#installation-archived
1 messages · Page 6 of 1
@fallen lichen how do you recommend I fix my config?
@fallen lichen I am not sure what you mean by curl command. Can you please explain further?
The wget command
wget returns try wget —help for more options.
wget -O - https://raw.githubusercontent.com/hacs/get/main/get
@fallen lichen wget: invalid option — ‘0’ usage: wget [option]… [url]…
That's a letter O
It's literally the same command on the HACS install page
Just asking for the copy/paste output
@fallen lichen My bad.
— 2022-11-09 22:23:10— https:raw.githubcontent.com/hacs/get/main/get
Resolving raw.githubcontent.com (raw.gitbubcontent.com)… 213.227.149.193
Connecting to raw.githubcontent.com (raw.githubcontent.com) |213.227.149.193|:443… connected.
OpenSSL: error:14094410:SSL routines:ssl3_read_bytes:sslv3 alert handshake failure
Unable to establish SSL connection.
That is not the command I provided
I feel like the root of your problems may be not having an SSH session where you can copy and paste commands to
I have HAOS installed on a ADATA 256GB SU750 attached to RPI4b using a StarTech.com USB 3.0 to 2.5" SATA III Hard Drive Adapter Cable w/UASP via a powered USB 3 hub. RPI4 is powered via a CanaKit 3.5A adapter.
I lose connection to the HAOS, during which times I can ping the rpi4 but can’t access the HAOS homepage. Even add-ons are inaccessible
If I restart just the rpi4 HAOS starts normally and remains accessible for a few hours.
which logs I should look at to find the cause of this?
hi,
I've just upgraded my truenas core server to truenas scale and have to reinstall all my app/plugins including home-assistant
I have a dataset with my previous homeassistant installation (following these instructions: https://github.com/tprelog/iocage-homeassistant/wiki/external_dataset)
How can I mount the dataset in my new homeassistant app (using the truecharts app installation)
EDIT: alternatively, I could settle for overwriting the current default homeassistant configuration and database file from my recent home-assistant installation from my backup in the dataset. Then I would create a cron job to backup the configuration files
You should be able to just add your existing dataset as external storage to the TrueCharts app but last I checked doing so falls outside the support of TrueCharts. They only support using a PVC. Probably best to get help straight from TrueCharts on this one. (I would love to help but I'm not using any of those SCALE or TrueCharts apps)
Currently on docker (core-2022.7.7) and saw that there is a newer release (https://www.home-assistant.io/blog/2022/11/02/release-202211/) how do i update to it?
Going to Settings -> System -> Updates, says that there are no updates available, in the 3 dot menu for the updates page it has on option to leave beta channel, so should i do that?
If you're running it from docker then you'll need to pull the new docker command and then reload the container
Which hopefully you have a docker compose file to make it easy
this right? https://imgur.com/a/DuD28mk
That's a docker pull command yes
I can add the dataset as external storage (a simple mount point).
The question is given the base directory tree of home assistant, what should be overwritten from the dataset.
I currently have two PVCs:
- db: with a
datadirectory containing a postgresql database (a bunch of files with apg_prefix) - config: with the rest of the installation, namely the
.storagedirectory, a bunch of yaml files andinit,tts,deps,custom_componentsandblueprintsdirectories
I don't know how to override the base config directory such that it is replaced with the dataset (anyone know how to do that?), but I guess I could replace the files with symbolic links to the dataset files in a mounted directory.
which of the files must I replace such that my previous installation is restored?
Ultimately all of them, everything in your config folder
You can skip deps/ and tts/
thanks. will try that.
what about the postgresql db?
No idea, I don't use PG, but I'd assume that you have to point your new PG server at that folder
This is a good opportunity to think about how you backed all this up before, and how you were planning on restoring it
on 2nd thought I don't think my plan is a good idea, since truecharts manages the configuration directory by overwriting the files.
so the next time the app is updated or some system configuration is changed, the symbolic links would be overwritten by a new file
the way I see it, I have two options:
- copy the files from my backup to the new app, and make a periodic job that copies the files to the bkp dataset
- find a way to point the PVC to the dataset
the 2nd option is more robust (and cleaner), but requires PVC/K8S knowledge that I don't have. Any advice on how to do that?
PVC is persistent volume claim... True charts app runs in kubernetes
apparently it is "Persistent Volume Claims", some sort of kubernetes terminology for storage
but for all I know, it might as well be plastic...
Yeah, I don't know anything about TrueCharts/TrueNAS at that level - and I doubt there's a lot of folks here that really do (other than Troy)
And Im still using a hack to run just native docker without all the kubernetes complication.
Do you have instructions for native docker installation on true nas scale?
BTW, I copied the configuration files, and home assistant is up with almost everything. But I am missing automations that I wrote.
In what file are the automations?
automations.yaml
My docker hack for TrueNAS Scale: https://gist.github.com/tprelog/7988dc6b196775f33929beb19f0090d7
it (automations.yaml) has been copied, and has the sctipts.
The script in question triggers an action on a Tuya device when a Shelly relay is pushed. both tuya and shelly devices are recognized by homeassistant and can be operated manually.
Any clue why the automation script does not work?
How would you debug it?
I'd start by checking to see if the automations are found and loaded by HA
I see that your script states: "You CANNOT use SCALE Apps while using this script!"
Does that mean I have to use the docker hack for all apps (I have a few media management truecharts apps)
Yes sorry. It's intended as all or nothing.
automation is found in ha. but not working
are relay clicks (push up/down on a two state button connected to a shelly i3) supposed to show up in ha logbook? I am not seeing the clicks, although the relay is found by ha
@queen dagger 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).
Now you can follow the test advice in the #automations-archived channel's pinned messages
I'll have a look
and sorry about the pings (I don't see the @keen rose/@off option)
It's right there when you hit Reply - on the right side
Exactly where the image shows
Please don't tag me 🙂
You have the most unfortunate Discord handle
That's my name 🤷♂️
(except for the other Tinkerer, who gets random pings...)
Ah, the joys of a global community
Guessing by the name I would say Hanan and I are from the same country 🙂
@keen roseYesThatsMyName
Huh?
That was it btw 😄 Though I had planned on opening up the machine and cleaning it out prior to putting it in its permanent home and configuring HA, and in the process of disassembly completely biffed it. C'est la guerre.
the corollary to "don't release the magic smoke" is "don't remove the magic dust"
Anyone know where to get more info on installing onto VMware? It seems like there should be a VMX file and I cannot get the VMDK file to boot
VMWare Workstation on Windows? https://www.home-assistant.io/installation/windows should have everything there
I didn't even get to that point, I couldn't get it open to clean it, there was a magic screw that I just could not find.
I am on core, is it possible to edit the configuration file from inside the homeassistant browser page?
without addons? you need file editor/browser addon
or just another web-based editor with an iframe button in the sidebar
I think he's running core with no addons?
yes, that doesn't change what i said though 😄
google for your case; there's probably instructions. what kind of pc/case?
I had to google how to open my lenovo tiny pc thing yesterday 🙂
What's that got to do with it
I thought you said you had problems opening the case? Couldn't find the magic screw?
or rather, theoreticalsandwiches
Was replyiing to
[8:25 AM]TheoreticalSandwiches: I didn't even get to that point, I couldn't get it open to clean it, there was a magic screw that I just could not find.
Yeah you didn't reply to anyone so it was very confusing order
Since you can do reply but not ping if that's your concern
Sorry 🙂
@still hedge I converted your message into a file since it's above 15 lines :+1:
Surely something simple, any help would be appreciated. Thanks 🙏
Hi there,
After Upgrading my new Yellow to 2022.11.2 the following 3 repair warnings showed up and I wanted to know, what I can do about those. Any help would be greatly appreciated.
Unsupported system - Systemd issues
Unsupported system - CGroup version
Unsupported system - Operating System
If your Yellow is running Home Assistant OS then try a reboot
Surely something simple any help would
Hey guys does anyone maybe know how I screwed up duckdns for my server? It doesn't send information to smart things or aws
DuckDNS doesn't send any information to anybody
DuckDNS resolves your hostname to your public IP address - the one from your ISP - and nothing more
I mean it keeps timing out
But I can fully access with my phone or pc
Can you access with your phone when you're not connected to your WiFi
Yes
And https://www.yougetsignal.com/tools/open-ports/ shows it as open, and https://www.ssllabs.com/ssltest/ returns a valid certificate?
It does say the port is open going to check ssl labs quickly
Ssl labs doesn't support port usage like :8123
Then Google around and find one that does 
Then you know the problem isn't at your end
But it isn't working for smart things or aws
To set up alexa
But when I used nginx' http port with aws it worked kind of but it kept saying unauthorized in logs
Maybe those both require you to be using port 443?
And unauthorized could mean many things, without seeing the actual logs we'd be guessing
I used the correct tokens for alexa it was supposed to be working and ai did exactly the same before my hassio died
it worked before my hassio died
And when I used nginx I couldn't access my server on my wifi but could access using mobile data
That shouldn't have caused more problems, if anything it should have simplified things since you can go direct (no SSL) locally if you want
Exactly what I'm breaking my mind about
And from the logs this is what smart things is telling basically the same as aws
[{'code': 'TargetTimeoutError', 'target': 'https://hassio.duckdns.org:8123/api/webhook/377da71965dc1f5e01a329482e5562ab8f5608c6489e9bbdd87f1e0c18b5f958', 'message': 'Upstream target timed out', 'details': []}
Or I think I found a problem I 'll put this in pastebin this time
That's coming from something trying to set up the same webhook ID as already exists
Typically that's from an automation in my experience
looks like smartthings
I think I had exactly that error with exactly that integration
So do you know how I can fix?
iirc I uninstalled the untegration and restarted HA before reinstalling it
I'll check if can find something of iirc
I couldn't find something with iirc but I'm rebooting to see if the addons and integrations I uninstalled fixed it
Could it be the fact I had nginx stopped but not running?
