#installation-archived
1 messages · Page 40 of 1
It finished, amazing how seemless it is to move to new hardware
it be like that
my 650mb restore took about 2 hours
Yeah that was my impression as well - I've so far moved from supervised on rasbian to ubuntu then back, then to HAOS on bare metal on RPi 4 and most recently bare metal x86. Each time I just do a fresh install for the corresponding platform and restore backup, no problems whatsoever
Noob here. Just got the Zooz 700 Z-wave usb stick. HAOS sees it and adds it to the Z-wave module. However, it's not discovering anything. Has this occurred to anyone else ?
Stupid question. I’ve been using HAOS on a raspberry pi. I spent a good amount of time working with it. I’m looking to move over to the HA container on Docker. Is there anyway to make a backup on the OS version and just load it up on the Docker version so I don’t have to start all over?
Yes
Just extract the homeassistant.tar.gz file from the backup into your new config directory
This was you last night 😂
I’m just trying to get everything moved over but am struggling.
Noob here Just got the Zooz 700 Z wave
I'm trying to use a data disk connected via USB. HA runs on a VM on top of Fedora Server. I have made the necessary configuration on the VM that this host USB disk is available to the VM (guest).
Now I'm trying to understand how to add this storage device to HA (Settings > System > Storage) via https://www.home-assistant.io/common-tasks/os#network-storage
but it looks fairly complex. The be clear I'm simply trying to add extra storage (as I want to store footage from my Reolink camera) and not move the data disk.
How to go about this?
Buenos días
estoy valorando la opción de comprar el kit yellow de home assistant
es recomendable? se que tiene bastante tiempo en el mercado y no se si van a sacar una nueva versión o si está algo obsoleto
Good morning. I have to complete Home Assistant configuration but, my machine is interfaced outward internet by a proxy server. How can I configure Home Assistant to include proxy authorizations settings before to try an update of operating system, or Core or Addons ? Best regards.
You have to install a proxy like a complement and include the proxy configuration
I supose nginx proxy server is a good option
I have just configured proxy server in my network. I don't know where I have to set proxy address, or proxy username within Home Assistant configuration. In practise Home Assistant aren0t use proxy.
Why you are configured proxy server in your network? do you need it for other use than home assistant?
you can use https in home assistant
is not easy but is possible
I recommend you to use cloudflare with a register domain
and connect to your home internet trought zeru trust service
one problem please. How can configure proxy settings in Home Assistant ?
you can found more info in youtube (search this: Hassio cloudflare)
where have you install home assistant? pc or rasperry?
it is installed on a server
zero trust could be complicate, but home assistant has a cloudflare integration (HACS) where update your dinamic ip in a subdomain
and you can include proxy server in cloudflare in order to hide your real ip address
No. Not Cloud only local use for security
so if you only use in a local lan, I think you don't need proxy server
anyway, I never installed proxy in a home assistant
Proxy server shadows network implementation outside of your office. Is important against attack surface
Ok thank you for your time
How to solve this problem, I installed homeassistant on a mobile phone without root permission.
someone has installed home assistant in windows plattform without virtualization?
Howdy all - I've set up a brand new HA instance, complete with the HA script available on the HA website, and am getting the same systemd-journald: failed to rotate [...] read only file system error again. Clearly this is some Proxmox thing, then? Any ideas welcome, thanks in advance!
What script?
I can't help much, but I can say I ran HA for a while in VirtualBox on Windows and it frequently had hangups and locked itself up. Strong don't recommend, for what little it's worth towards your query.
oops wrong brackets
It doesn't embed anyway, bacon saved
You can use <link> to not embed/preview
Yeah, I meant to do that
I didn't want it to boost the page too much, but embeds are off so it's same same anyway
You could run dmesg -T | less and scroll (arrow keys, q to quit) through it from the start to see if there's anything useful.
