#installation-archived
1 messages Β· Page 12 of 1
I've been attempting to install Home Assistant on my home network with no luck. I've tried inside of a VM and on baremetal and both give me the same repeated issue . I googled around a little and I believe its either a DNS issue or MTU issue. I tried setting a static DNS to 8.8.8.8 but that did not fix the issue. Any help/suggestions would be greatly appreciated.
22-12-20 21:38:43 ERROR (SyncWorker_2) [supervisor.docker.interface] Can't install ghcr.io/home-assistant/generic-x86-64-homeassistant:2022.12.7: 500 Server Error for http+docker://localhost/v1.41/images/create?tag=2022.12.7&fromImage=ghcr.io%2Fhome-assistant%2Fgeneric-x86-64-homeassistant&platform=linux%2Famd64: Internal Server Error ("Get "https://ghcr.io/v2/": net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)")
22-12-20 21:38:43 WARNING (MainThread) ```
Do you mean you tried setting static DNS to 8.8.8.8?
Seems to work! Thanks much!
correct
Glad to hear!
I've not experienced that personally, but I've heard the same that it's generally DNS resolution issues
its likely its something with my router/switch etc but am unsure what or where to start
I've usually heard that being a problem w/ Pis though.... not VMs
The Pi is the only thing I havent tried. It seems to be a somewhat regular issue
For the 8.8.8.8 DNS... did you set that on your router to hand out via DHCP or another method?
I set that in home assistant, using the command:
ha dns restart```
I dont know if that is the same DNS server that is set in my router. I can check and see if it is if you think it needs to be
Is Ipv6 possibly enabled and your network doesn't have that configured?
ha network info
no, ipv6 is not enabled
i can see its trying to use ipv4. I can also access it via the web interface. It seems to be able to ping both 8.8.8.8 and www.google.com so it looks like it has internet
really weird, never had that problem
i wonder what causes this for all those people that seem to have that
<insert DNS haiku>
its either resolving the name to wrong ips or there is something blocking the https connection
maybe, just im just using the firewall on my router
so basically block all inbound
allow all outbound
Haha yes lol this is indeed mtu in this case
Iβve seen this issue with too small windowss
Some domains work fine while others just seem to not respond data, just like refusing a connection
odd how would that ever be a problem with an isp router
Yep weird, last time I have seen such thing was when whole sites where connected and internet was routed via vpn
any idea how to fix it?
Temporarily change mtu in your router see if it gets things unstuck? Do you have any special settings active?
No special settings that I know of. What are you thinking I change it to? its at the default of 1500 right now
try setting a smaller (e.g. 1420) and then change back to 1500
ok. ill try that and let you know. Thanks for the help
restart your router in between
mstone was it you that recommended I use swag?
may have been rob or petro I cant remember now
Hi, now I have a solution, the SSD was damaged, I bought new ones, restored backoup and everything is OK π Thank you
Yes
hi. I'm trying to restore a backup of my Home Assistant in a fresh installation on a RP4. After choosing the backup tar file, the page shows this:
(it's showing "Upload backup" with a spinning wheel)
after this, it asks me if I want to restore a full or partial backup. I choose full and it shows only the page to create a new account..
is this doing the restore in the background in some way?? Or am I having an issue here?
this error ocurred before I asked here. now it says:
Upload failed: Failed to fetch
and now I don't have access to my <ip>:8123
Usually the recommendation is to set up a new install, install the samba add on and then upload the backup through that and restore it. The upload backup is restricted in size.
Hi, How can I upgrade python from 3.9 to 3.10 in a development environment docker running from vscode? I'm running the "dev container for HA" for developer purposes and there is a warning indicating that I'm running on python 3.9
Hello. I have an issue, trying to setup HA on raspberry 3B, but getting these errors. Can undervoltage cause this? because I use 2A 5V power adapter but as I read it's recommended to use 2.5A. Thank you!
22-12-21 17:33:22 ERROR (SyncWorker_2) [supervisor.docker.interface] Can't install ghcr.io/home-assistant/raspberrypi3-homeassistant:2022.12.7: 404 Client Error for http+docker://localhost/v1.41/images/ghcr.io/home-assistant/raspberrypi3-homeassistant:2022.12.7/json: Not Found ("no such image: ghcr.io/home-assistant/raspberrypi3-homeassistant:2022.12.7: No such image: ghcr.io/home-assistant/raspberrypi3-homeassistant:2022.12.7")
22-12-21 17:33:22 WARNING (MainThread) [supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30se
Well, the lack of power probably isn't the cause of that problem, but it's likely to cause you a lot of other ones
404 Error for docker usually is a result of one of these things:
-
The image tag you are trying to use does not exist (if this is the case you will see the word 'manifest' in the error)
-
There is not enough space left on the device to download the new container.
-
There was a networking issue preventing the image from being downloaded.
Either you're running a too small SD card (minimum is 32 GB) or you've got a DNS problem
I am using 32gb SD card
Can you tell me more about DNS problem? How should I check? I'm very new into this π
Yes, HassOS. It's the same right? Where should I find log file? In SD card?
Yes, and you can find it on the console, but also via the UI (click the blue circle if it's still setting up)
Hello all,
does any1 have any idea why i cant connect to home assistant gui in browser on default port 8123 but I can ping the machine?
Hello everyone, I have a question. I installed HA on vmware with the .vmdk, I get an IP address and when I go to the IP in my browser with :8123 the page cannot be displayed. (I have a HA in virtualbox and it works fine) I want to switch to vmware to fix issues im having but for some reason i cant figure out this seems to be simple step...
in vmware I have bridged and NAT networking setup correctly with other machines so they get internet and my HA gets the correct subnet range on bridged... I cant connect to home assistant gui in browser on default port 8123 but I can ping the machine?
yeah it will be assigned another IP
get an IP scanner free one, and you can find it that wauy
my local is 2, but vmware uses 30
my HA has an IP address and i can ping it
Try port 4357 instead of 8123
on the main HA page what does it say the IP is ?
do I have access to my CLI my vmware one my ip address ipv4 is 10.0.0.21:8123 I have also tried port 4357 and i also tried the homeassitant.local versions
my home subnet is 10.0.0.0
It is 10.0.0.0/24 and not 10.0.0.0/8?
its a /24
can you show the vm settings network pane?
you said vmware
yes
the settings for this virtual machine, especially the network preferences would be interesting
network is bridged, Its workplayer so I have checkbox replicate physical network connection state checkec (prob doesnt need it but..)
network editor -> VMnet0 subnet address 10.0.0.0 subnetmask 255.255.255.0
vmnet0 is ?
the default bridged vmware connection network
maybe there is your issue, usually you don't need to set a adress / mask
actually i see my vmnet0 is bridged then set to my network adapter dhcp is - and subnet address is -
but under machine settings it shows 10.0.0.0/24
Heh, stumbled upon some interesting bug. When onboarding HA presented me with some devices and integrations it had found (HomeKit, Roomba etc) and I set them up as far as possible. Then there was a button "More" that I was silly enough to tap. It dimmed the web page a bit as if to pop up some dialog. Nothing ever happened and there was no way to cancel. So, I tried to reload the page and now it won't accept the credentials I created at the start of the onboarding. Not a particularly smooth experience.
I guess I'll have to reflash my SD card and go again.
i should be able to have 2 instances of HA on my network no issues right? just diff ips they would be
Yes
ill shut down my prob version and see if that fixes why i cant see my vmware version, my prod is on virtualbox
prod*
If it does solve the problem then you've got something really broken with your VM setups
lol
i have other VMs running and i can connect to them and have internet on them
so i dont think this will fix my problem but ill try it out
haos is what I have on my vms right? i downloaded the .vmdk
yes, use that .vmdk as harddrive
yes i did that and its running, has an IP and i can ping the IP address and verify it on my router dashboard that its Home Assistant
just i cant access ipaddress:8123 or :4357
my interface says enp2s1 instead of what i normally see eth0 does that matter
can you connect via the homeassist.local
nmcli:
ip4 default
inet4 10.0.0.21/24
route4 default via 10.0.0.1 metric 100
route4 10.0.0.0/24 metric 100
homeassistant.local goes to my prod server my main one thats running
Hi everyone. Homeassistant hangs itself once a day. The frontend is no longer accessible, but it shows up in the Fritzbox and I can ping it too. What can be the reason for this? System is running on a rasp pi 4 with ssd.
will running the supervisor install script reset all my HA configs? I actually haven't ran the script since my initial install over a year ago
There's no supervisor installer script, at least not an official one
It's a deb you install
yeah that's the one i'm talking about. Just rather deal with not being able to access host logs and the like rather than reconfiguring everything. Although I could just pull a backup before running it if that's the case which is why I ask
I'd take a backup... if you've left your install to rot to the point where you want to re-install then there's potential for a lot of things to go wrong
If your install pre-dates the new way of doing Supervised there's also a high chance things will break horribly
literally the only two things are systemd journal and resolved not actually passing health check that actually matter.
Only other repair checks that fail are unsupported software in other containers and putting it on a different OS which I won't do.
so I just went through the script and found the commits for this service when it was added to the install script and how it's configured with is non-standard which is why all the repair doc's have you rerun the install script, just manually making those changes.
so after shutting down my prod version of HA i still cannot connect on my 2nd instance
any help with local tuya, I got the tuya working, but when I use lcoaltuya it keep asking for the cloud api config ?
integrations would be beter
according to the YT vids, when you add localtuya it should just find them on the network
k
Hi, i will install in Debian 11 supervised. But have a Warning that βld-configβ is not in PATH . Same with Start-Stop Daemon. Anyone who can help? Found Ni Solution in Google.