It still doesn't seem like its working let me try updating hassio
Still not fixed
And I unlinked the cloud because it said something of webhooks amd I don't want to pay every month
Why aren't you using https over 443?
Is it even possible with duckdns?
Probably with port forwarding I can try
Still the same using 443
8123 is also enabled
When I go to the webhooks page smart things is trying to access it says method not allowed so it is trying to access it but something is breaking my head
Hmm unsure then
Yeah and I also have set the internal and external urls in the config file
DuckDNS doesn't care what ports you use... it doesn't even know. All it does is turn a hostname into an IP, it has zero involvement in the actual communication
But it must be something with lets encrypt
On duckdns
Is there something I should add to the aliases?
LetsEncrypt just gives you a cert, it isn't involved int he connection either
Then I really don't know what to do I'm so damn confused
I actually want to let someone on to my server and let them see for themselves because I'm clueless
Do you think you can maybe help me out?
You already appear to have working remote access, since you say you can access HA remotely with your phone
If you have valid SSL certificates and can access HA remotely then the bits I can help with are working
I'm not even sure what you're doing any more, whether you're using a proxy or going direct to HA, or ...
I'm going direct no proxy
I think something is up with my servers api because alexa wont work either but google assistant does work
hi to all, i have an installation running on a PI4, i'ts not an HAOS, how can i check if it's a doker installation or a core installation? i don't remember it
Not sure which install you're using? Check the Home Assistant info page
So [external public IP] is being used?
To confirm it is public, can you confirm it does not begin with 192, 10, or 172?
ok, I'm running a supervised installation on raspbian, and from the last 2 updates I've already noticed pending repairs (journal, OS and AppArmor). I only need to run a few things on the same machine alongside HASS, so I can't use a dedicated PI for HAOS, and I can't use a second board just to have an OpenVpn client and some iptables routes. Do you think a new installation with the Containered version can solve these problems?
Translated with www.DeepL.com/Translator (free version)
Okay so that's a super unsupported system
those are private classes, so not publics
Correct. That's why I said confirm it does not begin with those. I have seen someone try to connect remotely to 192.x.x.x lol
you talk about mine?
Yes
why Super Unsupported?
Supervised means you have to dedicate your machine to HA
Raspbian is not debian 11 either
Switching to Container from Supervised is the only sensible solution
^^
ok, perfect. as i read, only issue i may have will be to expose manually some devices, like zigbee dongle?
Yes
no no, not issue, just it's how contaniners works
a full backup will work also restoring to different install method?
You'll have to manually restore it
of course, after fresh install i'll upload backup and start a restore
You'll have to restore it manually
question is if after upload and restore, all works again or if due to different intall method, somethink must be manually restored
There's no UI support for restoring backups in Container
None of your add-ons will come across
oh.... nice..... seems will be some differences in this kind of install. i need to find something more to read about it
so, supported os to run supervised, actually are only HAOS and debian 11. Raspbian, is not debian, but i'm reading i can install debian 11 on raspberry pi 4, so i can next rebuild my actual environment: hass supervised and my other stuff on same machine, with a fully supported os
is it right?
Yes, but did you read the rest of the ADR?
You can't install any other software, none, zero, zilch
sorry, probably i'm missing something: supervised means i install hass in my OS, not in HAOS, right?=
so i have full access to system, why i can't install something?
and as i sayed, something stand for vpn client a some iptables stuff
You can't, because that's what the requirements say
If you install other software then at some point it's going to flag as Unsupported and/or Unhealthy and either it'll stop allowing updates, or worst case it'll simply stop working
ok, i understanded it now
so, at this point, what is the advantage to use my own os if i can't use it for other purposes but i still have to manage it?
None
as i thinked!
Supervised only makes sense when you feel that you really need add-ons, have a computer that HAOS won't run on, and have strong Linux and Docker skills
ok, it's clear now
my only chanche, before to dedicate the bord to hass, will be to try to install in supervised and assure openvpn will be recognized as an unwanted software from hass. if recognized..... i have to find where configure my vpn
Just don't. Dedicate the pi and use wireguard for vpn (there's an add on for that) or you run container
wireguard will be a good alternative, but iptables rules can't be solved
What do you need iptables rules for
If wireguard can't do what you need then you'll want another machine or deal with container
Because you haven't read the requirements and things will break
tanks to all for explains mates
just last thing: actual install on raspbian as supervised, will have issue on short times?
or i'll just have alerts?
You'll likely have issues
ok, thanks
By all means, ignore us and the ADR and do what you want, but don't expect much help if you do
Either do HAOS or Container. Avoid supervised
haos will be my choice and i'll use an old PI3 just for vpn
thanks to all
need to find how to boot from SD and move os to ssd
It is public only
is everyone runnin OS 9.3 already, just noticed its a release, and no longer in beta?
asking because in the community, people seem to have issues, https://community.home-assistant.io/t/ha-dead-after-upgrade-to-9-3/480525 and the repo only holds 1 closed topic on that matter https://github.com/home-assistant/operating-system/issues/2219
Hi all, completely new to HA and after 4 days of struggling (tried to instal a VM on nuc without succes) managed to instal HAOS ! How long does the prep in browser usually take? (Max 20 mins but its been 25+ already 😅)
Depends on the speed of the computer, the storage device, and your Internet connection
Click on the blue circle to find out what's going on
Thanks! I can see it’s been stuck on something for the past 10 mins but I’ll give it some more time. Already happy I can see more happening than before!
I'm not sure what I did wrong but I clearly did something wrong.
I neglected my HA for a while and tonight I decided to get back to it. I did all the steps for remote access, DuckDNS, Lets Encrypt, DNSmasq (my router doesn't support NAT loopback) and I had everything working fine, except I couldn't get access on the app on my iPhone when connected to my home network, only when using cellular data, though my laptop worked on my home network.
But now I have an even worse problem. My HA IP was 192.168.1.69 but now it seems to be 192.168.1.73 according to my router and when I go to https://192.168.1.73 it wants to me setup HA as new. I know very very little about networking but I assume this may have something to do with a DHCP lease. How do I force my HA IP back to .69 so I can access it again because right now I have no access to it at all it seems. And once I do get it back, how do I make sure that HA stays with the same IP without changing again?
Thanks!
Are you sure you don't have a second HA install?
IP allocation is handled by your router though, make sure it's allocating a static lease
I only have 1. I was working on it as .69 then all of a sudden I tried to login and couldn’t. Logged into my router and it was now .73. How do I ensure static lease is allocated?
That's handled on the router, whatever it is
Click around the DHCP settings and see if it offers static leases
There was something about assigning an IP to a MAC. I tried that and assigning .69 to the MAC for HA but HA still shows as .73
It won't take effect until after the DHCP lease is renewed. You could potentially force that to happen on the host depending on the host OS
If it's macOS: System Settings > Network > Click on your connection > Details > TCP/IP > Renew DHCP lease.
Substitute "System Settings" for "System Preferences" on pre-Ventura releases
Alright I’ll give that a shot, thanks all!
Hey Guys, I am trying to update a 2021.2.2 build, but I am having problems accessing the supervisor dashboard... I also don't have the ssh addon installed, though I cannot update or maintain it through the cli. Is there anybody that can help?
Hey all. I'm trying to update home assistant core on a Raspberry Pi 3 from 2022.10.5 to 2022.11.2. The update fails with the error "image doesn't exist" when trying to pull the docker image for rpi3 2022.11.2. I went to the github page where the docker image is hosted and it does indeed exist. I'm stuck and unsure how to proceed. 😦
Can someone help me by answering a few novice questions? I bought a nuc with with windows pre installed for HA (new to this, i thought it would be easy) tried installing vmware (bootdisk) but didn’t recognize my network drivers. People on reddit recommended HAOS given that I’m new to this “would be easy” they said. Should I uninstall windows from my primary drive as it starts up windows by default or should I always reuse the bootdisk? (What do you have?) next issue, earlier i thought I had managed, nuc actually gave me the URL however… on my phone it took more than an hour to set-up so I cancelled wanting to try this again. Is this common or am I really the black sheep (self pitty)?
Hey everyone, I installed a new router today and got assigned a new external IP address. The internal IP of Home Assistant has been set to a static IP. I am using the Nginx Reverse Proxy addon to access my DuckDNS configured Home Assistant instance. The weird thing is, is that I only can access my instance externally via the DuckDNS domain. Trying to access my instance via DuckDNS internally doesn't seem to work. My http config is listed below:
http: server_port: 8123 use_x_forwarded_for: true trusted_proxies: - 127.0.0.1 - 172.16.0.0/12 - ::1 - 172.30.33.0/24 ip_ban_enabled: true login_attempts_threshold: 3
Trying to access my instance via DuckDNS internally doesn't seem to work
That requires your router to support NAT Reflection or Loopback NAT
Do you only want to use the NUC for HA? If so, just installing HA OS on the bare device would be much better. Alternatively (and what I do), is run Proxmox (hypervisor) on the NUC, and still run HA OS in a VM, then you're free to make other VMs as well. I'm not entirely sure what you're asking in your second question and not sure what you were setting up for over an hour.
