so easy moving a whole installation or when an SD is corrupt. HA ❤️
I wad running on an Rpi4 for several years
moving to proxmox is day and night
but also entirely depends on how big an installation is
True, my whole backup was 450mbish. How is proxmox setup? desktop?
See that's nothing. I had ended up with 2000 entities, a 4GB MariaDB and 2GB InfluxDB 😅
Backup alone took hours
hoho! Damn
I got a HP Elitedesk
installed Proxmox and everything in VMs or LXCs
but at the moment Rpi4 install is still running in parallel
okey.. First, 2000 entities? I got 400. All the lights in my home, half dozen ESPs. How? 😄
I ask that myself tbh 😂
So proxmox > generic x86?
OVA image essentially
the generic is specifically for flashing on a drive and booting
For proxmox is called qcow2 I believe
yeah iam running mine on proxmox too, just so much better to manage than physical installations
it has been so much better
like earlier also so easy to spin up another instance to test
I'm keeping the RPi4 too though for the Zigbee stick at the moment as it's at a better location than where proxmox is
I could see that, since I had to reflash 2-3 times. Troubleshooting network issues I have to connect monitors and keyboards, all that would be easier with VM
but me being lazy I just disabled all automations and kept HAOS running as is 🙈
hehe, I see
I never had network issues but e.g. making backups would lock up the db till I'd restart
that also being a personal issue having such a big db in the first place though
Ah, that's not really an issue (yet) for me
I have to say, even if I've had to fresh start my HA multiple times during the last month, the backup restore in HA is top notch. Kudos
VMware workstation seems to work
hey uhhhhh question: I just moved into a new home. Before I moved I installed HA on a RPi 4 and logged in, just to make sure it was working properly. Now I'm in the new place, and I try to access via Homeassistant.local:8123 and it doesn't see it... RPi is connected wired to the modem/router via a network switch. Do I need to reset it somehow since I logged in in a new home?
@placid sage try the ip_address:8123 instead of homeassistant.local:8123
no such luck 😦
IP may have change ?
So either look into the router to see what is connect or not.
Or plug a screen to the rasberry pi to see the Io written t in the terminal ?
I'm not sure what any of that means...
Probably because I can't speak proper English 😂.
I can access my internet router using 192.168.1.1. I suppose you have something similar at your new home ?
It's probably mostly because I am ignorant about this. I have a Google Home router. Typed that in and nothing happened
you have a phone app for your google router i assume?
Okay I don't know of this specific one work. Using the app Google home you should be able to see all device connected and their IP
how did you know what the ip address was when you tried it?
you didn't literally try "ip_address:8123" did you?
🤔
yes
does it list all devices which are connected?
Yeah I don't see the Pi on there...
hmm, yeah, I'm not sure how to do that. It's connected wired to a network switch.
and the light on the switch is flashing like it communicating
Check all connections by disconnecting and reconnecting at both ends of everything. Weird stuff happens after a move
Is there a way to reset the Pi without being connected?
pulling power is the only way if you don't have a keyboard and monitor attached to it
But really, don't think about it, and disconnect/reconnect both ends of the ethernet from pi to switch, and from switch to router. Just in case. Wait and see if it shows up, and then if not, pull power on the pi, wait a minute, then power it back up
"shows up" in the app, I mean
Yeah I've already done all that, unfortunately
no, I just turned it on and logged in and then put it away
ok so you set this up, did basically nothing, then moved
yes
without access, it's hard to guess what is going on
ok thank you, I will figure it out
so it's either get access, or reflash
I have a Sense energy monitor running in my home and it is now connected to the Home Assistant virtual machine. It is taking a long time to get the status/ analog readings to get reported to the Home Assistant interface. What will be the reason? #986757441985323059
I'll reflash it, seems the easiest option
no trabajo 😦
I will but it's wired. I even connected it directly to my laptop and no dice.
I was saying that because I specifically remember people coming here with Google Wifi and having that exact issue of not showing up on the DHCP
connected what to the laptop?
The Pi
that anyway wouldn't have worked
that's for connecting the laptop to ethernet. It doesn't just work if you plug a client to the port
🤷♂️
as in laptop wifi --> etherner port --> magic 😛
Hmm and where would the "network" be coming from in the scenario? 🙈
Try rebooting the Google Wifi device in case that works since it worked for the last person that had exactly that issue and go from there
I remember it so vividly because 3 different people on the same day came here with Google Wifi and that issue 😄
rebooting didn't work. Adding https didn't work.