Consider running HA OS or HA Container instead
Going back to this error, I tried setting the MTU to 1420 with no luck. Still errors out when downloading. It is possible to skip this initial download so I can poke around the settings inside home assistant?
you can download the images and export them with docker save from a machine with internet
then docker load them on the haos machine
Ok Iβll try that
did you try setting a known good dns server? like googles?
Yes. It currently has 8.8.8.8 as its DNS
so if you run systemd-resolve --status|more on the haos console it shows its using that as the primary dns server?
Correct
Odd. What Type of machine is this running on?
any1 have any idea about an issue with my vmware HAOS vm, I can ping it, its valid on my network but i can not access the GUI on the web browser, :8123 or :4357 and the ip address.
An older intel NUC. 4gb RAM, i3, 256gb SSD
And you use Ethernet?
Yes
Whats the Error you get?
I have a strange occurrence that just happened overnight. I moved my HA OS instance to a NUC6i5 on Proxmox about 2-3 weeks ago and everything has been speeding along until today. I noticed I had a couple of updates and tried to install and I get an error that there's no internet. I reboot HA and get the same. I reboot the VM and it hangs at the time sync thing and then continues to the emergency console. I exit out of the console and Supervisor loads. I get the UI on the IP:Port that is setup but still no internet. I triple-checked all firewall rules on pfSense and even added a couple to see if it would help but no luck. I can ping google from the ssh add-on. Nabu Casa is checked in and working and so is cloudflared tunnel. I just can't get the updates to work.
[3:43 PM]
oh, and Eufy doesn't connect (not a big deal) and neither does the SmartThings app for my Samsung TV expanded info (not a big deal either).
i got this warning "System is unsupported because the wrong version of Docker CGroup is in use. Use the link to learn the correct version and how to fix this." i just did a fresh install to move to Debian 11 , it's not clear to me from the Learn More section how i change this cgroup (is it because i did a full restore ?) also no addons want to start now
Are you running Supervised?
yes
They could be running a docker container π
Which would be much easier to support
System is unsupported because the
Is clearly Supervised or HAOS π
it was a supervised install
You really should use HAOS instead
The only time Supervised makes sense is when the hardware can't boot HAOS
i run on an old laptop
@storm sluice i dont get any error. I just get nothing on web page, page cannot be displayed
maybe when i restored something was changed
shouldn it be correct since it was a fresh install ?
i am running HAOS and page cannot be displayed from my VM, but the machine can ping and i can access CLI and it gives a valid IP.
@sage orbit a valid IP in the same range as everything else on your network?
yes
same subnet on a /24 all thats is correct, it is pingable and on my router dashboard it shows at HomeAssistant just like my prod version, I have a prod version running in Virtualbox, and now this is vmware, I am workin on transfering my HA to vmware but this beginning issue i'm having
does it work from another device?
umm ill turn on another vm and see if i can ping it
its pingable on my kali vm
im checkin open ports
still cannot be opened from a different machines web browser
I have a strange occurrence that just
if you do curl localhost:8123 on the vm console, does it give anything back or also give an error?
good idea lets see
doing an nmap scan none of the normal ports are open like my prod version
if i could send a pic in here id show you but i only have 2 unknown tcp ports both closed state and on my prod version I have over 10 open ports
so could the new download 9.4 of HAOS have something wrong with it?
dont think so, check if its running any containers
and if not check the docker daemon logs
docker ps? command in CLI
yeah
hello all. I have installed HAOS to raspberry pi 4 and followed the steps on the installation page. After install completes I am unable to access port to begin with onboarding. the browser gieve a "IPAddress" Refused to connect error.
@keen hornet sorted? Sometimes needs reboot. Login and check logs
Accidentally overwrote my laptop with Hass lol
ooops! need to read the dialog a bit closer.
Just to post for a quick FYI, first time setup is missing a unit on elevation. Am I defining this in meters or feet? I'm assuming meters based on what I remember from last time I set the elevation on my HA but π€·ββοΈ
I assume it's in meters if your unit system is set to metric; and ft if it's set to "us customary" (heh)
Having a quick look around the code, I can't see that it's used much.
Doesn't it auto populate according to location? I could be wrong here
but I think sun relies on it
I can see that the parameter is passed, but couldn't follow how it was used within that.
And a couple of integrations
I think it's only metadata for sun stuff
eg
if event not in ELEVATION_AGNOSTIC_EVENTS:
kwargs["observer_elevation"] = elevation
try:
return cast(_AstralSunEventCallable, getattr(location, event))(date, **kwargs)
actually "met" does
if not self._is_metric:
elevation = int(
round(
DistanceConverter.convert(
elevation, UnitOfLength.FEET, UnitOfLength.METERS
)
)
)
of course, air pressure levels for weather forecasts
that code implies that it's in meters/feet depending on your metric/us setting
how to increase the swap for HAOS?
Can anyone help me install on an x86 system?
It seems the USB boots up but won't install to the system
oh nvm i think i understand...
there is no installer
Holy fudge!
It was taking 25 min to "boot" with my usb.
Took 25 sec to boot with my ssd
lel
I pray that this setup is easy enough i can manage remotely and be fully robust. Wifi switch to power cycle
Hi! Might be a stupid question, but i seem to cant find it in the documentation.. whats the default backup location on a HomeAssistant Supervisor installation? I know its in the root folder of the installation.. but i cant find it..
/backup or /backups ISTR - though if you mean where on the disk that's different
Thats what i thought.. but i dont have a /backups folder
i got a find running now on "configuration.yaml" xD
It's /usr/share/hassio/backup
Thank you very much!
My homeassistant lost its ip config for some reason.. i cant send a screenshot here.. My network manage config is gone.. tf
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.
Are you using HAOS or Supervised, or something else?
Im using Supervised, i think my Network Manager config is broken
Ah... you fell for the trap
hahah, i restored my config and it works again
What do you mean?
Supervised ... it's a trap for the unwary
The only time it makes sense to run it is when you're an expert with Linux and Docker, and can't boot HAOS on that hardware, yet really want add-ons for some reason
In other words, it pretty much never makes sense
It's fragile, restrictive, and ... not recommended by anybody involved in its creation or continued existence
I wont call myself an expert with Linux and Docker, but i can handle them fine. I like to option to install addons π there might be something fun π
The only thing i get out of your message is, if your lazy dont pick this one.
As long as you're complying with the requirements, and keeping current with changes, it shouldn't break
I agree π Its not a plug-and-play version π But i like the "control" i have
Most people who run it deliberately ignore the requirements, then complain when things break
Mistake number one.
And yet, with Supervised, you don't get control...
ok
Hopefully you did find and read https://github.com/home-assistant/architecture/blob/master/adr/0014-home-assistant-supervised.md
Yes. thats why i picked Supervised.
@normal cedar 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).
thank the Discord devs for that π I just turned of Discord notifications untill they fixed that tbh.. your the first one who asks me to turn it off π sorry man, just not used to it.
i dont like Discord in general..
You can't turn it off and be done, you have to do it every time, or use something like Better Discord to disable it
Hi! I'm new to this, having problems, I want to install home assistant on an old laptop, I have booted from a live Linux diatro, downloaded balena and a correct url for x86/64 and targeted my internal drive. When I then reboot without the USB in and only get a flashing marker and nothing shows up in my network. I have waited hours but nothing when i search in my network
Do I need a bootloader?
And in that case, witch bootloader is easiest to install and install the ha distro on?
Does that laptop support UEFI boot?
I think so yes? How can i check this?
I wanted it to be easy, just boot the live Linux distro and then just install and use the laptops internal ssd as a big fat drive
can i run home assistant on a pi zero?
@fierce laurel no
Pi2, technically, but you'd be daft to do so
Pi3 is the functional minimum, and even that's pushing it
so basicly a pi 4 that is almost impossible to get hold of today.
You could do a Core install of HA on a Zero, it'll probably run... if very slowly, and as long as you don't enable too many integrations
any alternetives?
Just about any PC from the last decade
as of now i only have 2 shelly swithces. the shelly plus 1pm
If it has at least 2 cores, 2 GB of RAM, and a CPUMark of 1,000 - and supports UEFI - then it'll work
would love to test them out π
Run HA in a VM then
Hi, I just did a fresh install on a RPi 3b but getting this when trying to reboot. Any ideas?
cannot import name 'check_config' from 'homeassistant.helpers' (/usr/src/homeassistant/homeassistant/helpers/__init__.py)```
Did you flash the right image to the Pi?
Hmm, I went with the 64-bit, maybe I should use the 32-bit? Could that be it?
I'm running it from an SSD
It could also be power issues if you didn't use a powered USB hub for the SSD
Just running usb power from the pi, using the official power supply
But previous owner of the pi has ran Debian OS with that setup so I doubt that a bit
I wouldn't
I've seen many complaints from people who switched to using HAOS and found that their existing power supply wasn't enough
Alright, thanks. I'll try the 32 bit first but otherwise external power to the ssd. Thx!
Maybe this isn't the thread for it (although its a consideration for my installation) - could I run HA from an SD card and direct logging (and possibily backups) to the connected SSD - to reduce write wear on the SD card?
@hearty hatch 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).
Nice, thanks! I assume logging is one of the most wearing tasks for the sd card within HA?
Well, that and the database for history
Actually, installing the 32 bit version instead of the 64 bit seemed to have fixed the issue! For now at least.
After you install it to the SD card, you can use the Move datadisk feature to move most of the data storage to the SSD
https://www.home-assistant.io/common-tasks/os/#using-ui-to-move-the-data-partition
Hello I am about to get started with Home Assistant and have bought a NUC for the purpose. I want to run it in a docker, and i was wondering if i could ask some questions about the OS setup. So its not exactly HA related, is that allowed here ?