Right now I only want to use HA, get familiar and comftable with it etc. in 3 to 4 months we're moving and I can consider doing other apps/things alongside it. Initially I tried installing a VM (VMWare) but I got a message my networkcards were not found. Making a custom bootdisk for this was impossible both from a lack of experience (never used powershell) and I tried 2 videos, kept running into problems.
From what I read it's a very common problem with NUCS and there's community drivers etc but the powershell part is what I got stuck on and why I decided to try HAOS. - edit- my second question is, after the initial set-up on my NUC I get a link that I can visit (i did it through the app on my phone)in which it's setting up what I assume to be the interface. there's a lot of lines processing the set-up and a message saying it can take up to 20 mins. but for me it got stuck.
I nuked my sff pc windows 11 and put proxmox (a hypervisor) on it, and installed a few VMs/containers, including HAOS. Love it!
You can run HAOS in virtualbox if you want, then it's completely isolated from the host OS. That works pretty well too.
Did you click the little circle thing to see the logs, see what it was stuck on?
If you got HA OS set up, you can click on the blue circle on that screen and it'll show you the log.... sometimes it can take longer than 20m
But I'm also not clear.... you're not booting HA OS from a flash drive, right?
(cause you shouldn't)
HAOS on vmware should be pretty straightforward, and not depend on your physical hardware at all
Guess it depends if it was Workstation or ESXi, kinda assuming it might've been ESXi if there were driver issues and a bootdisk was mentioned
ESXi is it's own OS, the OP mentioned he was running on windows?
So I'm assuming workstation
Lots of questions 🙂
He’s an obvious candidate for haos no need to complicate it
Yeah, if he doesn't mind nuking windows, and doesn't need anything else on the box, go HAOS
If you installed container then you don't have the supervisor so no add ons
and that, too
mistakes have been made
That could be clearer
You still won't get addons, though
You can use zigbee with ZHA without needing add ons
You can install whatever you like alongside your HA container
So I renewed the lease and the IP is still showing as .73. In my router I see: homeassistant MAC ADDRESS
Ethernet Port 3 – 1000 Mbps Full
192.168.1.73 DHCP
In DHCP reservation I put: homeassistant MAC ADDRESS 192.168.1.69
Any ideas anyone?
Did something else grab it?
Did you try just rebooting the box
The router?
No the box w/ HA
Oh, no. Should I?
Wouldn't hurt just to rule that out.
Okay doing it now
Okay so I did that and it came back online now as .69. But I still can't login using duckdns link or IP...
correct
https://192.168.1.69:8123/ gives This site can’t provide a secure connection192.168.1.69 sent an invalid response.
ERR_SSL_PROTOCOL_ERROR
use http
That brings me to the HA onboarding page, not my HA setup that I had
What is it installed on
Proxmox
DuckDNS link brings me to the HA loading page but just spins forever then errors out.
What happened between it working and it restarting onboarding
So yesterday it was working on .69 then all of a sudden it logged me out and I saw in my router is was .73
You don't happen to have >1 HA VM?
The only way I know of to restart onboarding is to nuke a bunch of stuff in .storage
No I don't have more than 1
Not sure then if you're saying you did nothing at all between the 2 events, seems like something may have gotten corrupt.
Yeah I didn't do anything abnormal I was just logging in, redid the steps for making duckdns work, I ran some updates, and tested a few devices. All was good
I did have a problem in the past where I would get the duckdns link to work and then a few minutes later it would stop working and I'd have to access it via the https and IP instead of duckdns URL
And that same problem happened again yesterday too
Is my old HA gone now? Do i have to upload a backup and start that way?
Possibly? I mean in Proxmox you can try typing
login
ha core logs | more
and see if there's anything exciting there
Morning all. Just starting out got a Raspberry Pi 4, 8GB SD card & fresh install of Home Assistant thanks to the Raspberry Pi imager. At the screen telling me 'Preparing Home Assistant'. So while I wait is there anything a noob like me needs to know?
SD cards aren't recommended 🙂 But if you must, update the recorder config and severely limit what you log and the commit interval
64GB USB3 stick any good & how would I move over to it?
That's basically going to be the same as your SD card
I feel like a real noob right now, how do i login to proxmox again?
Just to check. What install method did you use?
However your proxmox is configured? https://[IP]:8006
Not sure what you mean exactly, sorry
Yeah I have no idea, did it a long time ago
Did you install HAOS, container, etc?
Surprised you don't remember. Should know how to update proxmox lol
I followed this video: https://www.youtube.com/watch?v=42gopfIrPBY
I have backups set to my google drive

So I would say best thing is restore your backup then
Since something in your storage has gone wrong
Okay thanks, i'll do that. Is there any way to ensure this doesn't happen again besides assigning that IP to HA?
Just got a new backup in my Drive from my old HA so it's still running...
Trying to restore from backup, I upload the backup, click restore, it asks if I'm sure, I click restore, it just takes me back to the onboarding page asking me to make a new account or select from backup...
Try restarting your proxmox host entirely then. Just see what on earth is happening
If it's doing backups then you're on a secondary instance
Also wait, you're accessing your instance at [ip]:8123 right?
How? and yes I'm on :8123
Question for you. Why did you install it through proxmox
By restart host do you mean just restart the computer it's installed on?
If you don't know how to use proxmox
Because that seemed to be the most recommended option and after following a few tutorials I felt like I had the hang of proxmox well enough that it seemed like a good option
You don't have the hang of proxmox if you didnt even know how to access it.
Are you running any other VMs on this computer
I did know how to access it, like I said I haven't had to deal with proxmox since I initally set it up. At the time, I could work my way around it pretty easily. But ti's been a while.
No
@fallen coral when you do a restore, it looks like it's not working, with no progress bar, for like 20 mins. Then you might even need to reboot.
If you have question about proxmox + haos, then I can probably help (that's what I have).
Yeah it would just reload the page and bring me back to onboarding
Can I just create a new account and upload my backup in there somehow?
Honestly if you don't have anything else running and don't have a plan to, then just install HAOS directly on your hardware
it takes long time. looks like it did nothing. but it's working in the background
Yes
Where within HA do I go or how do I do this? I did google it but didn't see a conrete answer
Under backups
But seriously would recommend nuking your install and just running HAOS directly on the hardware instead of proxmox if you don't plan to run other things on the hardware
If your proxmox is working fine, then it's really no hassle.
Except for unnecessary overhead and complication
shrug
overhead is small, and it's flexible. plus you can do a one-line haos install
bash -c "$(wget -qLO - https://github.com/tteck/Proxmox/raw/main/vm/haos-vm-v4.sh)"
Which they didn't do anyway
nah they didn't 🙂
And I'm just not a fan of saying virtualize when a) we don't know the hardware being used, and b) when it's only HAOS there's no advantage to proxmox
but if he's asking how to create an account, he doesn't know HA either 🙂
No, that's not the issue. His issue is he has a working install somewhere except now it's asking for onboarding steps
Whilst also working
@fallen coral just to sanity check, are you using the latest backup or a backup from before everything started to go wrong
hasn't he clicked the "restore" and uploaded a backup? Just need to wait a while. It's sucky UI for restore.
That's also what he's tried
I tried both.
I would say a reinstall of HAOS at this point
And basically start everything over from scratch?
No? You still restore from backup
Just seems like something is heavily broken with your install
Except trying to restore from backup did nothing. There are only 2 backups in there and I tried both
And use the one line install mentioned here if you do reinstall
Are you sure you gave it long enough to do the restore? It's pretty slow, and has no feedback
And you may need to browser-clear cache etc
I find creating an account, then restoring from the backup to work better
Is there anywhere I'd be able to see if the backup is uploading? Anything in Supervisor?
Nope
It's crappy UI
It asks for a backup password, is there any way to know if I'm inputting the correct password? Or do I even have to put one in?
I also get this in Supervisor:
22-11-11 20:50:24 ERROR (MainThread) [supervisor.homeassistant.module] Can't read tarfile <securetar.SecureTarFile object at 0x7fc526d8e8f0>: not a gzip file
22-11-11 20:50:24 ERROR (MainThread) [supervisor.backups.manager] Restore 1dc66eca error
Sounds like a problem 🙂
Can you send a screenshot of what your backup files look like in Google drive? Just the file name and extension is what we need
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.
In my google drive they're just .tar files
❯ ls -l ~/Google\ Drive/Home\ Assistant\ Backups
total 2619104
-rw------- 1 simonroberts staff 312954880 4 Nov 08:06 4 November 2022.tar
-rw------- 1 simonroberts staff 261068800 2 Nov 14:25 Full Backup 2022-11-02 14:25:02.tar
-rw------- 1 simonroberts staff 373534720 7 Nov 08:06 Full Backup 2022-11-07 08:06:39.tar
-rw------- 1 simonroberts staff 393420800 11 Nov 08:35 Full Backup 2022-11-11 08:35:21.tar
Yeah and I'm guessing these aren't otherwise that error wouldn't be there
It might be talking about one of the gz that are inside the tar...