You have to enter login first to get ouf of the HA CLI
I wouldn't know what's useful and what isn't, none of this makes sense to me - I'm not familiar with Linux or it's various common run-ons
Does scrolling through this start you at the top, by going down once?
it appears to start from the same time as we see most recently in the list
I can definitely see variations in the messages - two main things happen - but nothing clearly different
Pasting an example of the two kinds of messages in that link
What type of storage are you using and how is it connected
who is your daddy and what does he do
NVMe SSD internal on an Intel NUC (yep, the one HA was baremetal on last time we spoke)
500 GB NVMe SSD and I do have a backup SATA disk in there but it's not initialised at all
HA has it's default 32 gb plus I expanded that to 64 by resizing the disk, but that's all on the same internal disk
I should have answered the most important questions first my bad
I just installed HA on a RPi and I get the logs when I press the blue button, but it's not clear how to get past the logs and onto the rest of the setup. Can someone tell me how to get to the next step, please?
It shows successfully running in the logs.
Are you expecting a gui?
Go to homeassistant.local:8123 on your lan or ip:8123
Specifically what do the logs say
If they've pressed the blue button they're presumably in the GUI webpage already
For all I know he’s pressed a blue button on his toy car on his desk
Fair point
Just migrating from a pi3 to something with a bit more beef. there I use cockpit and cockpit-machines to start a HAOS vm.
but here I start having issues with OS and qemu-virtiofsd. I try to mount a folder (/opt/homeassistant/config) to /config in the VM. sadly I can not find a way to mount the shared folder. (ha-cli and terminal addon)
any ideas welcome
I dont think that I can get any acess in the ha-cli, when using the addon, and querying mount, I also can not see the shared folder, maybe I have to install findmnt to see
nah, findmnt also dont dind it, probably because the addon is already a container...
you... are mounting directories in a vm? usually thats not a option without guest tools and sounds like you are setting up a container
so HAOS does not have guest tools then..
Nah, just using addons
it does have guest tools afaik
but they might not be compatible with cockpit is what i want to say
cockpit-machines is just a wrapper around qemu and libvirt
urgh, then some unholiness with sshfs and systemd I guess is my only way
why don't you just use a proper disk? and restore the backup like everyone else? 😄
Good Morning. trying to install HA on a RPi4 (HA OS style). Guess I'm seeing the same 'issue' as @safe thunder before. In the logs (blue button), errors are showing up, related to 'https://ghcr.io/v2/' . I've tested it in a normal browser, and there seems to be some authentication error (unauthorized). Is there perhaps some version missmatch between the Pi Imager, and newly release version?
you mean dding the qcow file out to a thumbdrive?
no?
no backups in ha container via compose 😄
it gets more confusing
then I dont what you mena with prper disk
sorry, mixing here,
old system, pi3 with docker-compose
new system fedora, and moving to HAOS
no other mounts
ah, yea, my current mount disk is the qcow file stored on disk
Doing a restore from a full backup onto a new clean installation on a different system. I selected restore from the onboarding page (didn't create an initial login). I know this could take a while (4gb backup tar file), but is there any way to check and make sure that it's actually restoring? I've read that it can take anywhere from 15 minutes to 2-3 hours.
wasnt this what you meant here?
yes, no other "mounts" needed
kk, was hoping to havethe same stucture as I'm used to with cotainer, maybe I ll go down the route with sshfs with the terminal addon to have this functionality
thanks a lot! saved me a couple of hours banging my head against the wall
you will hit limits I guess, thats not how haos was intended. its pretty much one way with the config files in haos, you only get backups out of it. but hey you can try and see maybe it works
Hello there. First post here so be gentle- also i'm a bit of a linux idiot. Just upgraded from rpi3-> rpi4. Running on a usb ssd (Netac Portable SSD 250GB). All seems good but lost access to pi completely last night and had to reboot this morning. Now lost access again through web (but can ssh in). Can someone direct me how to get the logs off it and where to put them so someone who understands whats going on can help me get it stable?
you first have to tell us how you installed the pi and Homeassistant
Mine isn't showing any errors, though. It's just sitting at the log file screen. Nothing in the instructions say how to get past the logs.
@safe thunder If you look through the logs, do you see any 'errors' as I did?
Im my case, that's probablu why my install got stuck.
Oh you're right. Looking farther down in the logs, I do see the same error. Get "https://ghcr.io/v2/ error
navigating to that page says there's an authentication error.
Just as mine. Bummer.