Hello all, so this must be something with my vmware, I am creating a new VM with the haos.vmdk when i run it I get an IP and it says :8123 and CLI but when I actually go to it and i run a port scan on my ip no ports are actually open..
vmware workstation, I am using bridged
its bridged to a ethernet port right?
is it on windows? do you use any weird security/firewall software?
i have vmware on windows yes
its a vm on my pc
i have other vms running and everything works fine
so I have a virtualbox HAOS vm that works fine its currently my prod HA, I now have vmware I want to switch all my vms to vmware, and this with this one I do my normal process create a vm, then attach the .vmdk as a hard drive to that vm start it up, it gets an IP, on my router dashboard it shows as Home Assistant just like my virtualbox vm but when i go to the ip:8123 i get page cannot be displayed, when i do an nmap scan on this IP the ports like 8123, 4357, etc are not open. and on my virtualbox vm all those ports all open
Have you tried logging into the HAOS console in vmware to make sure the services are running, check logs, try pinging out, etc?
i can ping it I also do have access to the CLI so i can check the logs, but where are the logs at?
i have not checked services
If you run ha at the CLI, you'll get all the available commands. There's ha core logs, ha supervisor logs, etc
ha info will show you whether things are running and so on
ok
ha core logs shows everything started successfully no errors in that
is there a way I can send a snippet pic to this chat?
ha supervisor logs
last lines
starteding system autofix at state CoreState.RUNNING
system autofix complete
updated Home Assistant API Token
yeah so it says homeassistant is running
and other vms in vmware really work? like you can access services running in them?
Hello I am about to get started with
so heres an update, I just tried the same .vmdk in my virtualbox and it works!
why can I not get it working in vmware.
i can access my ubuntu vm in vmware yes. and access internet, etc updates
wait are both running on the same computer?
yes
Oh, depending on the type of bridging VMware does, you may not be able to access it
I have also tried shutting off my prod instance of HA and it still did not work
they both install some sort of filter drivers to do the bridging stuff
it gets the correct subnet on vmware and I can access other vms just not my HA and they have internet everything is bridged
Have you tried from another computer on the network?
i had to change my subnet because it tried to give me a 192... and im using 10.0.0.0/24
i logged onto a different vm and tried to connect to it, I can ping it but cannot access web UI :8123
change where?
change the vmware virtual network editor at first internet did not work but after changing settings I was able to setup bridging and NAT to all my VMs in vmware
why nat? if it uses bridged ethernet?
im not using NAT anymore but i just wanted it working
i only use bridge
on default install of vmware no network connection was working
but after changing the virtual network editor i was able to get everything to work
looking at netstat -tunlp on my HA it shows the correct ports :8123 for python3 is open but doing a port scan on the vm from a different machine shows its not open
From the HA console can you ping to the internet?
umm
so i tried ping 8.8.8.8 and it said its alive but i tried pinging local ips on my network and says no response
sorry for long delay went and grabbed lunch
Is HAOS on the same subnet as the local IPs on your network? Kinda sounds like it could be a routing issue, maybe subnet mask, something like that
yes its on the same subnet
my subnet is 10.0.0.0 and it has the ip address 10.0.0.21/24
when i go on other VMs they get an IP in the same subnet and can access internet and ssh to each other
so heres a weird thing that i do see
when i do the command ip a on my HAOS
it shows inet 10.0.0.21/24 brd 10.0.0.255 on interface enp2s1
is brd suppose to be netmask?
brd is the broadcast address and that would be the correct broadcast address for a /24. It really sounds like some sort of vmware-level network config issue rather than a HAOS one, especially since you had it working on virtualbox
id still think its using nat or something, or something else is messed up with the vmware networking
so running ip a on a different machine its all the same settings
it shows inet 10.0.0.21/24 brd 10.0.0.255 on interface ens33
this makes no sense to me, it appears all networking is setup correctly the machine is pingable on the CLI it shows netstat
0.0.0.0:8123 so the port should be open but doing a scan from the outside onto the machine it shows no ports open but then creating the same VM on virtualbox it works and can connect on :8123
why does that other machine have the same ip?
i copied the line it actually .26
it shows inet 10.0.0.26/24 brd 10.0.0.255 on interface ens33 for a ubuntu server of mine
and that runs in vmware too? and you can access its services? like ssh?
yes
im double checkin to be 100% sure
so ssh did not work!
so its gotta be networkin on vmware
yeah mostlikely
thank you for lettin troubleshootin this with me!
now i gotta figure out my vmware network
its probably due to the nat network you changed
if it uses the same ip range your actual network that will fuckup the computers routing table
ok damm ok when i setup vmware no internet worked at all
How do i use HA OS?
My time seems warped, but 20 minutes is lasting 12 hours.
[supervisor.jobs] 'Updater.fetch_data' blocked from execution, no supervisor internet connection
22-12-22 20:27:59 WARNING (MainThread) [supervisor.homeassistant.core] Error on Home Assistant installation. Retry in 30sec
@crude inlet yuo
yup
THats how i can view it in the web browser π
This was so much easier in Unraid :D\
Probably DNS then
hmmm oki doki
ill look into that thx
maybe a pihole issue
@crude inlet looks like when my cat was laying on my phone, she accidentally clicked to pause internet for HA
π
What? lol
My dog is so much easier than your cat. Never messes with the internet
Though they do sit on the remotes, changing the input for the tv while I'm watching something
I had google wifi open to find the ip last night, and she plopped herself down. π
She is loved. keeps my hands warm in the winter
[supervisor.jobs] 'Updater.fetch_data' blocked from execution, no supervisor internet connection. Check for cat.
I need some help getting HA installed on Unraid. I had it working month ago and then something happened and it stopped working. Lots of troubleshooting later now I am just remaking it, I have a recent backup so it's not that much of a loss. I'm trying to make a VM to run HA and it just doesn't want to work. I'm using the .qcow2 file from the HA site and its currently in a boot loop and ends with "unable to read id index table". Ideas what I can do to get past this or should I use one of the other files?
(note: i'm not a support person, im mostly clueless)
If you're in an Unraid enviornment, why are you using a VM instead of a docker?
I saw at one point that running HA in a VM was better than running it in a docker
It's not, to be clear.
Each installation method has its own advantages and disadvantages
I think it was something to do with add-ons? I think that might have been why I went to a VM and not a docker
Any ideas what might fix my issue?
id think the image is broken or something like that
Add-ons are docker containers π
Trying to install HA on an intel nuc. Already used dd to write the decompressed image to my nvme. Unfortunately my bios does not pick up the installation. (uefi boot box is checked in bios) Any idea what I missed?
sure you did that correctly? pretty sure the guide recommends something like etcher as it also verifies the image was written correctly. any reason not to do that?
@vast geyser sounds like you might try again. VM in Unraid should work fine
you can also boot to ubuntu live desktop and use the built in disks utility instead of etcher
but I don't think there's a cli etcher
dd should be ok
I'm on limited bandwith here. Downloading trillion gigabytes for a simple dd alternative is somewhat not an option. π
how did you run your dd ? any special bs or else?
dd if=haos.img of=/dev/nvme0n1 bs=1M
So just download the qcow2 file again and try again?
fdisk -l showed my a handfull of partitions afterwards.
thanks for your help, by the way. Much appreciated.
not sure, sounds ok. try a smaller bs of 64k maybe ?
I've seen it matter once, really just a random guess
π
last version i tried was 9.2 and it worked fine. I can try to confirm with the latest
@vast geyser you used haos_ova-9.4.qcow2.xz and not one of the other qcow files, right?
not aarch64 or generic x86-64
Was I supposed to unzip the file that I got from the website? Cause I did. Yes that looks like the file name
making sure. I wasn't paying attention once and used one of the others mentioned
Mine worked fine with 9.3 (proxmox + qcow2 file)
I didn't even get that far
It get boots to slot A, goes through some of the stuff there and then gets stopped saying "unable to read id index table"
Are there settings in making the VM that I should change for Unraid?
@vast geyser I shared my settings in DM
Getting this error in my supervisor after trying to restart HA:
[supervisor.misc.tasks] Watchdog found a problem with Home Assistant API!
Then it tries to restart the container and then a few mins later I get an error saying "tried to kill container, but did not receive an exit event". Then supervisor logs:
Home Assistant watchdog reanimation failed!
I've just restarted the host in the past, but wondering what's causing this.
Here are the last logs from HA:
s6-svwait: fatal: timed out
s6-rc: info: service legacy-services successfully stopped
s6-rc: info: service legacy-cont-init: stopping
s6-rc: info: service legacy-cont-init successfully stopped
s6-rc: info: service fix-attrs: stopping
s6-rc: info: service fix-attrs successfully stopped
s6-rc: info: service s6rc-oneshot-runner: stopping
s6-rc: info: service s6rc-oneshot-runner successfully stopped
I just bought a SkyConnect stick today, my HA is running in Virtualbox.
I'm a complete starter in this topic and got stuck at the step where i expose the USB to the VM. Tried filtering and attaching it, but i always receive error that it is already in use in Virtualbox.
Figured out i should do this step, but cannot get the "docker run ... --device /dev/ttyUSB0:/dev/ttyUSB0 ..." command working in HA CLI.
Could you help out a beginner user in HA pls? π
That step is not for HA OS
What?
could you let me know how or where to enter that command to enable the stick in the vm? π
Like I said, you don't use that command if you're running HA OS
oh
It just needs to be passed through to the VM. Unfortunately I don't have virtualbox experience so someone else will need to help you there
after 2 hours of googling i didnt come to any solution, thanks for saving me another hour by cutting that topic short though π
You'll have to look in the virtualbox documentation, but it's usually called USB redirection or USB passthrough or something similar
yes, already configured that
Ah okay
enabled passthrough, set it on usb 3, set up filter for stick
that part is done, i get the error after this
From what?