❯ tar tvfz ~/Google\ Drive/Home\ Assistant\ Backups/Full\ Backup\ 2022-11-02\ 14:25:02.tar
drwx------ 0 root root 0 2 Nov 14:26 ./
-rw-r--r-- 0 root root 1276 2 Nov 14:25 ./a0d7b954_sqlite-web.tar.gz
-rw-r--r-- 0 root root 5026 2 Nov 14:25 ./a0d7b954_ssh.tar.gz
-rw-r--r-- 0 root root 189 2 Nov 14:26 ./addons_local.tar.gz
-rw------- 0 root root 1251 2 Nov 14:26 ./backup.json
-rw-r--r-- 0 root root 3744 2 Nov 14:25 ./cebe7a76_hassio_google_drive_backup.tar.gz
-rw-r--r-- 0 root root 1667 2 Nov 14:25 ./core_configurator.tar.gz
-rw-r--r-- 0 root root 261027251 2 Nov 14:26 ./homeassistant.tar.gz
-rw-r--r-- 0 root root 182 2 Nov 14:26 ./media.tar.gz
-rw-r--r-- 0 root root 182 2 Nov 14:26 ./share.tar.gz
-rw-r--r-- 0 root root 180 2 Nov 14:26 ./ssl.tar.gz
And you only have 2 backups in g drive?
And when i unzip it, I can see all my add ons that I had in there like duckdns, etc
Yes
Don't unzip it to install
I'm not unzipping to install. I unzipped to see what was in it if it was from before the issue or from after
I've successfully restored from my google-backup tars a week or two ago. Worked OK.
And you're restoring the backup by creating a user account -> then going to backups and clicking upload/ restore/ whatever it's called?
Yes I am in the new account, went to backups, it shows both my backups there, I select the one from 2 days ago, it asks for password and restore then I click restore then it says are yous ure and i hit restore then that window goes away, HA loses conneciton for a bit then says HA is back online after like 30s
Sounds like restore worked. Did you do a hard-reload at that point?
Maybe even restart HA too
I would reboot the host after a full restore
I rebooted everything, no dice
still asking you to enroll? sorry I have no idea man
Sounds like try a reinstall as your install is broken somehow
Then restore from backup
Isn't that what he's doing? Except the restore is failing?
No? He hasn't reinstalled anything
OK I didn't read the start of the thread
If he's in proxmox, run the script, get a new install 🙂
Okay i might have found the solution. I found a note in my phone with all my HA passwords and found the proper password for my backups. This one seems to be taking longer than the rest so fingers crossed.
Nope. Can't even get the page to load now. And can't even see it connected to my router. It's showing homeassistant with 2 different IPs but neither of them work
To do a clean install I need a keyboard and monitor for my pc right?
Did you restore a duckdns SSL config? You're using https:// if so?
Of course I was... I wasn't, but I am now and it worked. Thank you!
And thank you to @copper steppe and @manic bane for your patience with me. Very much appreciated!!
Wahoo.
Is proxmox difficult to install? I tried esxi but had driver issues, tried fixing but this was way beyong my skill.
Currently I am cause on restart it automatically starts. I followed the tutorial on haos website and made a bootdisk on usb to install. Windows however is a nightmare to remove ( or prevent from starting)
Its a nuc indeed, esxi gave network driver issues. As the nuc came with windows pre installed I’m not sure how to change this.
I don’t mind but sadly I don’t know how. Esxi failed due to drivers but when trying to install haos it starts windows regardless.
You probably need to change your bios so it boots from usb, rather than disk first.
The write proxmox to a usb, and boot/install it
Nuking the entire disk (including windows) is pretty much the default install. Get it booting, we'll walk you through it
TLDR download the iso, write to a usb, do basic config (network in particular). After that, I tend to use the proxmox helper scripts to do haos install - https://tteck.github.io/Proxmox/
What do you mean? After installing HAOS? Or when you're trying to boot the install-disk/usb/cdrom?
Install is super easy. Takes 5 mins once you have usb written
I did it last week
It does boot from usb but on restart if it boots from c its windows again. Should i keep booting from usb? ( genuine question)
No, you didn't install it correctly I guess. It should have nuked the C: completely, should be no windows left.
I'm currently running HA, pfsense (firewall), heimdall (dashboard), kuma uptime (uptime monitor), node-red (uh, decision engine), and an ubuntu container. All works nicely.
This was before i tried HAOS i wanted esxi as VM but this gave big trouble with network drivers. After s while i gave up and wanted HAOS without vm
"esxi as vm"?
You have no idea how happy I am to hear, I’ve grown to hate windows with this whole drama
Esxi is a hypervisor itself, you don't want to try and run it inside something else
proxmox is a fairly-lightweight virtualisation manager, broadly the same as ESXi. Install both on bare metal.
I have to go in a few minutes but will you be around later today?
Maybe for the next hour or so. It' s 6:20pm in my timezone
I can tell you in s chronological and better way what i tried and how it failed
Ahh
Follow https://www.proxmox.com/en/proxmox-ve/get-started - it's fairly straightforward.
I’ll give that a try.
This is not windows fault lol. Your device is still set to boot from the main drive first (which given nothing else is installed on it makes sense) and esxi is meant to only be used on business/ enterprise hardware so driver issues aren't surprising (had them myself too)
Yeah, proxmox is way more easy-going than esxi
That helped, seems somehow the previous reboot left the device in a limbo state somehow... Thanks
I was not aware 😅 - edit- does explain why their so eager to advertise/contact.
Hi all,
I posted last night but was in a bit of a rush and seeing how there were some questions I figured I’d write out what my struggles are and hopefully someone with more knowledge/expertise can help me fix this. To give a good context I have to start at the start. I know a bit about IT but I’m no expert, never used a VM but have always been interested in tech and eager to learn.
- Few days ago (Last Tuesday) My nuc arrived. I ordered it from amazon.de (despite not being german) - Mini PC Intel Celeron J4005 2x2,0GHz 16GB RAM 1000GB HDD Bluetooth 5.0 HDMI RJ-45 USB 3.0 Windows 10 Professional.
- It came with Windows (in german) pre installed, as I do not speak the language I struggled for a day trying to change language settings but eventually downloaded the iso and re-installed it in English expecting HA to simply be a program I could run in windows (What do I know?)
- Wasn’t sure what version of HA to use and why I would need a VM but eventually was told by a friend to download ESXI (8.0) and make a bootdisk. The ISO was giving me driver issues (network) and so a dead end.
- “Download the ESXI ZIP and make a default ISO” was the next step, I followed 2 YT Videos both using powershell to integrate the community drivers but in the first videos the commands weren’t being recognized and the second wouldn’t find the VM (according to the video I should have been able to see a list). Again a dead end.
- Posted on reddit and was told I shouldn’t over complicate and install HAOS directly. Made a bootdisk for HA. Installation on the server was successful I could see the URL. Used my phone to download the app and do the set-up on “that side” however after an hour nothing happened. (tried 3 times, sorry I didn’t note down the error I eventually found)
- Posted here, download proxmox. seemed a promising solution. Downloaded, placed on USB and made bootdevice. Installation ran for 3 second then black screen.
With all the above what has been consistent is windows. Each of these methods should have cleared windows (right?) and replaced it with either the VM or HAOS but upon restart there it is again…
Does anyone know what I can/should do? Would anyone have some time to guide me through this cause this is going further and further above my knowledge and it’s come to a point that I feel rather shit for spending 250+ on the nuc thinking the install would be easy.
No if you're not installing it over the internal drive you're not going to wipe windows. If you've installed and told esxi or proxmox to boot off the USB (and HAOS has to be flashed to the drive directly as per the docs) then you're not wiping the main drive
Windows won't just reinstall itself, you're just not wiping/ installing on the main drive of the nuc
Also, that seems wild it comes w/ a 5400RPM drive, maybe a good first upgrade 🙂
Pretty sure I followed this video where he shows just installing Proxmox from a bootable drive and then how you get the image to Proxmox to install an HA VM, https://www.youtube.com/watch?v=42gopfIrPBY
I have installed HA about 4 times now and some of the app just doesnt work. the Addons page stays blank and then finally errors out AND the zwave plugin will not start, just error after error. i have tried the OS version and the debian docker installs.. both do the exact same thing.
Docker install does not have an add ons page though...
Unless you mean you're doing supervised in which case stay away
i did do the supervised
Yeah don't do supervised install. Just try HAOS
And I imagine something is wrong with your install method then
i did the vmware version of the HAOS first and it does the same thing..
it has supervised too... right?
Yes HAOS is a better option than supervised
addons page does the same thing.. errors out.. and zwave never installs or works
What's the error?
Logs would be helpful
[supervisor.docker.manager] Image homeassistant/amd64-addon-zwave_js:0.1.74 does not exist for addon_core_zwave_js
Are you restoring from a backup or doing fresh install?
fresh
Then provide more logs (through a text sharing site) and someone should hopefully be able to help
Please use a code share site to share code or logs, for example:
- http://pastie.org/ (select YAML for the language)
- https://dpaste.org/ (select YAML for the language)
- https://paste.debian.net/ (you guessed it, select YAML as the language)
Please don't use Pastebin, since it can randomly add spaces to the main view. Please also don't share text as images since it makes it harder for people to help you. Remember that others may have colour blindness, impaired vision, etc.
Super quick yes/no question: when I run HAOS in RPI4 for first time, it shows some checks and green "ok" next to tests extremely fast(cant read but maybe 1 or two fast) then goes to a black screen. I didn't see any NOKS. Is the screen supposed to be black while it does the "up to 20 minutes" first time setup?
might have found the issue.. when its downloading amd64-addon-zwave_js from docker.. its going extremely slow.. in the last 20 minutes its downloaded 1.638MB of 39.94
is it possible to set the backup directory (https://www.home-assistant.io/integrations/backup/) ?