Am I supposed to reboot all of them? I have 2
the one that acts as you main device
plugged in to a modem or the wall to your ISP
https wouldnt' work but what are you typing exactly if you don't have an IP?
(maybe I missed it this is a fresh install right?)
What is says in the tutorial: homeassistant.local:8123
well it is now
Also not coming up in the app
Focus fist on getting it to show up on the app
if it doesn't show up there with an IP the homeassistant.local will never work
So you setup is Google Wifi and then you mentioned a switch and the Pi on the switch?
By router you mean then main google wifi device?
yes
Well now the Google WiFi thing stopped working. I'm thinking I need a new router.
How did it stop?
To be honest I don't have the greatest of experience with them based on people's feedback
they are good for silly networking
How was it connected to the network before?
Flashing red light. No internet access. Plugged in another router and it did the same...
okay reset the modem and got internet back. Plugged in a different router with Pi plugged directly into it and still no dice.
Hmm not sure anymore
without a monitor on the Pi as well to make sure it boots fine it's difficult
Was it working before on Google Wifi?
Yep
I have a docker image of home-assistant I have been using in bridge mode no problem. Today I decided to use host mode, and the 8123 port is not being listed with netcat -an on the host when the container is running. I docker compose exec hass-server /bin/bash and curl localhost:8123 and it works. Is there some issue with my /config/configuration.xml http: key?
http:
use_x_forwarded_for: true
server_port: 8123
trusted_proxies:
- 127.0.0.1
- 192.168.0.0/16
- 172.0.0.0/8
This probably has been asked several times before and hopefully has been answered several times before. I've installed the latest version of HAOS on an USB ssd drive connected to a Pi 4B (8gb), It boots normally and I can access the dashboard. I can also see that Supervisor has been installed too ... but the supervisor menu is nowhere to be found 😦 I searched the internet for answers and checked the menu structure over and over again but I cannot find the supervisor option. Am I overlooking something or should I reinstall HAOS?
The Supervisor menu is no more since 2022.5 (or maybe even .6).
You'll find the Supervisor Logs under Settings > System > Logs
You'll find the Reboot Host under Settings > System > Hardware (3-dot menu)
..and so on.
@modest lake Thank you! One last noob question; how to install the user/community add-ons which are not in the standard list?
@viscid spade If you are only seeing the Ofiicial Addons and not the Community Addons you need to enable advanced mode (it's under your profile settings --> bottom right on the circle)
is host networking non-functional in docker for mac?
can we still add the usb option for config and authorized keys somewhere?was on the supervisor panel before.. (to shell in with 22222)
hi everyone, i have been run HA on an Rpi and recently had to upgrade to an SSD, i lost all my backups and now during the initial install process i get an error message stating that the supervisor failed to get update and the system is unhealthy...... any clues as to why??
this is the exact message in the install logs
22-06-12 09:10:21 INFO (MainThread) [supervisor.jobs] ResolutionFixup.run_autofix ' blocked from execution, system is not healthy
Thx
it runs in a vm, so if you use host networking it will open the listening sockets on that vms ip address.Pretty sure thats not accessible by anything external to your host. so host networking is kinda useless on docker for osx/windows
was looking for this https://developers.home-assistant.io/docs/operating-system/debugging/?_highlight=22222#home-assistant-operating-system wasnt in my memory like that.. but still possible 😉
Thanks I had it responding on ip:8123 but might have been bridged mode
I've given up on host mode on mac and think I'm going to migrate to a machine running k3OS for better docker support
I installed home assistant „the easy way“ (HAOS) on my rpi4. How do I get to run a fsck on the next boot? touch /forcefsck won’t let me ( read-only file system )
no idea, cant you just run the fsck while its booted?
Hi all, not sure if this is the right channel for getting help with the installation
sorry, installation for the nest integration into hass where I'm using duckdns and ssl to get public access to my instance
If you look at the top of each channel (or right panel on mobile) each channel has a topic describing what it's for
this one is for installing home assistant
no, just says that it‘s mounted
what are you trying to fsck? / is a squashfs image you should be able to run fsck on /mnt/data and /mnt/overlay but why do you even want to do that?
do you see any evidence of filesystem problems?