@thin gorge You said "Tried filtering and attaching it, but i always receive error that it is already in use in Virtualbox." Does that mean the attach was not successful?
i believe it wasn't since it does not show up in ha at devices
virtualbox returns the error when i try to attach it:
"Nabu Casa SkyConnect v1.0 is busy with a previous request, please try again later"
Something else might be using the USB device
Do you have more than 1 VM on this box? Perhaps another filtering rule attached it
running virtualbox in win server 2016
also tried disabling the device in windows
didnt help
yeah, sorry... π
Dunno if that command works on windows. I mean it might
Or you can go into the GUI and it should probably show you attached USB devices
according to virtualbox the usb device is captured, but not attached. when clicking it i get the error above
let me check
if you open a shell in windows can you run vboxmanage?
I run it headless on Linux so I'm not as familiar with the vbox UI
not recognized unfortunately, but thanks for trying to help me π
same error with a flashdrive, i guess i have to go after virtualbox, not ha
probably windows sits on the usb
Have you tried unplugging it and plugging it back in after redirecting it to the VM, so that Windows doesn't have control over it anymore?
if by redirecting you mean the filtering then yes, several times
@thin gorge try USB 2 instead of 3
let me try both, ty
@queen bronze
the showinfo command did not work, however i've used the "vboxmanage list usbhost" command and got this:
It is captured, but not attached, upon attaching still get the "skyconnect waiting for previous request" error.
UUID: d78fd4f3-2382-46d8-80a9-cd9b9ffa888f
VendorId: 0x10c4 (10C4)
ProductId: 0xea60 (EA60)
Revision: 1.0 (0100)
Port: 1
USB version/speed: 2/Full
Manufacturer: Nabu Casa
Product: SkyConnect v1.0
SerialNumber: 64659376ab14ed11a52cb68be054580b
Current State: Captured
@cinder marshTried switching it back to USB 2, but results with the same error: "skyconnect waiting for previous request". Cannot attach usb π¦
ok solved it, there is a workaround for running homeassistant in virtualbox in windows and ataching usb stick.
i just leave it here if it would be needed π
https://www.virtualbox.org/ticket/14287
Hi guys, I'm investigating my options for installation and I was planning on doing the hyperv HAOS option. However, it has a warning that hyperv doesn't have usb support. 1) Is USB typically needed for HAOS installs? 2) Is that a Hyperv limitation or a limitation of the particular implementation?
Do you have anything that you need to connect to via usb? my haos doesn't use any.
I mean a ZWave or Zigbee radio would be one common device you may add to HA OS
@soft hawk limitation of hyperv
im not faminiar with docker and my installation seems to be not working. container keeps restarting (using docker contaner ls -command to check). How can I look "inside the container" to see what is going on?
Check the HA log (and container logs) to see what's going on
If the container is restarting it's likely because HA is crashing
thx.. its "ValueError: source code string cannot contain null bytes"
There should be more than just those half dozen words
It could also be that you've got a corrupt image, you could try stopping the container, removing it, removing the image, and pulling again
there is python tracelog with some indications to config.py tracing to auth package. anyway having some logs gets me bit further
ah.. with docker compose without -d i can have it log to stdout
Without -d it runs in the foreground
with docker container rm it removes the container runnig. but how can i do the container "building" again?
docker image ls -a to see all images and then docker image rm IMAGE to remove one
that is fetch the packages etc
ss
aa
thx
container = something running, and image = environment
Pretty much
Think of the image like a little file holding the software that you're going to be running, because that's really what it is
thx
does docker have separate repos for docker compose, and docker run type of execution? just wondering because i did docker run first and tried docker compose type starting later
No
same images, just different type of config input..
Yup
thx
Docker manages images the same no matter how you interact with them
pulled all new and no errors anymore
I seem to have borked my install upgrading to 2022.12.8, and got to the point where I'm trying to re-install (Supervised on RPi4)... I removed all docker images and directories, updated the OS and isntalled all the pre-requisites. But when trying to run the install packege, I get this generic error: "dpkg: error processing archive homeassistant-supervised.deb (--install):
new homeassistant-supervised package pre-installation script subprocess returned error exit status 1
Any ideas where I can find out what this actually means? Is there a log somewhere?
I think itβs tradition in this channel to ask why you arenβt using HAOS, especially on a pi 4. (The updates are image based so you donβt get this kind of borked half upgrade, and the actual requirements for running supervised are large enough and change often enough that you are almost always doing something βwrongβ that could brick an update).
If you're using Supervised so you can run something outside of the HA add-ons, that's going to cause you problems and is one of the things the requirements tell you that you cannot do
But that error means a script in the supervised package itself returned an error code. Itβs probably made an assumption about your host os that isnβt true, otherwise everyone would have hit it.
I've had it running fine for a couple years. Reason I didn't want to install HAOS is that the Pi was already used for other things, namely a Unifi controller I didn't want to kill, backups from a PC and video hosting.
Ah, so I was right π
The extra processes haven't been a problem up to now. I just wish there was a way for the install script to tell me what its problem actually is.
You could ask in a Debian support channel, there might something in the dpkg or apt logs in /var/log
βUntil nowβ is a bit of a wrong way to think about it. If you had a car with faulty brakes that failed after 3 months and it was a design defect, the car manufacturer doesnβt say βwell itβs been fine until nowβ. Though Iβm sure theyβd try tbf.
I have zigstar LAN gateway on /dev/ttyUSB0 but im failing the new device installation. under what user the docker instance is running? with the one i started "docker compose up" with, or root when using --privilege settings? Is there a way to test the functioning of my device beforehand using telnet to /dev/ttyUSB0 or something similar?
If you're running the official Docker image, it runs as root
roger... so its not my /dev provileges then
GM! Have been using HA for a couple of years, though recently it crashed. I acquired an new microSD card, installed a fresh install, and at the screen to "Create Account", I am trying to load a backup from the old microSD card in my Ubuntu laptop with Firefox. After selecting the backup file, it gives me the error "Upload Failed - 401 Unauthorized". I have a screen up with the RPi HA CLI screen, HA OS 9.4, home assistant core 2021.12.10
I did the basis RPi install, not the HA container or Docker Compose
Or should I just create a fresh new start altogether (uname/pwd/etc) and then restore from backup after that?
Am I not asking questions in the correct manner?
afaik the recommended way to restore on haos is with the samba addon
restoring bigger backups likely fails at the http upload in the onboarding process
I just tried addons info samba, and it says Addon samba does not exist
my backup file is 13.9MB
thats not too big, no idea why it fails. someone else can maybe help you out
I'm following the instructions;
"Restoring a Backup on a new install
You can make use of backup which you have copied off of a previous install to restore to a new installation during the onboarding process. Follow the link at the bottom of the account creation page to upload your backup from the previous installation."
In your position I would set up throwaway uname/pwd try getting the Google Drive Backup add-on from HACS, upload your backup to Google Drive, and use the restore feature within the add-on.
Ok, thanks, I'll go that route, just got the call for breakfast, will attempt after that
The add-on defaults to a folder called "Home Assistant Backups". You can make that preemptively, and the add-on will ask "Folder already exists. Do you want to use it?"
Thanks, that's quite helpful
Hope you've managed to get things sorted.
Ok, after a trimming of my configuration.yaml of a custom panel I'd had in there previously and a couple of reboots, the backup seems to have worked after all, even after the 401 error.
Thank you @placid sage and @low notch for your helpful assistance. Happy Holidays!
Hi, something is failing during installation on a RPI3
Is here the right place to ask ?
yup
Well I have 2 kinds of errors that occur. Both errors happened on 2 different RPI (both are 3B), different SD cards and 2 different power supplies. So I hardly believe it is a hardware issue.
First problem was (and it occured several times) that is got stucked on wait 20 minutes page for more than days (both were LAN connected)
and tried both 32 and 64 bits version.
Second problem was that 2022.11 was not able to upgrade and left with an error ( for the few times problem 1 did not occur)
there is no Internet connection errors, many PC and TV are connected to the internet
you need to click the blue dot to view details of what fails.
possibilities are:
- DNS fails
- Power Supply is too weak
- Raspberry pi dead π
well RPI, both are live and kicking.
Power supply is OK (I have SD with "bullseye" installed, with no power supply notification showing)
be aware that HAOS demands a lot of processing power and most folks having trouble in combination with a pi don't have the proper power supply.
(what i want to say is that debian may very well run and do things with that power supply - but haos goes on full draw, starts docker and pulls images -> which often fails catastrophically with these PSUs)
Are either of the power supplies you're using the official one?
could totally be the same reason
π«
I am attempting to complete a fresh install and when the software boots there seems to be no ipv4 and ipv6 addresses. I have a ethernet cable plugged in and I have attempted to set up my network wifilessly via a boot file (CONFIG/network.my-network.txt). I have also attempted to turn on the method eth0 (name of my interface) with the command: "net update eth0 --ipv4-method auto" or "net update eth0 --ipv4-method static" the result is this prompt "Error: Can't create config and activate eth0: Connection 'Supervisor eth0' is not available on devcice eth0 because device has no carrier". If anyone has any suggestions fire away.
seems like it does not see a cable (if there is one plugged in)
weird...
i figured it out it had something to do with the case I was using since it didnt have direct connection to the ethernet port
thx for your help @low notch
Hello all! I'm trying out Home Assistant for the first time and downloaded the 64-bit Raspberry Pi 4 version and I'm having some issues. When I power on the Pi on a clean image and go to homeassistant.local:8123, I see that it is completing the initial configuration then once it's done, it continues to onboarding and it breaks. The error I get is "Something went wrong loading onboarding, try refreshing." I've of course tried that but the error persists. For grins, I looked at Java console and I see errors about the access to a script has been blocked by CORS policy.