Looks like there was a code push 9 hours ago: https://github.com/home-assistant/core/pkgs/container/raspberrypi3-64-homeassistant
Maybe mine was trying to pull while the code was being updated? I'm rebooting the RPi to see if it tries again.
Updates are 'semi' atomic, so if you do a pull when someone is pushing there should not be a conflict. My guess is more that the HAOS tha'ts found on/in the Pi Imager needs an update.. Alternative, somebody messed up the read rights on the folder... (ghcr.io/v2)
Invalid time zone passed in. Valid options can be found here: http://en.wikipedia.org/wiki/List_of_tz_database_time_zones for dictionary value @ data['time_zone']. Got 'Asia/Shanghai'
I installed homeassistant on a mobile phone without root, and it prompts the above when I configure it.
Looking further down, I see this:
"23-08-05 03:07:45 WARNING (MainThread) [supervisor.jobs] 'GitRepo.pull' blocked from execution, not enough free space (0.2GB) left on the device"
I used a 4GB card. Now I'm going to use an 8GB and try again.
32G are the recommended minimum (for HAOS) AFAIK
as dfirected by the website- burned an image of 64 bit rpi4 HASS to the SSD, then booted from it. Then restored my previous backup to it.
thats then HomeAssistant OS or HAOS, when you can do SSH, that probably means that your system restored and you connect to the addon
try to run ha core check for starters
logs should be in /config
says processing done. didn't come up with anything
its all accessible again now. Weird. Was unable to connect to it for about an hour
Resolved this issue by following this guide: https://community.home-assistant.io/t/solution-to-a-hanged-preparing-home-installation-screen-during-new-install-on-a-raspberry-pi-4-dns-fix/273803
Just wanted to post the resolution in case anyone else runs into the issue I had!
should I install home assistant with docker or home assistant core on ubuntu?
The former. There's no point in the latter
so what's the difference between docker or OS?
I'm installing this on a Microsoft Surface w/ ubuntu latest
Would like to use the surface as a server and control panel
@quick oracle HAOS is a minimal OS based on buildroot for embedded systems. It includes docker and is managed by the supervisor for managing other containers (add-ons). With docker you are the supervisor and choose your own containers if needed
Hello! I am installing home assistant on an Ubuntu server. I had it running in a virtualbox vm, but after a server reboot, I cannot get it up. All the errors point to virtualbox-drv or dkms. I'm not a Lennox person, so I don't know what those means. I have tried to reinstalling those packages, with no success.
This is the error message I'm getting:
haven@haven:~$ sudo VBoxManage startvm "homeassistant" --type headless
[sudo] password for haven:
WARNING: The character device /dev/vboxdrv does not exist.
Please install the virtualbox-dkms package and the appropriate
headers, most likely linux-headers-generic.
You will not be able to start VMs until this problem is fixed.
VBoxManage: error: Could not find a registered machine named 'homeassistant'
VBoxManage: error: Details: code VBOX_E_OBJECT_NOT_FOUND (0x80bb0001), component VirtualBoxWrap, interface IVirtualBox, callee nsISupports
VBoxManage: error: Context: "FindMachine(Bstr(pszVM).raw(), machine.asOutParam())" at line 721 of file VBoxManageMisc.cpp
What happens if you try to install the packages?
Hi all! I'm trying to install HA via Docker on Windows 10 and I'm using the online installation guide. Currently I think I have docker running since the image is showing as "Running" in the Containers menu. However, I can't seem to access the Web UI? Any advice? I'm attempting to access it via "127.0.0.1:8123" on the local PC.
It seems to be listed as an option for under the Generic x86-64 menu though, would that not work?
(i.e. I have an Intel NUC and would like to use the Home Assistant Container option listed there)
That's not Windows and not a container install
It's completely unrelated to what you said you were trying
😅 I'm a bit baffled, sorry! I'm new to most of this
Install Linux on it and you can use the container install
Or just use the generic x86 install of HAOS
I suggest the latter
Either way, you'll replace Windows
Hmmm. That's unfortunate. I'd have liked to keep Windows!