Lots of
EXT4-fs error (device sda8): ___ext4_find_entry:1612: inode #131493: comm python3: reading directory block 0
and weird behaviour of home assistant from time to time
Running on an SD card?
hi @high token, @storm sluice and @wide wind
Sorry for the global ping, but i wanted to inform you that the docker bug is a closed bug from January : https://github.com/home-assistant/operating-system/issues/1706
A little help to Open this back up with a comment would be very nice :)
That issue is open
did i miss a thing ? https://www.imgpaste.net/image/K15qrs
OOOOOKay ... login off made it open ...
I'm not going to question that
i'll edit the comment in that way
damn ! loggin back in it's still closed !
It's you
no, i refuse that, i will say thank's microsoft !
anyhow, i'll update my comment
uploading picture is a reserved right, right ? i had to find a hosting website, but you didn't have to, that's not me being dumb ?
Correct
The advantage of being a mod
The downside is having to deal with the tsunamis of spammers
No, an SSD 😦
Worked for 3 months without problems, though
Hey all, my home assistant on my raspberry pi I think died and I want to restore from a backup. I have a bunch of .tar files on google drive from using the auto backup integration but I want to make sure I restore correctly.
Right now, I can't access the home assistant dashboard from the network, and alot of the guides i have seen use the dashboard to restore.
What is the "correct" way to restore a backup by just copy/pasting the google drive backup onto the sd card? I hope to not lose any of my settings with my backups and I want to do it correctly.
you would need to reinstall home assistant fully first, the button to upload a backup is on the first setup page when you start home assistant for the first time
Thanks! Just wanted to make sure there wasnt an easier/safer way first before formatting the corrupted sd card
Be warned, if the SD card is corrupt it may also be failing
Reformatting it may get it up and running, for a few hours/days/weeks before it becomes worse
Thanks for the heads up! If it turns out to be terminally corrupted, should I get a new sd card or a new external ssd? (or is it just personal preference?)
SSD > any form of SD, but there are better and worse SD options
Industrial SD cards should last many years
SD Cards for high endurance - like those designed for dashcams - should last years
Everything else is a roll of the dice
Nice...
thanks for the help/advice!
good thing, i got to test sandisk warranty system, and they are nice
Hi guys
Could someone help me please? I have installed HA onto an SSD and used a SD card to install the boot from USB onto it.
I power the PI up but cannot connect to it through http://homeassistant.local:8123 or http://homeassistant:8123 I have also looked under my devices in my router and cant seem to find it. Any suggestions?
Have you connected a keyboard and monitor to it?
No just power and ethernet
That'll allow you to see what's going on
i just cloned my previous sd card onto the ssd, booted directly from the ssd, no sd card installed, with no problems since 3 months, until today.
Is it actually recommended to run the os on an SD and only move the data disk to an ssd instead?
Pure SSD is fine
At least, if you have enough power for it, and it's not a crappy one
in my virtual HA environment in virtual box, Id like the Hostname to change to ha-virtual and click save. restart, and I still see homeassistant.local:8123 as the address...
should I do something extra?
its just a green screen
Question, once I have inserted the sd into the PI with the USB boot file on it can I then remove that?
If you connected the screen after you powered on the Pi you'll need to power cycle it
I took the SD out and now I have a wall of text with just the SSD in
ha host reboot did the trick..
is the text moving? 👀
can i send a picture?
imgur
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.
Command '.images' was invoked by: Thanasis
can we use another image hosting platform it isnt sending me my verification code
Hmm I would just try reflashing the SSD and see if same thing happens again
So I do not need to keep the SD card in?
No the SD card is a one time thing
you used it for the bootloader right?
not that you flashed the SD with HAOS
I installed HA on SSD and then Installed USB boot on the SD card
ok then yeah the usb boot sd card is only needed once
you did that first though right?
without the ssd
it probably was fine though or it wouldn't have been seeing HAOS stuff on the screen (even if with errors)
You think I should just install both again and put SD in first then turn off and ut SSD in
I'm not sure how recent the image for HAOS is in Pi imager so try instead using Balena Etcher like mentioned here https://www.home-assistant.io/installation/raspberrypi#write-the-image-to-your-boot-media
You can try again with the SD it won't hurt if you want
either it will do nothing cos it already did what it needed to or it will do it this time
it should only take a few seconds
PI imager is fine actually
I just saw on mine it's OS 8.2
If it still doesn't work I would go with HAOS on the SD card then use datadisk to move HA into the SSD
feeling a bit silly as I cannot find a way to shutdown HASS gracefully on a RPI. Is there a power off button in the UI somewhere?