I would like to periodically backup to a mounter directory on my file server
the pain...
Not with that integration
- would it be possible to overwrite the
backupdirectory as a symbolic link to the desired directory? - you said: "not with that integration". Is there another integration you recommend that provides the additional flexibility of setting the directory?
How do you run Home Assistant? HAOS, Container, Core, Supervised?
a container (a TrueCharts app in TrueNAS scale)
Then you can use any external backup solution you want
can you give an example?
rsync, rsnapshot, urbackup, duplicati, kopia, duplicity, ...
almost an hour of docker trying to download an image... https://imgur.com/a/qYZhN6I
everything else but this works fine
Oh, and borg, or restic, or ...
thanks. If I understand correctly, all these backup options are not integrated to home assistant.
Which files/folders do I need to include in the backup?
Right, so you backup whatever you mapped to /config
thanks.
?ban @shell burrow
:dynoSuccess: Ferosin#1906 was banned
thanks
does it make a difference really if i install cloudflared and adguard home inside my home assistant VM vs separate docker containers?
cloudflared is supposed to be some % faster than regular proxying but I have no experience with that
doesn't really make a difference if you install adguard in haos or any other machine
@zenith lodge did you get the install sorted?
thank you @low notch - ive been playing around with my "new" nas, HA, nginx, cloudflared, etc., trying to figure out what works and doesn't so I just factory reset my router and NAS and am going to start fresh
apparently it didn't like my * in my domain - still learning yaml I guess :d
Hi there, I've had home assistant up and running for a few months now, it's worked perfect. but the last couple of days my https url isn't working, I can access via homeassistant.local, but accessing vis https isn't working.
it's a managed via self hosted nginx proxy manager, I can access my other sites fine via the SSL certs made, but home assistant isn't working, I've deleted and recreated the cert and still nothing, checked site url settings in home assistant, all checks out
it's on a vmware vm
hmm, restarted vm, got an error, but it works.
I've been running HA Core. I'm trying to migrate to container. I've stopped my core installation, then i copied all my files to the docker file directory, and i've started a new docker image (via portainer) that is the same version as my old install. But its giving me the onboarding page. Am I missing something obvious here?
Is the docker directory mapping correct?
I mean thats the most obvious answer. Its weird. I first went through the onboarding and made a temp installation. Then i copied over my old files and it didn't work.. only the new temp installation login was accepted
but i cleared out that directory and then was presented with the onboarding page again
could it be permission related?
Who knows with portainer
Now its weird i see logs that are clearly from my custom install in the portainer thing. But i'm still getting the onboarding page. Should I not be using portainer? Is it recommended to just use docker without any other layer on top of it?
It is a lot easier. But this may also be browser cache if the logs are saying otherwise. Try opening your HA in private browser window
No dice. I'm going to try to set it up with the standard installation method on the documentation
wtf i'm getting the same thing that way. i'm upgrading from 2022.7.6 by the way. this time i tried to skip to the latest stable version of HA figuring the DB would update too
actually ive got a directory issue let me fix that and keep debugging soryr
anyone know where let's encrypt addon in HA puts the generated certificates? thought it would be somewhere like /data/letsencrypt/live/ like in the log but i can't seem to find where that's at
/ssl?
alright this isn't working.... docker, portainer both do the same thing. if i look in the docker log it shows messages that are clearly from my installation. but when i go to the website it looks like a clean install and I can't get past the onboarding page. Note when i turn my core installation back on it starts up right away. if anyone has any ideas what i might be missing let me know.
For now i'm just sticking with running my core installation
Post your docker config here
Back. Ready to help:)
Ever time I try to install home assistant from an image it does not boot. Tested the SD card with R PI OS it works just fine anyone have any ideas?
The installer does not boot, or haos post-install?
I get nothing when I boot I can see files but nothing appears no installer no text nothing
What Pi do you have and what image are you using?
I've tried on a pi3 (HA OS 9.3 Rpi 3)and Banana pi BPI-M5 4GB HA 9.3 RPI 4/400) through R PI Imager
Weird
Nop, now trying through ubuntu but having some issues here as well.
I don’t remember your story. But trying ten things isn’t usually the solution. Try one thing and persist.
Can you give me the tldr?
Hello, has anyone encountered problem of installing hassos 9.3 in VM, it does not run if there is no Internet? I noticed that the time daemon does not start during loading, and page Preparing Home Assistant is displayed indefinitely. I didn't notice such problems before, what should I pay attention to?
Thank you! I'm already looking at the log files, he wants to download a lot. Yes, unfortunately without the Internet to install it will not work.
The home assistant OS needs to download the docker containers that include HA
Any install method needs Internet for the install itself - as will adding any integration (since it downloads the required Python packages at that time)
im literally dying trying to get my nginx proxy manager VM on HA to talk to my domain!! i keep getting bad gateway when it tries to redirect so I think i have to change something in my nginx conf except i have no idea where i can find it
Are you using the NGINX/NPM add-on?
Because, if so, you wanted #add-ons-archived
(also to visit ER, dying isn't good for you)
I have installed HAOS on the eMMC of a Raspberry Pi CM4 using the original CM4 IO board. So far it is working. As soon as I add an NVMe using an m.2 to PCIe adapter (as planned data disk) it doesn't boot anymore. When I look at the HDMI output, the OS crashes after it is loading env (current device mmc0) - the last message is Starting Kernel ....
Anyone got an idea what went wrong and how to fix it (except removing the NVMe 😆 )?
So, I took a StarLink, for a while it provided Internet. https://github.com/home-assistant/operating-system/releases/download/9.3/haos_ova-9.3.ova Preparing Home Assistant - displayed for an hour. Hmmm .... I have little time, Internet needs to be returned. I will try others , https://github.com/home-assistant/operating-system/releases/download/9.3/haos_generic-x86-64-9.3.img.xz
It won't help, you need functional internet until all of HA is downloaded and configured
Any time you add an integration, you need functional internet
thank you. im actuallly military nurse so dying def. not good lol
So, not literally dying then 😛
a second at a time 😉
Ain't we all
Internet doesn't helpHome Assistant observer
Supervisor: Connected
Supported: Supported
Healthy: Healthy
Apparently the reason is different
Click the blue dot, see what's going on
Is there a way to install nginx proxy manager on a no supervised hass version?
I am currently on synology docker hass
Sure, NPM existed before add-ons did
Just Google for instructions for your setup. If you're able to run any Docker container, follow the NPM docs for Docker
so I need a separate container with nginx ?
Yes, and that's all that add-ons are
isn't possible to just add it via ssh
I mean, you could try that, but as soon as you upgrade HA, or restart under certain conditions, it'll vanish
You don't ever install software inside a Docker container and expect it to persist
damm, supervisor makes it indeed a better setup than other versions
alright, I will pin a another container and look for intructions
is there any tutorial here or shall I google it?
It's not HA specific, so there's not likely to be any HA specific docs for it
(personally I prefer traefik, but YMMV)
If you're using Docker Traefik is pretty trivial to set up - it may not be on the NAS though
It's also already capable of getting SSL certificates, and handles websockets by default
http:
routers:
homeassistant:
entryPoints:
- "websecure"
rule: "Host(`homeassistant.example.org`)"
tls:
certResolver: zerossl
domains:
- main: "homeassistant.example.org"
service: homeassistant
``` and boom, up and running alongside my other containers
Its seems that the HAOS is not really compatible with ESX 6.7 ??? If i install it on VMWare ESX 6.7 then the addons page doesnt work and zwave docker image will NEVER download correctly.. however, if i install it and use it in VMWare Workstation - Everything runs fine! Addons page works, zwave docker image downloads....
in ESXi, other things dont work like Backups that page never loads either
Do you mean the docker command I was running? If not let me know what you mean. My command was the following (and i experimented with omitting the user_id stuff (this is my pi account user info-
docker run -d
--name homeassistant
--privileged
--restart=unless-stopped
--device /dev/ttyUSB0:/dev/ttyUSB0
--device /dev/ttyUSB1:/dev/ttyUSB1
-e TZ=America/New_York
-e USER_ID=1000
-e GROUP_ID=1000
-v /home/pi/docker_files/homeassistant_docker:/config
--network=host
ghcr.io/home-assistant/home-assistant:stable
So is it best to install via the OS method? I currently have set up pi hole and homebridge on my sd card and wanted to install home assistant but looks like I can’t have add on support or supervisor support unless I start over?
If you don't need add-ons that's fine
You can run without them. Integrations are more relevant than add-ons
Integrations being the accessories/devices? Add ons I didn’t no if I would need
#integrations-archived integrate Home Assistant with devices or services, or provide functionality within Home Assistant.
#add-ons-archived provide additional software or services, which an integration could possibility integrate with. Add-ons are for Home Assistant OS and Supervised only, other install methods can install software other ways.
Addons provide software outside of HA
Other servers, other services.
Like Adguard, MQTT broker, ZwaveJS, etc
Hmm ok maybe I’ll back up@my homebridge and start over anyways just incase I want them before I get too much set up@too want to lol
hey guys, i got this warning
[warn] Could not find /etc/default/grub or /boot/firmware/cmdline.txt failed to switch to cgroup v1
any solution_
?
Are you using Home Assistant OS, or something else?