Hello, I wonder if anyone can shine some light on my issue. I installed Home Assistant Supervised on Rpi4 64 bit Bullseye and I have 3 issues with Unsupported system. Apparmor issues, Network Manager issues and CGroup version. I have tried to follow steps to fix this from different sources. I reinstalled everything maybe 10 times and still the same problem. Home Assistant seems to work ok, but I don't know if its ok to have those issues open.
Are you 100% certain you installed debian (not any debian derivatives either)?
Also what is the point of running supervised on a Pi? If its to run other software then its not supported
100% ? I'm not sure now when you ask, how would I know for sure
Well it depends on how you installed it...
but again, why are you installing supervised
because of add-ons
You are supposed to know whats needed to run the hardest installation method, its in the requirements (of the hardest install method, again)
yes I went through it many times
So you are a expert now?
You can achieve the same with HAOS
you mean the full OS?
Yes, the OS install is easy on a Pi and has the exact same abilities as supervised
With far less maintenance required by you
Yes, and it also takes over the the whole Pi, right ?
if I can manage it, yes
You cannot do that with supervised. if you read the requirements it says that nothing other than HA or add ons can run
You can use docker yes, even on the Pi. You just won't have add ons (but add ons are just fancy docker containers anyway)
That's what I was doing with Pi, installing with Docker
No you said supervised
both run in docker
yes, supervised , in docker
Yes, but ultimately you're running supervised. Which as per the requirements dictates the device is dedicated to HA
Synology will not let you run supervised on there unless it is in a VM
People run it that way, don't they?
Or HAOS in a VM rather than on bare metal
Eventually my main install will be on Synology, my Rpi setup is mostly to play with it and discover things, but once it is on Synology I'd like to have it supervised. So you're saying it will have to be on VM.
he says you should go with haos. that's what I'm trying to tell you also
how do you install haos on Synology ?
there's literally no difference from haos to supervised from a consumer perspective,
except you have to do A LOT more maintenance work with supervised
In a VM
you install haos in a VM
Also given you wanted to run other programs other than HA on a supervised install tells us you did not read the instructions and requirements of supervised
that's still an assumption though π
this is a bullshit warning you can easily just ignore, lol I get HAOS not letting you but supervised is a piece of software, not an OS, so it can coexist with anything else on the system
Except there are plenty of people who come to this very discord with broken installs because they installed other things that conflict with HA
The simple answer is just run a container install if you want multiple other things to run
Add ons are just containers after all
You do not need a supervisor unless you want click button go, in which case HAOS is probably more useful
Except there are easy fixes for those issues and becuase you don't want to actually troubleshoot anything you try to say it can't be done, when in fact it can
I was doing a clean install on a new Debian OS, nothing else was on it yet.
And I asked was it actually debian or a pi distro of debian
hassio died a long time ago π
you literally get the same warning when you install that whole thing
And supporting multiple people who are trying that is a lot of effort, given most of them have never looked at a command line, and have just clicked a tutorial link and followed instructions blindly with no clue what they have done
can you give me a link to "actually debian" and not distro? I'll tell you if that's what I used
why can't you just cat /etc/os-release and tell us?
I mean, we can take the additional requirement that was added on in 2022.11 of systemd-journal-gateway service needing to be configured and ran for unix sockets, that broke so many installs because the upgrade didn't mark them as requirements, nor did it update the supervisor to actually bind to that service on the host.
Maybe i'm just salty because I finally sat down and actually figured that one out by actually having to apply the updates to supervisor manually because I didn't wanna rerun the install.
except you have to do A LOT more maintenance work with supervised
PRETTY_NAME="Debian GNU/Linux 11 (bullseye)"
NAME="Debian GNU/Linux"
VERSION_ID="11"
VERSION="11 (bullseye)"
VERSION_CODENAME=bullseye
ID=debian
HOME_URL="https://www.debian.org/"
SUPPORT_URL="https://www.debian.org/support"
BUG_REPORT_URL="https://bugs.debian.org/"
that is debian
There is "official way" to install Supervised , but it seems to be broken somehow, I followed every step.
All dependencies were there, everything as per instructions.
the "official way" is exactly as hard as the "unofficial way"
and the reason why people chose haos over supervised
https://github.com/home-assistant/supervised-installer
and I tried some other ones as well
Alright, I just saw your original post
Well that is the only correct install guide
<> can you run a docker info, you're looking for cgroup version mine looks like this
Supervised is the expert installation method assuming expert knowledge of linux and docker
Logging Driver: journald
Cgroup Driver: systemd
Cgroup Version: 1
Should be a bout midway down
Again, these kind of feature updates shouldn't requiring a full reinstall and feature updates like an additional service requirement should be included in the updates made between version changes. It's just lazy maintenance. Anyone else running updates with new requirements can mark those new dependencies. No reason this should be any different.
I think I'm done for today, just way to tired at 4am
All good. for those repairs the gudies they give are pretty decent starting place, then you can use "ha su info" "ha res info" to help dig in a little deeper.
@upper talon Can I ping you another day for some help maybe ?
inbox is open
I have no clue on how it works. I stray away from supervised as I don't see the need for it. Container install does it all for me with limited effort needed
I have just one little intel NUC running ubuntu that runs pihole, plex, nut, ntpd and Home assistant all in one power efficient package. If I want or need something else along with that I can just drop something else on there or spin up another docker container. I don't need dedicated hardware for each service which is expensive to procure and expensive to run.
You can do all that without breaking the requirements that supervised needs (eg nothing else running and requiring debian 11 not ubuntu) just by using containers
Which you seem to already do anyway, don't need the extra hassle that supervised brings
I have Zigbee2MQTT, MariaDB, Mosquitto, HA, Traefik, Authentik, Frigate, and a couple other things all running on containers on a single headless debian install
This whole thread on "supervised" may be worth a documentation clarification on "supervised". I read it and was confused about the linux/HomeAssistant definition of supervised versus the general definition of supervised. It seems myself and the OP assumed that "supervised" meant that it doesn't take over the whole box. However, Home Assistant Supervised (IMO--free to correct me if I'm wrong) is about running Home Assistant AND providing the supervision. If you plan to supervise using something else (docker, supervisord, etc.) do not use supervised?
The docs for HA Supervised says that it manages the whole system: "The Supervisor is not just an application, it is a full appliance that manages the whole system. It will clean up, repair or reset settings to default if they no longer match expected values."
I'm looking at the docs for HA Installation: https://www.home-assistant.io/installation/ which doesn't mention it at all. While we do preface that supervised is for experienced users, it's not clear why or whether a general user is "experienced enough" to use it. If not at the top, we should look at the "Compare Installation Methods". By looking at the green checkboxes alone (I know, but bear with me)--supervised is the second best option. Many people tend to be scared at containers still, but this is where I think we could add some clarity.
The top of the Supervised documentation (which is on GitHub not the main site) does state what you need to be proficient in
And the Supervised sections on the website all start with:
This way of running Home Assistant will require the most of you. It also has strict requirements you need to follow.
Unless you really need this installation type, you should install Home Assistant OS (this can also be a virtual machine), or Home Assistant Container.
Step 1 of the install instructions is "First make sure you understand the requirements." which links to that Github page
Where did you see them mention VMs? 
that was an assumption with their previous posts
Can anyone help me with installation of GitHub automatic backup?
Hello,anyone.. is there a way for someone to help me setup HomeAssistant and add my multiple dif smart devices, and get paid to do it? over teamviewer or something.. DM me
I'm trying to update to 12.8 on Docker within Synology. Currently have latest tag selected, downloaded from registry, but when I start the container the About shows I'm still running 2022.9.2. Tried manually selecting the 12.8 tag and installing, but still nothing. What am I missing to check?
you should be grabbing stable tag and not latest, but it doesn't sound like that will make a difference
I can try that real quick and see, thanks
Stopped running container. Switched registry to download Stable. Waited for finished download. Matched settings of previous container with mount etc, which are just the defaults from https://www.home-assistant.io/installation/alternative/#synology-nas. Looks like I'm on Frontend 20221213.1 - latest now - but the settings all reset. I think cause a new container was created when I switched tags?
I would double check the mount
Oh, you're right. Older was set to homeassistant/config on added folder. new is just homeassistant and then /config - probably didnt see the ending when I originally set it up. And then realized I had done this a few times trying to set it up a while back while trying to learn docker. I think this latest release is a good excuse for a fresh start.
Can anyone provide a link to the last version of Home Assistant Supervised that could be installed on Ubuntu? I read Here: https://www.home-assistant.io/blog/2020/05/09/deprecating-home-assistant-supervised-on-generic-linux that "the archived repository is still there", However, I am not able to find this version. If anyone has the link I would appreciate it.
You shouldn't install it on Ubuntu that's not supported
Why do you want to install Supervised?
Hey, i got a usb stick with hassos for my rasp 4 which are no longer boots, not sure why, but the rasp just resets itself everytime. Got a new stick with a fresh hassos installation and it works.
My question:
Which files/foldes i have to copy from the old stick, do get my HA running again?
Ideally, restore a backup
Yeah, not sure if the backups are okay, got only one old full and 4 partially after that. They are somewhat up to date, but not really π
I saw many docker stuff, not sure how the OS and HA instance are separated.
Is there a way to use a fresh OS/stick with my old HA instance, without backup, just moving some files from the old stick to the new one?