-> System -> Hardware -> 3-dot
Soo on my old install all of a sudden I was unable to access the add ons menu. I wiped and on a fresh install fully updated I still can’t access the add on menu. Everything else works. Running on rpi4 4gig.
Full log says error main thread frontend.js.latest.202206010 :0;0 script error.
so I moved to a Mini PC with HAOS install without too many issues. and yes it is faster, but tbh I am somewhat underwhelmed. it has taken off about 2% of my processor and still takes over 2 minutes on startup (was 4). Is this what can be expected from a 8GB/128 eMMC pc?
since I am on the move, I mght want to up the ante a bit more. should I add an SSD and move the DB there (had that on the RPi4 too), or should I up the processor to get most out of HAOS. (now celeron J4125)
Heya,
having a slight issue with trying to update my Home Assistant Operating System.
Am stuck on version 7.6 trying to update to 8.2, everytime i click the install button it does some work, restart my pi and then tells me i'm still very much on v7.6. Was there an upgrade path that needs to be followed?
evening folks. I'm trying to install home assistant as a docker container on my synology nas
I have other containers running fine, but whenever I try to download the home assistant image it gets most of the way through and then fails with no useful error message
it spools up to 367 MB downloaded, sits there for a few minutes and then fails
I'm attempting to use "homeassistant/home-assistant:stable
Hi i'm totally new to Home assistant, and want to try it out. I saw somewhere that i could install hass on an old android phone. (samsung galaxy j5 2016) just to see if hass is something for me. But i havent bin able to install it or find i guide that works. please help
There's no such option, at least not supported
Some people have hacked it, but it's a really terrible idea
why so?
Because:
- Mobile phones aren't servers
- Mobile phones use WiFi, which isn't great for a server
- See (1), no really
okay, can i then use a pc (just to try it not a final platform)
Yup, lots of people use those for their install
thanks i will try to fin a guide the
thanks
it says download the appropriate image, im not really familiar with VM's witch is best to choose
Well, what OS are you using?
windows 11
VirtualBox is probably the easiest option (at least, it's free)
Yes
windows host?
if youo have a computer running HA OS is there a way to create a new user and not yeet all the data and already saved connected things like lamps and plugs and such?
cause i replugged my old HA but it doesnt take any password or username
i read this documentation yesterday, it does not tell me if yes or no i loose the data of all the already connencted things
You won't lose anything, it's just about resetting a password, not the whole computer
It doesn't mention anything about losing data since it'd make no sense to mention that
so i need to disconect the thing, get into the drive with my normal computer, yeet the files that they specfiy on the site, and then when i remake an account and the connected devieces will still be on it?
so if you start a new onboarding process, do the already connected devices still connect or do they all need to be reconnected?
i cannot do such a thing
Why not?
'If you know the username, but not the password and you can access the Home Assistant console and use the command below:'
i dont know the username
how please tell me
Look in .storage/auth_provider.homeassistant
so i need to connect the storage of my HA machine to a computer? or is there a way to look into it directly via console?
You can do all this in the console
so i should type the file location in the console?
or is there a specific command i should use?
The file location, which is somewhere under /usr/share/ I think
If i type it into the console it gives me an error
@viral marlin Please don't say I have an error or describe an error. Share the whole actual error message so we can help you.
Command '.error' was invoked by: Tinkerer
You can find the config folder at #installation-archived message
Ha > /usr/share/
Error: unknown command "/usr/share/" for "ha"
Yeah sorry i dont know
Im just trying to get it work :(
Type the command I provided
Then you can cat auth_provider.homeassistant to see what's in the file
It tell me cd is an unknown command
Ah, type login
It changed from ha> to a white hashtag
Now try those other commands
I did this one, it gave no output so i think it worked
Now the other one
I did. Euhhh
I think something is wrong
The password is like 25 characters
And the username is admin
Well the password is not in clear
Now you know the username, and can reset the password
Oh of course!
Thank you a lot!!!!
Euh, if i try the command given on the site it says
/bin/ash/: auth not found
Logout gives me the same answer
Exit worked !!!
im now makeing the vm. Using the HA guide but i have trouble with step 3
What trouble?
it cant find the existing virtual hard disk file
Did you download and unzip it?
yes i downloaded it from the top of the page
And unzip it?
yes