Hey! I hope you guys can help me. I set up home assistant via VM. Downloaded the app on my iPhone and everything was working well. Then the app on my phone only provided me with a white screen. In an attempt to correct the problem, I uninstalled and reinstalled the app. Now my phone will not connect at all. Any suggestions?
Is the VM still running?
Yes.
Can you open http://homeassistant.local:8123/ from browser on your phone?
It's not looking like it. It's been loading for a while now.
..and it just timed out.
Then your VM may not be running, or the network can't find it at least
It's working on my PC in the browser. I will try and restart the VM
Well what's the difference network wise between the PC and the phone? Assuming your PC is running the VM, the VM may not be exposed to the rest of your network.
Is your VM network interface in "bridged mode"? Won't work in local-only etc
It should be, it's all connected to the same WiFI....Yes it is in "bridged mode".
what IP address does "homeassistant.local" resolve to from another device on your lan? (not the pc)
192.168.0.114
And if you try opening http://192.168.0.114:8123 on your phone?
Doesn't work. Same thing.
Seems like problem with your VM host
Then assuming this works on other devices, the problem is your phone. If it doesn't work on other devices, then you'll need to investigate the network situation on the PC with the VM. As well as check the Home Assistant logs.
Which VM platform are you using? Virtualbox , vmware, other? You may need to do some "enable external access" checkbox
Does the web url work from another PC (Not the one hosting the VM)
I'm using Virtualbox.
So it works from my other computer
I have also tried 2 different Iphones
Do you have private relay or a VPN enabled on the devices?
And are you trying Safari or just the app?
Hi all, I am preparing to transition my HA to a brand-new VM on VirtualBox, since the current production instance grew to some amazing sizes and I can't seem to get it to shrink down. In doing so, I'd like to make sure that my backup is not having any issues with MariaDB, as it's been the case before.
@brazen herald No, and I have tried the app and 3 different browsers.
The MariaDB seemed to have issues with transferring the database when a backup was being run, since it had to be fully stopped to create a proper backup. Has anyone tried running HA from a full backup with MariaDB lately and can confirm it to work??
Or should I stop MariaDB and create a backup, which will cost me about 1h of data stored?
And you don't have two home networks or something?
So installing home assistant there isn’t a desktop is there?
Do you mean a user interface on the device you install it on?
Ya I just have the dos looking screen is that right?
that should no longer be an issue https://github.com/home-assistant/addons/pull/2063
@brazen herald I have a 2.4GHz and 5GHz. Everything is connected to the 2.4GHz
Discord isn't like IRC, you don't have to tag people on every response. Keep in mind that every time you tag somebody, they get a notification ping. That can very quickly become annoying and people may block you.
When using Discord's new Reply feature it defaults to pinging the person you reply to, click @ ON to @ OFF to stop this - on the right side of the compose bar.
So from the dos like screen is where I’d install homebridge as well?
No, because you can't install HomeBridge on Home Assistant and don't need to
If you want to share devices to HomeKit: HomeKit Integration, if you want to add support for another device, check HACS
I’m wanting to use both for some different accessories
Yea there’s no need for a gui. Once it’s setup you just access it from your browser to its ip or hostname
That doesn't change the truth of my statement. HAOS won't let you install HomeBridge, and Supervised would, but it breaks the terms of installing Supervised
You misunderstand, she’s saying homebridge isn’t required because home assistant does the same thing without needing it
Sounds fantastic! Do you by any chance know if the fact the tables are locked results in the data being collected in the background and at the time the backup is finished the data is back populated in the DB? The reason I am asking is my automated backup at night takes about 45-60 minutes but I am not loosing any data there.
This makes me think the DB is either not locked or it is, while the data is being buffered for that time period.
No I get that, I’ve seen lots of posts of people running both HA and HB and was wondering how you go about it
That doesn’t make sense
The question is: what will you get by running both at the same time except multiple systems to maintain?
Because if you really want both, you'll need to do the container install of HA
Besides the question above, have there been any improvements to allow backups beyond 1GB? My backup is 6.9GB total, so it's always a pain to run a new HA, created an account, install SSH or Google Backup Addon, load the backup, etc. as compare to simply using the 'alternative' to login and loading the backup file.
I have also made a second VM with the same issue. The error code is: NSURLErrorDomain-1001
I was just thinking incase something wasn’t compatible with home assistant, someone on Reddit had said it wasn’t bad incase HA didn’t have one of the devices you were wanting to add. I guess I could worry about it then if that happens. Is it possible to set up my pi hole using the HA OS?
No, but there is an Adguard addon
I have yet to find a single device that HA doesn't support and HomeBridge does.
You can integrate an existing pihole setup on your network or add an unofficial addon to run it on haos
Ok well I’ll give it a shot then, I’ve only got a few devices so far but was using HB no issues so will try this out
You’ll never look back man
Ok thanks I’ll take a look for that. Seems confusing so far just cuz I’ve never used a computer system with out a desktop OS lol
Plenty of good YouTube tutorials for almost anything you’d want to do
You can google home assistant [thing you want] and get pretty far
sorry, had a situation
i installed this one
https://www.home-assistant.io/installation/linux#install-home-assistant-supervised
Ok just went out for lunch so when I get home I’ll start messing with it. So just to be clear in my understanding I can still control everything with the home app as well on my phone? Until I’m used to HA and my wife will prob only want the home app lol
Can't understand how to fix this, any help? already checked the learn more link
Unsupported system - AppArmor issues
System is unsupported because AppArmor is working incorrectly and add-ons are running in an unprotected and insecure way. Use the link to learn more and how to fix this.
You add everything to home assistant then use the HomeKit integration to share devices etc with HomeKit and then they show up in the home app
You’re running an unsupported install method and it’s rare people are willing and/or able to help you with it
Ok thank you
It’s a ‘go have fun wild wild west’ install method with supervised man
I followed the guide step by step :/
It breaks over time if you aren’t literally keeping up with updates and requirements
Your best bet is probably copy files off, install latest, restore files
And then watch it break again in the future
i justed formated and installed, but it's ok, im not to worry to be honest, and about this on?
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.
any idea?
Either use the container or haos install in a vm
those are the 2 error that i got
im using Raspberry, but can't use OS, whats the best way ? (if possible with addons, thats the problem )
HAOS
Can’t use os?
HAOS, because I need rasp for other thinks, or it's ok with the HAOS and add my thinks on in? never used HAOS tbh
What other things?
container with node apps for example (home made)
and maybe plex server with all the stuffs on it, still not sure if i will add on the pi or not
Pi and Plex isn’t great man. Pi4 with ssd boot is an okay option for haos or container install
I think you’d like the container setup it just won’t be able to run addons (which are just containers themselves)
So you can run your own containers that are equivalent to addons just more diy footwork
I had plex on PI4 and was working well ...
i will check that thanks, but, HAOS work with addons / containers?
If you want to run your own containers then HAOS isn't an option
Only the Container install method (native Docker) is
A kind reminder, since it got buried....
On yet another note, has anyone been able to run the proposed VMbox command to minimize the VDI file size effectively:
https://community.home-assistant.io/t/installation-says-by-default-virtualbox-does-not-free-up-unused-disk-space/426814
I've tried but I am stuck with the syntax.
My VMs live inside E:\VMs
Should I navigate to my VM folder or the Oracle\Virtualbox folder, as indicated?
Second, the <VM name> should be kept inside the parethesis as in my VM name is Home Assistant so it would be <Home Assistant> and controller "IDE" (that's the default name).
anyone know the secret on having HAOS run correctly inside of ESXi 6.7? i can get it to run and boot but nothing works... addons page, adding zwave plugin, backup page, etc.. are all BLANK and come up with an error.. it seems to work fine when running in VMware workstation, all those items work as expected.
"with an error", mind sharing?
Nah
the error mostly is: Could not load the Supervisor panel!
with troubleshooting page.. check the observer, try a reboot, check system health
I don't know esx, but did you see https://community.home-assistant.io/t/hass-io-on-vmware-esxi-6-7-step-by-step/151419 ?
no matter what i set the hostname to.. when i click "check the observer" it sends me to http://homeassistant.local:4357/
tried hard-refresh on browser?
i have set the hosts file on the vm itself and on my pc, that has no effect
I assume the network stuff comes from the hostname at http://homeassistant.local:8123/config/network
i would agree.. but changing that doesnt change the link on the error page - still links me to http://homeassistant.local:4357/
I never used observer url before - but hard refresh?
HAOS works OK on proxmox, which is functionally pretty similar to ESXi
Both type-1 hypervisors
hard refresh doesnt work.. i have reinstalled 20 times, same issue
if i install and run on vmware workstation player - everything works
i can then copy stright to esxi - the same working vm... quits working.. same issues
sorry I have no idea on esxi. sounds like either a networking issue, or maybe not allowing privs for supervisor
its so weird.. i even installed a full Debian linux install and deployed supervisor.. same issue
i have this virtualization server for a reason.. so i dont have to have a bunch of PI's and crap everywhere. i cant belive i cant get this vm working
Seems like a problem with esx+supervisor. Sorry no idea.
Why not just do a docker install?
checked esx logs for the vm to see if anything "denied"?
i tried that too.. ive been trying to install this damn thing since Friday
i dont see anything really in the esxi logs... all looks ok
Sorry, certainly seems like an ESX problem?
couple of months ago set up my pi 3B+ to run on SSD instead of microSD due to all the warnings I was reading about SD cards always failing.
Anyway, my SSD packed it in yesterday! TF for backups hey!