If the data is corrupt due to a failing usb device your backups would be the best bet, otherwise yes some folders you may be able to copy (though seriously better off just doing a restore and copying the changes to configuration.yaml and other related things
Hmm okay, i will try the backups ty
Is it normal that it take ages to restore the backup? It's 600mb compressed and the stick should make 40mb/s.
You're running it off a USB stick (bad idea), on a Pi, uploading over the network, its slow.
You may just have to wait, or the 600mb file is too large
Yeah, already bought a ssd π
But it seems to work now. Ty again ^^
Would've just swapped to an ssd after a USB failure. They are very unreliable
Still not sure if the old stick is corrupt, will test it tomorrow.
Hello everyone! I have installed home assitant, in docker on debian. but now I cannot find the supervisor tab... is it possible to add it?
No. Supervisor tab is only on HAOS and supervised installs. However if you want add ons they're just docker containers that you can set up yourself without needing a supervisor & dedicating the machine to HA
because I wanted to install nginx, but I can't find it from integrations. and I can't find the file editor either. can I install it anyway? because I can find duck dns
You don't need a file editor. Just edit the config from the command line or SMB or whatever.
Ngnix isn't an integration, nor is duckdns. You can install both as docker containers I presume
There is a duckdns integration but naturally it doesn't do what the add-on does
Hey together, I'm new to HA and I have a little problem with my OctoPrint in HA.
The integrated webcam view in HA shows nothing, it's empty.
I already figured out that I can add an "image path", but that doesn't change anything.
The webcam stream works oc. I tried the paths for "stream" and "snapshot"
http://ipaddress:8080/?action=stream & http://ipaddress:8080/?action=snapshot.
All other information from OctoPrint are available in HA.
How can I fix that?
Could I get just a quick check-in on whether a "full install" on a Rasp 4 would still give me the most convenient type of HA? That's what it sounded like to me from all I read. Or is there a better supported install procedure/hardware available?
(Sidenote: With an install like that, I could still do other unrelated docker stuff on that Pi, right?)
Thanks!
HAOS is recommended, or container
If you want to install other things there are two choices:
- If they are available as #add-ons-archived, use HAOS and install them as such
- If they are not, install a decent Linux OS (Debian for example), and then install HA and the other things as containers.
What I do not recommend is installing Supervised: installing anything else will break the requirements for the install so you'll immediately be unsupported. Plus the requirements to maintain it are very strict (for a good reason)
Thanks so much! I thought HAOS was the same as supervised. Need to read into that then.
Supervised is HAOS but managing the whole thing yourself, so you have to make sure all the libraries are what is needed (not older or newer), etc.
It's only recommended if you're an expert with Linux and Docker, and even then you're almost certainly better off running with containers as you can do whatever you like π
Alright, thanks again!
I'd strongly recommend getting an SSD for it, either a portable SSD or even just an internal one with a USB-SATA cable. You can move your data (minus the OS itself) to the SSD so that you don't destroy the SD card too quickly
Thanks for that input! I hoped to use the SSD (have several old 2.5s laying around) for everything including the OS, not a good idea?
It's possible to install to and boot from a USB drive on the Pi, but not really worth the effort. You can just flash the SD card with the OS, then after it's installed there's an option to move your data disk to the SSD
Explanation why not to boot from the USB SSD: https://www.home-assistant.io/faq/usb_boot/ There's also a link there on how to setup the data disk
how can i delete my home assistant in the docker so that i can install supervisor home assistant?
what commands should I give in debian? just delete the folder?
I fear that the fact that you're asking that question means that Supervised isn't the right installation method for you
I wanted to try, after I reinstall the normal version. but how can i remove it all? via nano
Hi All - interested in some advice or perspective: I have a Synology DS1515+ currently running HA in docker and a Windows Core i5 box that runs Plex. Is it worth moving the HA to the windows box with Ubuntu? What do I get out of that over the docker container? Thanks!
is someone using the samba nas addon from Lucio Tarantino (dianlight) with a supervised install?
well if it is your only option?
It's never the only option
If you can install Debian you can run Container
If you can't install Debian... you can't use Supervised either
maybe i wrote it wrong
Still, if you need help with an add-on, the #add-ons-archived channel is where you want to be
HAOS: not working for me
Debian + supervisor + os agent: working good so far
i will give #add-ons-archived a try
Hi guys. I managed to install the docker container (official chart) on true nas scale, but launching web UI gives me a "refused to connect" message. Anyone knows why this is happening or had the same problem?
I am still quite a noob would be happy if someone could point to a solution. π
p.s. - my plex and photoprism containers open the web UI just fine however
why should noop use supervision?
Supervised, the install method: https://github.com/home-assistant/architecture/blob/master/adr/0014-home-assistant-supervised.md
The one about which the developers said:
This method is considered advanced and should only be used if one is an ***expert ***in managing a Linux operating system, Docker and networking.
Not be confused with HAOS, which also has the Supervisor
I wonder why people who report βhaos doesnβt work for meβ think that changing to a completely different way of installing will change anythingβ¦
hey i am having trouble getting home assistant to work i install it on my sd card and everything plug in Ethernet and the power but when i try to access it through my browser it says this site cant be reached i can see it on my network and i can ping it but i just cant access it i am at a loss
How are you trying to access it? Try using the IP and 8123 as the port
and http not https
yeah did all that still nothin i just decided to use a vm and it worked so idk
Hmm, I'd be interested in helping figure out why that happened, so we could fix for other people; but if you've reinstalled etc, the evidence etc is gone...
it was just getting annoying i tried googling solutions but non of them helped i might try again at some point
Fair enough. I've seen a little of this recently. It's possible there's some issue with the installer, or a common misconfiguration...
i know its not supported but could i use a r pi 2b with home assistant
Not really. It probably won't even start.
hmm, ok cause i just have one and was wondering what i could do with it
You could try and use it for Room Assistant, there are lots of other projects out there, but HA is much more demanding than other things designed for the Pi.
ah ok i will look into that
Hello, after the last CORE install, 12.8 my home not start anymore, any ideas?
red light and green blink
what version were you coming from?
Sorry, i found the issue by myself, boot card was scratched and loose "start" file, i figured out with "raspberry pi manager", I installed SD card boot and it repaired my trouble card.
hello
i'm switching from rpi3 to a mini pc with haos installed on proxmox vm. Is there a way to migrate without having to repair everything ? i'm keeping the same zigbee usb bridge
thx for your help
If you're running HAOS on both then a full backup on the pi, can be restored to the mini pc
oh great thank you
i was looking for internal db or something else that stored data and ieee address of each device.
but if a full backup work that's better
I'm downloading the HAOS on windows through a VM and am stuck on the "preparing home assistant" page, I checked the logs and the last message was about 25 mins ago as: 22-12-26 12:03:46 INFO (MainThread) [supervisor.resolution.fixup] System autofix complete
Since then as far as I can tell the OS hasn't done anything
As far as I can tell there arenβt any errors or issues
Update: looked at the core check command and got: processing... ERRO[0000] Unexppected server response. status code 500
@pure pendant I converted your message into a file since it's above 15 lines :+1:
ping www.google.com and ping 8.8.8.8 both work from inside the cli
Hoping someone can point me in the right direction. My HASSOS has been going down every day. Someone mentioned I hook up a monitor to see if anything is displayed but there was nothing at all showing up on the monitor. Where should I be looking in the logs for an error? When I pull up the logs in the settings menu I can't go back past the reboot I just did. I am running HASSOS on RPi4
https://community.home-assistant.io/t/how-to-get-to-your-log-after-restart-restore/387662
On a Pi that crashes I always first suspect the power supply. Make sure you have one of the rated specs, try an alternative if you have one, maybe also try another usb cable, the shorter the better. Check for low voltage warnings. https://pimylifeup.com/raspberry-pi-low-voltage-warning/ (I'm not sure if this warnings work with haos)
Hello
I've switched from haos on rpi3 to haos in a vm on a mini pc (proxmox)
Since i switched i can't access anymore my ha with duckdns.
I changed the subdomain on duckdns, and i changed my port redirection to the vm local ip address
I'm still getting ERR_TIMED_OUT. Don't know why :/
any help ?
and for the info to switch from rpi3 to the vm, i made a full backup of haos on the rpi and restored it on haos on the vm
Ok I was able to pull the logs and here is the first error I get for the supervisor: "Dec 26 00:33:46 homeassistant hassio_supervisor[471]: 22-12-26 00:33:46 INFO (MainThread) [supervisor.jobs] 'CheckDNSServerIPv6.run_check' blocked from execution, no supervisor internet connection"
I am trying to apply Operating system 9.4, but when it reboots it still has 9.3
Anyone know why Balena Etcher just opens as a white window in my linux live boot? I am getting so beyond frustrated with this whole thing and I havent even gotten Balena Etcher opened yet.
Hey all, I'm running a blue, and over the last month, every time I try and upgrade any component, it just spins and spins and spins. SSH won't start either, so I'm kinda stuck.
What happens when you connect a keyboard and monitor to it?
I can get into the terminal that way for sure. I think I have a container stuck somehow (portainer maybe). I'll get it hooked up now
Are you running HAOS?
First time installing haos on a nuc. Now its stuck on the preparing home assistant screen. How do I check logs?
That part can take a while
I am, I can get to the ha> prompt
Ok ill just let it go
Then you shouldn't be using Portainer...
But using the CLI you can update core, supervisor, and the OS and you'll get logs
Currently running a 'core update.' After a bit 'core check' came back Done.
Ok my mistake on the sub dns
Is there an alternative to Balena Etcher for a linux live usb install? Balena Etcher just opens to a white window and I can't seem to get it to work.