Going back to SD card to wait until my yellow finally arrives
I found the rpi3 was just a bit slow, and a "real" box doesn't use too much power anyway. I measured it about 8-9W
define 'real' box?
problem is that I already sourced CM4 and nvme SSD for yellow, with estimated august delivery. So was alway sjust trying o get the 3B+ by. Yes, it's slow, but I'm also not asking a whole lot of it.
I bought an old lenovo m910q for like US$120 second hand, with SSD and 16GB ram. Small, fast, lowish power.
Yellow is still gonna suck
I’m glad mine got delayed until I gave up and cancelled then moved to a used laptop
I ran proxmox on an old (intel) macbook pro for a couple of weeks, worked fine.
But I moved to a sff box because it had good built-in NIC, remote management/console, and I run a few other things on it.
Put macos back on the laptop to run my CNC again
One of my requirements is small and discrete, which is why I like the idea of Pi/yellow. I really hope a NC supported solution doesn't suck. I have an 8GB Cm4 and am not doing video/media streaming or anything, simply automation, energy and dashboards
The lenovo stuff is like 7" square by 1.5" tall. Pretty small.
But your requirements sound very low. I did the same stuff on rpi3 no problem.
I don't do any "media stuff", but do run my pfsense (firewall), and a couple of tiny apps like uptimekuma and heimdall dashboard, along with whatever box/vm I feel like using for some experiment: like node-red, or ubuntu, etc.
NUC is another common option for ultra small form factor
Kinda glad there was a Pi shortage, otherwise I 100% would've went w/ a Pi and likely have been disappointed
Yeah same
I tried to order a yellow or something, saw the lead time, and went "forget it"
Yeah, I usually need things "now" 😛
A guy on #hardware-archived bought a lenovo box like mine a few days ahead of me, so help me setup a few things
Took like 3 days from ebay
I don't need it as small as a rpi, so nuc/thinkcenter size is fine, and way faster.
I feel like I'd still end up getting a case for the Pi and whatnot.... and then if you add an SSD as well, it's going to be nearly the same size and likely not as neat
Exactly
the lenovo etc is about four times faster (geekbench) than the rpi4
https://www.cpu-monkey.com/en/compare_cpu-raspberry_pi_4_b_broadcom_bcm2711-1796-vs-intel_core_i5_6500t-525
I still like used laptop as a suggestion too
You can buy used with a broken screen on eBay for dirt cheap
For sure, especially for trying things out and all that
and with proxmox installing things is super easy, especially if you use scripts like https://tteck.github.io/Proxmox/
What lenovo hardware is this?
That’s unnecessary for the majority of users
Yeah, true, but I play with other things, and don't want a lot of boxes.
Mostly the box also runs my firewall (pfsense)
Haos works fine and needing to manage a hypervisor isn’t what I consider a good option
True, I feel like with energy costs everywhere if you can go for some type of SFF/USFF, it's going to pay for itself for what we're doing with these setups
Maybe I'll look at NUC, just a shame when I've already invested in the hard to find hardware (harder again when you're in australia)
you can run 4 pis and still have lower power usage than this intel box
I'm in australia dude
Crikey
mstone: i disagree, was surprised to find the intel box only pulls 8-9W
it for sure does not when you run pfsense, ha and whatnot on it
you're still talking bare minimum os
sure it does, but I don't have the smartplug installed right now. maybe will do it later.
That would be pretty incredible.
anyway, horses for courses.
7.5 W TDP (PL1) 35 W
Even at crazy energy prices what are we talking here? An extra dollar a month?
Ah, where did you find this lenovo box you mentioned?
He’s not suggesting a dual Xeon
@jagged dock I got https://www.ebay.com.au/itm/334203502529 but it was AU$189 at the time (a month or two ago). Look around the seller shop.
Plus ebay keeps giving me coupons or discount codes or whatever
which save a little
I'm not too energy concerned other than I generally like efficiency and don't want to hear fans running in what is a quiet location now. Got huge solar and drive an EV so small computer loads are inconsequential.
Actually, I remember, I paid a bit extra to upgrade from 8GB to 16GB
Nice. much cheaper than the yellow stuff.. sigh.
I can't hear the box at all, except if I put my hear on the case. It's ssd and the fan barely kicks on.
I was pointing out mstone complaining about wattage when it’s still super small
we pay about AU$0.30/kWh and he said he's got big solar, so won't be paying during the day...
well I'm making $ every day with solar, I don't like to get less money off it
hahah, we get paid a pittance for export. $0.05/kWh
I export everyday from about 0700 until 1900 right now
Australia gov hates its citizens almost as much as us Americans and our gov
Update to my issue earlier. I left, came back, and everything is fine now. Who knows what the problem was. Thank you guys for your help!
heheh win
Google tells me that boxes like the lenovo/dell/hp use about 2-3 times more power than rpi at idle, and maybe up to 7 times more at 100% load (probably never)
https://www.techaddressed.com/reviews/mini-pc-better-value-raspberry-pi/
never said anything else, but saying you run at 8-9w is just plain wrong with multiple vms on it
Next time I'm up the ladder where the box is, I'll add a smart-switch to mine, so I can see what it actually uses. My box reports about 3-4% cpu usage, and most of that is haos (some fairly high-frequency polling integrations).
They're mostly just containers, except for haos and pfsense. But 3-4% cpu usage isn't much more than idle. Will put a plug in when I'm next there.
I don't remember if haos was installed when I measured 8-9
17 kW here with $0.20 feed in. makes up for the high energy cost easily
I'll put some thought into the lenovo box or a NUC after work today.
I've got 21kW but only get $0.05 so I only make a couple of bucks
https://imgur.com/L2ZxTSz
And my array is split east/west, so it gets power for longer, but probably less of it
where are you at location-wise?
I'm in cr*ppy-weather victoria 🙂
That's a big system! I'm all north, shade from 3pm but still maxing at 112 kWh or so ona clear day this time of year. MNC NSW. Getting OT here but happy to talk solar on the side.
#energy-archived for that 🙂
Have to get my #installation-archived back up and running so I can share stats from HA! I don't have a spare sd card lying around.
$0.30/kwh cost is excessively high unless you're on ausnet. Since the rest have far cheaper default offers
I'm happy to move if I can find a better plan? I pay 0.3346992 peak, and 0.1965997 off-peak and 0.054 export.
With $1.0279/day, and 13% discount.
I did, got the free $, I don't think anything was much better... But I was in the middle of getting 3phase+solar connected at the time, so didn't want to rock the boat.
Since the default offer is usually the highest and that's 0.3 for ausnet, rest are all 0.24 or so (this is for your supplier, not who you purchase from)
My supplier is United Energy
Funnily enough, I wrote a blog post about scraping their API for power data, just before they turned it off. Sheesh... https://innablr.com.au/blog/applying-cloud-sustainability-principles-to-non-cloud-choices/
😦
Energy used last week: 58.44 kWh
Energy exported last week: 468.96 kWh
Still get big bills because when it's cold/wet/cloudy, we use far too much power. And when it's hot/sunny we use basically nothing.
@manic bane not a lot in it https://imgur.com/SSok1M8
Their energy easy portal was old and crappy though. And a rip off from a few others portals too lol
I would still check what each one is charging per kwh. Found the general all year round price to be inaccurate
energyeasy is afaik a third-party product
Yeah and now they've got their own
I need a calculator to look at my historical peak/offpeak/shoulder usage, and multiply by the prices...
Hi. how to fix constant reboots of my HAOS installation? It is installed under virtual machine managed by proxmox. Hardware - minipc using N5105 and 16gb ram. It reboots a few times a day
Need to figure out the cause. Mine doesn't reboot.
is it running out of memory? how much did proxmox give ha?
Logs would be useful too
8gb is available to ha. Before a switched proxmox kernel to edge it was not rebooting, it was halted
So switch back. 5.19 has a few issues.
How to see relevant logs? A log viewer addon shows a clear log just after next reboot
You will probably have to use something like the vscode addon to view the log with a -1 in it
(aka before the last restart)
On that note, memory usage, is it normal for HAOS to consume all that it's given? It's a VM, is it just using it for cache whatever, and it's not in danger of OOM? Mine says 95%
Are you talking about what proxmox says?
are you asking me (95%) or Alexander?
Cannot add screenshot here. At time of halt a ha linux kernel provides a call trace and force a halt
use imgur to upload
Will try, thanks. Any log file a can see using terminal? I’ve checked /var/logs. It is free of any file
In terminal it's in the config directory
You
Hi Lyricnz, wanted to let you know that I finally managed to install HAOS, thank you and others very much for your support!
Excellent
Yes, reported by proxmox. https://imgur.com/Fwzij0w
Hi all, News question here and hope this is the right channel for this question. I had Duckdns working fine before and could reach my HASS instance via their service. I switched Router and thus redid the port forwarding for port 8123. But now, when I navigate to my DuckDNS-url, I land on my router's login screen. Obviously, I don't want this to happen. Does anybody know what I should do to land on my Home Assistant instance?
When I add :8123 to my DuckDNS url it does send my to my Home Assistant instance.
It's possible your new router can't forward the admin port it uses. Does it support changing the port?
Also check what it does if you're not on the same network (eg on mobile data) as your router login page publicly accessible is not good
Yes I tested this on my 5G network. It goes to the router. So now I disabled the port forwarding. For obvious reasons 😉
I will check the forum of my ISP to see. Thanks.
Check if your router has an external management setting
But also just get a new router. Put your current one in bridge mode (if you even need it)
I will get a new Router soon. I just wanted to have my home assistant app working again.
Cant you use the old one for now?
That was my old ISP's router and I had to send it back to them.
Will get a good Router that fits within my Unifi ecosystem.
Thanks for the help!
Hello, how can i edit the HA configuration file? (I installed HA using docker and portainer)
Edit it with any editor of your choice, outside of the Docker container
You can also use something like https://docs.linuxserver.io/images/docker-code-server if you want
But how can i acces the file?
If you mapped the container correctly it's on the host
You can access it with any editor
I created a volume
Right, so ... it's on disk there
Use #installation-archived message and the same volume
Yes, but ... you've told Docker to use a Volume, you shouldn't be messing directly with the folder
um, so how can i edit it then?
Copy your files out of there and create a proper volume bind mount for the container. Recreate the container with this and profit
so "proper install" shouldn't be in volumes?
It's even what's shown in the docs
awesome, i followed some stupid tutorial
You should always check what the docs say - many tutorials are misleading, outdated, or just boneheaded
Some of them set you up for failure
👍
okay
i think i will just do a clear install of HA
Are "the automation guy"'s videos good?
Videos get outdated very quickly and are hard to update a small section of
No, sorry, i didn't mean the tutorials how to install or do something
All videos date fast. HA releases new features monthly.
👍
Videos being "good' or "bad" is objective, but never assume that because someone has made a good video that all their things are good. There's more than one way to solve every problem.
Use things as inspiration, not a template.
That's what I meant, English is not my first language, I must be expressing myself badly, sorry
It's ok, lots of us speak other languages 🙂
version: '3'
services:
homeassistant:
container_name: homeassistant
image: "ghcr.io/home-assistant/home-assistant:stable"
volumes:
- /PATH_TO_YOUR_CONFIG:/config
- /etc/localtime:/etc/localtime:ro <------ What does the ":ro" mean?
restart: unless-stopped
privileged: true
network_mode: host
To format your text as code, enter three backticks on the first line, press Enter for a new line, paste your code, press Enter again for another new line, and lastly three more backticks.
```yaml
example: here
```
Don't forget you can edit your post rather than repeatedly posting the same thing.
Not really, same applies to it
Stop it, copy/move the config folder to somewhere sensible, update the config, start it
yes, but i will have to set it up again
No
I just don't get it, how can i move the files
Ah, linux basics...
cd /var/lib/docker/blah/blah
sudo cp -pr * .??* /my/new/location
Or
yes, i know that
Ok?
Then check the config for the container
docker inspect -f '{{ .Mounts }}' containerid
Error: No such object
oh missspelled
sorry
Okay once again, i am stupid
it's there
Stupid is unable to learn, you're finding your feet 😉
Everybody screws up at first, and most of us never stop
If you never make a mistake it's because you're not really trying
Or, to quote one comic, Failure is not an option. It is mandatory. The option is whether or not to let failure be the last thing you do.
adguardhome:
image: adguard/adguardhome
container_name: adguardhome
restart: unless-stopped
privileged: true
network_mode: host
volumes:
- /opt/adguardhome/conf:/conf
- /opt/adguardhome/work:/work
For some reason it doesn't take the config file when pasted there
Did you docker compose up -d adguardhome?
yup
Then I'd inspect the container and ensure it really worked
[{bind /opt/adguardhome/work /work rw true rprivate} {bind /opt/adguardhome/conf /conf rw true rprivate}]
Then check the container logs to see what's going on
Nothing there... last log was 1 hour ago
there is more than 1 way of doing this
you might get more luck with docker cp homeassistant:/config .
nvm, i just set it up again
but now i am just wondering where are the files stored
you could save a lot of time by just following the official documentation
instead of working around it
the official documentation does not provide any docker-compose help
I am talking about adguard home
what is the problem with adguard?
That after reinstall, it doesn't talk with already existing config file (pasted from before)
whats the real problem? setting up it up new is just setting a user/pw
all my settings
and yeah, this channel is generally about homeassistant, nothing else
your settings have to be in /opt/adguardhome/conf
they are
(at least if you use the compose from above)
well then you didn't read the official documentation again
says /my/own/confdir:/opt/adguardhome/conf
far from /opt/adguardhome/conf:/conf
The bit to the right of : is the path inside the container and you don't change that
volumes:
- /data/adguard/work:/opt/adguardhome/work
- /data/adguard/conf:/opt/adguardhome/conf
``` that's from my own config,
https://docs.docker.com/compose/compose-file/compose-file-v3/ is a good source to read
thanks
your problem is mainly another benefit of using docker. applications on linux can put their configs pretty much wherever they want, by using docker you can have them all in the same place if desired (I recommend this). so next time you set something up you don't have to look in a bazillion places where everything is
inside the container they will still be in different places (/config for ha, /opt/adguardhome/conf for adguard e.g.)
For migration, how should I get pass this page?
https://i.imgur.com/2ZOtGLL.png
#zwave-archived can help with Z-Wave
I've been asking in reddit, official forums and discord, but no response so far :/
You want the #zwave-archived channel, that's where the Z-Wave people hang out
Please don't crosspost to get attention, you'll get the wrong kind of attention
how should I get help when there is no response?
You waited two fecking minutes from your post in the Z-Wave channel
Why don't you have a little patience?
this is not the first post
You realise the whole world isn't lurking here, waiting for you to post and to spring into help
It is here for this problem
Many people don't also lurk on the forums, Reddit, etc
Timezones also exist. For most of Europe it's the middle of the working day. For the East coast the morning, and the West coast is still asleep. Australia, etc.,... not sure, but either way 😉
Posting the same thing in multiple channels doesn't mean that people will be more likely to answer, quite the inverse. Plus, if it was in the right channel and no-one answered, posting it in the wrong channel will just muddy the waters.
maybe if the documentation is better worded, it might be easier for everyone , just a sugestion
Good news, once you know the answer you can submit an edit to help the next person
This indeed is how most of the docs get improved - people see things that could be better and help out
Guys, I've moved my HA install from dedicated on Raspi 4 to Docker. On beforehand I've created a complete backup. (.tar file). Can I restore tis to the new docker HA?
Well, I am not that of an advanced user (yet), if I understand correctly I have to untar it and store these files on the SD where docker is running?
And then replace those files towards the location on DOCKER?
I have not indeed, first time on Docker
Ah, if you used HAOS before you do know that by using native Docker you get no add-ons?
Yeah I do
That's fine if you're happy to learn about Docker, but if you want the same experience you're not going to get it
I understand that it wont be the same experience, i'd really love to start working with docker (but not buy a second pi for the same). I just cant seem to find any way to restore my back up :/
So, pick a location that's going to hold your Docker compose file, and your HA config, and all the other containers, for instance my Docker host has:
/data/docker - compose file
/data/zigbee2mqtt - Z2M config folder
/data/homeassistant - HA config folder
etc
Then the https://www.home-assistant.io/installation/linux#docker-compose would look like
version: '3'
services:
homeassistant:
container_name: homeassistant
image: "ghcr.io/home-assistant/home-assistant:stable"
volumes:
- /data/homeassistant:/config
- /etc/localtime:/etc/localtime:ro
restart: unless-stopped
privileged: true
network_mode: host
Mine is nearly identical, but I called my folder docker. 100% recommend that approach
I can back up /data and get everything
This should be /home/piadmin/HomeAssistant/config
If that's what you want
Remember the :/config part doesn't change
You only change /data/homeassistant
Yeah that's fine I guess
Well, there's no guesswork here - if you don't use :/config then your config folder doesn't get mapped correctly, and when you restart or upgrade you lose everything
I think I do not follow completly
My HA is now running in Docker correctly, and I've mapped the following volumes
volumes:
- /data/homeassistant:/config
^^^^^^^^^^^^^^^^^^^
This is the path on the host
Volumes
Host/volume Path in container
/config /home/piadmin/HomeAssistant/config
/localtime /localtime
volumes:
- /data/homeassistant:/config
^^^^^^^
This is the path inside the container
You got it the wrong way around - what does your compose file look like?
I've set it up in Portainer, dont know where to find the complose file

Can i pull the composefile using SSH?
No idea, I have a strong dislike of Portainer and wouldn't touch it with any length of pole
You look though to have switched the paths
Host: /home/piadmin/HomeAssistant/config
Container: /config
Unfortunate I cant sent a screenshot, let me check if I can get it any clearer 🙂
Please use imgur or other image sharing web sites, and share the link here.
Image posting is blocked in most channels to discourage people from sharing text as images. Sharing text as images assumes that everybody sees the world as you do, which isn't the case. Some people are colour blind, or have visual impairment that means they can't make sense of an image of text.
Yeah, you flipped the paths the wrong way around
Host: The computer running Docker
Container: the container, duh
Shouldn' be to hard to remove the container and recreate the same with correct paths then
Let me check
docker has the power to create stuff while you spin up those containers, so you will end up with a new folder structure for every wrong attempt 🎉
@humble mirage https://imgur.com/a/Uo4lg7B
@gaunt forge Discord isn't like IRC, you don't have to tag people on every response. Keep in mind that every time you tag somebody, they get a notification ping. That can very quickly become annoying and people may block you.
When using Discord's new Reply feature it defaults to pinging the person you reply to, click @ ON to @ OFF to stop this - on the right side of the compose bar.
-> Repairs -> 