I've heard there is some dd command I could maybe use?
What OS are you running?
um. ubuntu? That's not helpful is it
I can do frickin anything on Windows machine, used to work IT support for 15 years, but my linux experience is limited to just this.
No worries, you can run dd
sudo dd if=<iso file name> of=/dev/<usb device> bs=4M status=progress
(likely something like /dev/sdb
it looks like I can right click haos_generic-x86-64-9.4.img.xz and choose "Open with Disk Image Writer", and then just write it to the eMMC SSD thats in the Intel NUC. Should I just try that first?
I had an old home assistant setup installed previously and it totally broke, so I used gparted a bit ago and wiped all of the partitions and just made a new one. So I could select that new 120 GB partition and just "Restore Disk Image" to that, right?
It depends what "it totally broke" means
I believe that I accidentally had some add-on recording all 8 of my Wyze camera's live feeds to the disk, without wiping the data when they hit 24 hours or something, and a partition filled up completely and so the thing just wouldn't boot
I tried to fix it and don't know anything about how to access the partitions or where what was stored or wtf "sda" is or anything so I said "fuck it" and left the thing powered off for over a year, and am just now getting around to trying to start it all over.
how do I update the operating system to 9.4 when the regular way fails
(ventoy, unetbootin and rufus are all alternatives to etcher if you prefer a gui) You should also untar the iso file 'tar xvf haos_generic-x86-64-9.4.img.xz'
I think this worked, btw. No etcher needed. I'm now accessing homeassistant.local:8123 from another network device.
I even tried ha os update and it claimed it was doing it, but it rebooted and I don't have 9.4
What does the log file say?
I agree, core update is still sitting and waiting after 20 minutes. Way longer than normal. Let it keep going or kill it and go to the logs?
the system auto reboots so any helpful logs are lost
How do you even have Portainer installed?
Probably through the old addon
I don't know that I do, I think I had seen a note about it screwing something up in someone else's configuration with similar behavior. I was hoping to I could get to the underlying OS and see what was or wasn't running through SSH to check. Since SSH isn't able to start I'm kinda stuck
Well if you're connected with a keyboard and monitor you're already there
I didn't think I could list all the running containers from the ha prompt. I thought that was separate from a 'normal' root login
you just type "login" on that prompt to get a root shell in homeassistant os
OK, I'm 30 minutes into a 'core upgrade' thoughts on killing it and looking at the logs vs. just letting it go?
Ok I need help, I'm trying to install HA in a VM but I can't get the vdi image to boot, I can only get to an efi shell? Any help would be great
(Or am I remembering the wrong way around that you have to log in to get to HA, I'm too chocolated out!)
Okay, I ended up killing it and if I do an 'os update' I get an error message 'OSManager.update blocked from execution, system is not running - CoreState.STARTUP'
Hey,
I am trying to configure OVPN from my HomeAssistant PC to my phone
ive did the following things
- configure ovpn server
- open portforward from XX.XX.XX.XX (HA PC ip) PORT XXXX
- export the ovpn file that was generate by the guide's script
but i am failing to connect to the OVPN. what I miss here?
Its been several hours, and I'm still stuck on preparing home assistant. I have a monitor and keyboard plugged into the nuc. What should I look at here to troubleshoot
Open the Home Assistant web page in another tab. The "preparing home assistant" doesn't ever go away on it's own, so you have to re-open it in another tab
Click the blue dot
Turns out that I was in the midst of this bug from last month and almost all resolved now: https://github.com/home-assistant/supervisor/issues/4004
Hi y'all! I am trying to install HASS OS on an old laptop and can't make it boot. I have flashed the file and getting this error: βnon system disk or disk error replace and strike any key when readyβ
Did you sort out the UEFI boot step?
In BIOS?
Yup
Yes
And secure boot is not an option on this computer.
It is a old netbook from HP...
This is guide I followed https://www.home-assistant.io/installation/generic-x86-64/
The guide says this: If the machine complains about not being able to find a bootable medium, you might need to specify the EFI entry in your BIOS. This can be accomplished either by using a live operating system (e.g. Ubuntu) and running the following command (replace <drivename> with the appropriate drive name assigned by Linux, typically this will be sda or nvme0n1 on NVMe SSDs):
efibootmgr --create --disk /dev/<drivename> --part 1 --label "HAOS"
--loader '\EFI\BOOT\bootx64.efi'
Text
Or else, the BIOS might provide you with a tool to add boot options, there you can specify the path to the EFI file:
\EFI\BOOT\bootx64.efi
I think this is what I have to do and I have booted with Linux lite but not how to this. I have next to none expercience in Linux. Is this command lines for the terminal?
Linux lite live...
From USB...
Did something (?) in a bootup menu before the computer booted HA. Was some options and menus. Not sure what it all meant. It says "Waiting for Home Asstant CLI to be ready".
And on my Android I downloaded the Home Asistant app and it found the notebook and connected.
Hi. I've started looking into rebuilding my existing supervised HA install because the base os is no longer supported (currently debian 10). I thought i'd see if i could start with the provided vm templates. I've downloaded the hyper-v vmdk and have it running. Can i now break out of the ha cli and get enough os level access to install additional stuff or is that not possible?
no
consider running HA Container if you want access to the host OS and the ability to run whatever other software you want
or spin up other VMs
ok, is there a description of the differences between container and supervised somewhere? i saw the pinned graphic but it's not particularly useful.
At the moment i need supervised in order to install the software needed to get usbip into the underlying os
HA Container is just home assistant running in a container on the host OS of your choice (preferably your preferred unix flavor)
Ok. based on that table for my use case i'd say supervised is the way i need to go again. The vm is dedicated to HA, the only additional stuff installed will be in order to get hardware devices into the vm so that ha can use it. For this use case using containerized lacks the supervisor which would handle auto updates which are desirable.
But updating a docker container is literally just docker pull, ideally docker compose pull
and if you really want auto-update which IMO isn't a great idea for a lot of things, there are programs like watchtower
Don't autoupdate, the one time you have a breaking change you will regret it
what i really want is the fully managed os version. but i can't do that while running under hyper-v because you can't attach usbip devices. superivised seems to get me the closest to that.
or HA Container
I've hit auto update problems in the past. One removed the sql-tds driver so i had to move my history from sql server to postgres. but that's my own fault for not checking the change log.
i'm surprised that this isn't a more common use case.
Again, HA Container is a thing
You don't need Supervised for anything that you described
Supervised requires you to be a Linux and Docker guru
It has extremely strict requirements which are about to increase in strictness, and if you do not stay 100% on the ball it will break.
Ooo I didn't hear about the requirements additions
This is how I update all my docker containers, its so simple to run
And this is what that script runs
Basically automated except I choose to update
i haven't found that to be the case. i've been running it for 2 years and have mostly just left it alone. Once installed and running per the instructions it has just worked until debian 10 was deprecated. rebuilding to spec with debian 11 as base is slightly annoying but it'll give me the same experience i currently have. ha as an appliance.
You know that the requirements change constantly right?
Supervised is the one most people shoot themselves in the foot with. Because they try to do something and it breaks badly
using it in container mode would seem to remove a major piece of functionality (addons) and cause me to have to use commandline ssh access to update it. that's less useful and more knowlege requiring from my perspective.
Add ons are just other docker containers, its not that hard to use a docker compose file with everything (and is usually a copy paste too)
Also updates are once, maybe twice a month
ignoring supervised also requires command line ssh access to keep all the dependancies up to date
i haven't needed to update os level dependencies so far. it may be that i've just been very lucky.
Then you haven't been keeping up to date
I use DIUNs, I think I can give Tinkerer the credit/blame for that!
https://crazymax.dev/diun/ This thing?
Yup
I'll take any credit, leave the blame for Tinkerer π
i suspect you're using a different metric than i am. I'm up to date on the latest HA. i'm not up to date on host packages. If i needed to be up to date on host packages if i knew what was needed i could get in and either update them or install them, but i haven't had a need to do so.
Well the metric should be "supported install"
for example ha is telling me that system resolved is missing or misconfigured. I could probably install it in debian 10, there will be a downstream apt package i'm sure. but i'd rather not since it's made clear that debian 10 isn't supported anymore. It seems better to me to update to debian 11 which is where all the testing is going to be.
Straight from the HA requirements
Looks like debian 11 was released in 2021, so you're well past the depreciation date
So the usual way? π
π
i am. and i've no argument with the deprecation. i'm a developer and i know the cost-benefit analysis that goes into such decisions. Until now my install has shown as "supported" so i left it alone. now it doesn't so i'm doing the research needed to identify how to rebuild. Part of that research is seeing if there's a better way than supervised since it's always been labeled as being advanced and not something people should casually use.
And we're telling you container is that method thats better to use...
You just aren't listening
It's more a "not something anyone should use"
There are very few legitimate use cases for it, and yours is not one of them
Maybe if that person hates themselves and wants to pretend they're hackerman constantly keeping on top of everything
based on the feature matrix and the description given here (i.e. that you can just use cli commands to install addons and update components) it doesn't seem that container is better than supervised. What are the technical differences behind your reccomendation? Is is mostly about os level dependencies as we've discussed or is there more?
Supervised requires a lot of effort to setup and maintain. And you have shown you're not reading the requirements at all
Container is a set and forget thing
Given you're a developer I'm sure you understand the hassle with keeping dependencies up to date vs something that is all handled in a single container
And debugging something that doesn't work is not something I'd put on my "I want to do this" list, as a developer myself.
https://github.com/atxbyea/docker-compose, an example from a grumpy but organised member of this discord @lavish anvil for his docker-compose files
I copy and pasted half of that to setup my stuff lol
Same thing π
i agree that it takes a lot of effort to setup. I haven't done any maintenance on it though and it hasn't proved to be a problem, as i said i probably just got lucky there. Apart from apt-get update and the occasonal new package install is there any other reason not to run the supervisor?
Aside from the fact it explicitly tells you not to do it?
They way you're running it (not doing any maintenance at all) is unsupported, so I'd consider that a downside
true, though it doesn't detect as unsupported. It probably should but i suspect it'd such a seldom seen use case it's not worth the effort to add that logic.
Since my current install is working i've got time to tinker. I'll try a containerized build and see what real downsides i hit. I'm not using a lot of addons but it is nice to be able to install them through the ui and not have to manually fiddle with docker.
You fiddle to set it up, but 99% of it is just read the docs and copy and paste the docker-compose. Then once the compose file is done do a docker-compose up -d
Then its just config, and set and forget
Bear in mind that the coming updates will likely change this.
I personally am a big fan of a single config file to get the containers working, nice and simple
I forget .env files sometimes lol
I have two env files, the shared one across all docker compose files, and the one per compose file
so is the reason that container version cam't deal with addons because there is no agent outside docker with the ability to create new containers? if so an agent that only requires docker on the host (not the other deps) could do that couldn't it?
How often are you installing containers that you need a GUI?
very seldom. i just intensely dislike interacting with docker. i'd much rather have something that knows more than me about docker do it.
plus it's nice to try out a plugin every now and again
A compose file is really not that hard to use
Though also, supervised requires a lot of knowledge of docker too
i'm getting the impression that the fact my ha has worked for as long as it has is miraculous.
Correct
Usually people come in here complaining around 2-5 weeks after installing
The fact you didn't touch it for 2 years including dependencies and the base OS.....
lets see, how do you get the install date on debian? is it a uname thing?
I have no clue tbh
I don't touch much on my servers except keeping them up to date. And docker is enough knowledge to make me dangerous
But its enough to have a stable system
2020-10-27 20:15:38.287137571 +0000 installer.sh so pretty close to 2 years now. It looks like i updated to a new agent about a year ago.
So i'm going to stop using the auto install update feature now.
what is the pros cons with installing HA over Ubuntu/Debian vs the HA O/S?
HAOS is effectively an appliance, it manages the OS, backups, etc
read up the last couple of hours for a long discussion.
By installing over Ubuntu/ Debian do you mean supervised or container or core?
But the basics are HAOS is an appliance. Container is the only other recommended install and thats just a simple docker container running
supervised
Do not go supervised
The fact you're asking whether you can install it over ubuntu shows you haven't read the requirements
Container and supervised are 2 different things
^^
Container install is just a docker container with just HA in it
Supervised is HAOS but with far more management needed
if you said that too start youwould have had me at more management
sorry it has been a hell of a christmas for me..... i have not been able to visit family since my furance died
im aware
Probably part of the issue is that Supervised looks appealing in the "compare installation methods" matrix because it's all the features of HAOS with the ability to manage the OS yourself, until you read the installation doc and see that it basically says "don't do this"
Plus installing anything other than HA or add ons = unsupported too
Which is another thing people get caught up with
Did you mean furnace? If so curious why that stops you from visiting family? Since I know it sucks to not see them
yes, furnace. i have no heat at my house and literally heating my house w/ space heaters for a few days. luckily they are coming tomorrow to replace the furnace
Wouldn't it have been a good idea to run away to family if they have heat? π
i assume the issue is potential burst pipes
Oh I forget that is a thing
But yeah that makes sense, sucks to hear it :/
and there is a 30% increase from last year (2021-2022) and then another 10% increase in 2023 from 2022
30% increase of what?
in $
Hahaha so if I'm currently using supervised, should I work on moving to something different?
Most likely, yes
I personally think most (non wizard) people should be using haos; either natively if the box is dedicated; or in a vm
I just bull-in-a-china-shop'd my way into building a server and setting up HA. So the fact that supervised hasn't blown up in my face yet is impressive
The rest of those people should be using container
I wouldn't have picked supervised if I had known that all the add-ons are mostly things that exist standalone anyway. Which is how I ended up running them, yet that also seems to be blasphemous. So I'm just asking for something to go wrong apparently lol.
Should be pretty easy to take a backup and move up HAOS
I'll definitely look into it. I mean, I haven't had any issues yet but everyone seems to think its inevitable.
And by pretty easy, its just take a full backup, spin up HAOS, then restore backup
Job done

Hi all, I have a question regarding hardware. I'm currently running HA on a RPi3 and an SD card.
I'm thinking of upgrading the hardware to something that is somewhat stronger as I'm running more and more integrations, add-ons and so on. This is the result of running it for 2-3+ years..
I'm thinking of the home assistant blue (to build myself) or would you advise me to go for a yellow?
A yellow or a blue isn't that much of an upgrade of what you have (especially a yellow, that's still a Pi). Recommend something proper instead
I've tried to install a HA instance in a proxmox environment (on an i5-3337), but this was extremely unreliable (might be due to my lack of skill) but I reverted back to my pi
That would be a far better choice if you want to upgrade. There are one line scripts that can install HAOS on proxmox for you too
@wooden smelt thanks, I'll give it a other try soon... I allocated 32gb for HA within 2 weeks it had a logfile that completely filled the 32gb .. so that was a bit of a disappointment, also it kept losing the connection with Nabu Casa... Whereas the RPi install (with backup restored of the proxmox install) doesn't seem to have that issue
So are add-ons really just other programs that already exist just ported to HA?
They're just other docker containers
i am currently using HAOS I think I would like it better anyways. It is a pain to do at times with VirtualBox with the USB passthrough
Oh so if I switch from supervised to something else then I definitely risk sacrificing some functionality like the Google Drive Backup add-on?
Sure, because either that's available as a container, or because there's better solutions out there
I'll take rclone over that add-on any day
Let me see if I can figure out how to properly word my question
https://rclone.org/overview/ - take your pick of destinations
I have a Dell R710 that has VirtualBox that is there.
google drive backup is an add on
i have to go add that back in
The add-ons I see in home assistant - Adguard Home, Plex, Grocy, etc - are all things that already exist as normal installs for whatever OS you're running.
So am I correct in saying that you don't NEED add-ons with HA if you can just install other software on the machine?
Supervised and HAOS are essentially the same thing, except Supervised is a ton more work for really no benefit. Your addons from Supervised will work on HAOS
Correct
If you use HA Container, then you have to install containers or other software yourself, there's no add-on menu or anything
That's what the documentation did not do a very good job of explaining. It just goes "add-ons, add-ons, add-ons" without explaining that your average user isn't even going to need add-ons.
Yeah, well, it's aimed at getting people onto the HAOS bandwagon...
Add-ons are the next great thing, the world didn't exist before them, you know π
Because that's the reason I went supervised, I didn't want to sacrifice any features or functionality.
Except, Supervised is just HAOS where you manage the OS. The moment you install other software outside of add-ons you're breaking the requirements and things are likely to fail on you
If you want to install software yourself, Supervised is a great big trap
They keep telling me that, but I haven't had issues yet
Hell, miss a package update and Supervised may fail on you - it's happened
i have a problem with home assistant remot access where do i need to be for that?
#general-archived is probably the best place
Unless you're using the Nabu Casa #cloud-archived service for remote access
i'm running through a container install. is there a precise set of requirements doc anywhere? for example in supervised i needed to set the docker filesystem drivers, just wondering if i need to do the same.
There's no specific requirements for Container
it looks like overlay2 is now the default and is configured in docker by the installer. that's useful.
any recommendations on where to keep the ha files on the host disk? i'm not a daily linux user so the distrinctions between the various top level conventions aren't known to me in detail
i could just go with /ha
I use:
/data/docker - compose file, environment files, etc
/data/homeassistant - HA config folder
/data/whale - whale config folder
etc
Then I just back up /data and get it all
Technically /media is for removable media π
Personally I've seen all sorts, as long as you're not using a folder name that's meaningful to the OS making up your own stuff is fine
how are you setting permissions on these folders?
wondering whether i should make a new group add myself to it and chmod it to full access for that group. i'm assuming docker will run as root (because of the privileged flag) so it won't have a problem
Depends on the image, some run as root, some run by default as UID 1000, others do other things
The HA image runs as root
that's all i'm concerned with at the moment. i'm just trying to make it so i never need to use root so to write to the config i need to add my user permissions.
bizzarely i'd have no problem with permissions like this on windows. weird that i find acl's easier
Once HA is up and running, set the group of all the files and folders to one your user is in, and set group write
sudo chgrp -R petunia /path/to/ha
sudo chmod -R g+w /path/to/ha
weird. no matter what i do i don't have permissions to create a subdirectory. I've even logged out and back in to make sure hte group permission was applied
what are you trying to do ?
Well this has been asked several times already i guess, but i am unable to find my config folder or the yaml file with the settings
I'm running HA in a virtual linux environment on my windows laptop.
Any directions would be appreciated π
Go there, then
and System Info
Thanks
for containerized how long should onboarding take? wondering if i'm stuck. it's been about 15 minutes and i'm stuck with just the ha logo and animated background.
Depends on what you're running it on
Something like a Pi3, or older, on an SD card will take way longer than something running on a vaguely recent i5 and an SSD
Check the container logs to see what's going on
ok, how do i check the container logs?
i also don't understand where it's keeping the config. i stopped the container in docker, deleted containers volumes images etc and deleted the entire contents of the /config folder to start from scratch and yet somehow it's still got my login information.
ah, forced browser refresh got me in.
docker logs is the command you want
ah, thanks.