Thanks for your help though : )
Hi Folks, I first posted a couple days ago but didnt really have time to debug then sorry. I've been unable to upgrade to HAOS 10.x ever since it came out. When I first started I was running HAOS9.5 and tried upgrading to 10.1 I think it was at the time. The upgrade didnt work because when i rebooted I was still on 9.5. I can't see anything in my logs but might be missing something. I've tried upgrading to 10.x several times with no succes. Sometimes it sends me back to 9.4 which I don't understand. I've also tried a clean install of 10.x and restoring a backup but that sent me back to 9.5 again. I also sometimes get errors about AppArmour but don't understand how to fix that. It seems to go away by itself once I attempt an upgrade again. This is all running on a Pi4 using HAOS I just rebooted my Pi to get a clean log - supervisor log is http://pastie.org/p/2drTQII82HjyANIEMUGY0A thats currently running 9.4 and says I can upgrade to 9.5. Upgrading to 9.5 was succesful this time and new supervisor log is http://pastie.org/p/70oHasaPD1ivCgDr2TmYQ2
When I started homeassistant for the first time, it prompted me that there were some code errors.
Invalid config
The following integrations and platforms could not be set up:
usb (Show logs)
bluetooth (Show logs)
zeroconf (Show logs)
cloud (Show logs)
assist_pipeline (Show logs)
mobile_app (Show logs)
default_config (Show logs)
Please check your config and logs.
What causes this?
If you're installing it on a phone then who knows
Could be anything
I installed Python, but its log error points to Python.
2023-08-06 06:24:20.197 ERROR (MainThread) [homeassistant.loader] Unexpected exception importing component homeassistant.components.assist_pipeline
File "/usr/local/lib/python3.10/dist-packages/homeassistant/components/assist_pipeline/init.py", line 26, in <module>
How did you install HA. Which method?
Also why on a phone
Ha official tutorial, my mobile phone simulates ubuntu system. I have a mobile phone with good performance. If it is discarded, it will be wasted. I want to use it as a waste.
Which tutorial did you follow, there are 4 installation options
It also doesn’t change that HA is not designed for a mobile phone, and phones are not designed to be plugged in 24/7 either
The title is Install Home Assistant Core
There are 4 installation options on that page, which one
Core, HAOS, Supervised, or Docker?
Okay so core
But this is the fastest way for students to get in touch with ha.
It’s also likely to break
Students are better off picking up a used PC and using that for HA
Since you get ethernet and USB built in
Use the phone as the controller for HA
@manic bane I solved my problem, but the version of ha is too new and my python doesn't fit.
So you don’t have python 3.10 then?
no,I'm Python3.10.3
But ha2023.7.10 doesn't work.
You want Python 3.11.4 now
Or just to use Docker, which s actually the fastest way to "get in touch" with HA
Heyo! Just wanted to check and see if you had any further ideas for me on this
hi everybody
If you could post your dmesg output, that may lead to more obvious reasons for that. Sometimes slow or erroneous storage or IO operations can cause file systems to go into readonly mode, but an NVMe SSD alone wouldn't fit that criteria.
someone are using the usb for radio frecuency?
What?
Many people use USB devices for Zigbee, Z-Wave, Thread, RF433, and more
Sure dude, I got you - added to the link, same as above (https://imgur.com/undefined)
It appears to spam more of the same? I get a huge rush of this
Oh RIP hold on
The beginning of the log will be most important
I can't post photos in here so there's not much more I can do
shelly use zigbee protocol?
Shelly devices are wifi not ZigBee based
The only thing i've done both times on these HA instances is expand the disk from 32 --> 64 GB while live, which from my VMWare experience is fine but maybe Proxmox doesn't like?
Post text? Using a code sharing site
hold up dawg
Nah, you're fine to expand disks under Proxmox and HA has no issue either
I can't copy and paste out of the VM, so re-writing all of that would be a hell of a thing :)
That's what I assumed, yeah
Good to know for sure!
Uh I might be nuts but, dmesg -T | less seems to only let me scroll through a certain range of outputs - do you know how I can scroll up to the top of the dmesg output? Or should I run the -T one THEN do dmesg?
Again, utter linux noob so I apologise for my likely obvious knowledge lack
I'm your average robocopy script kiddie and that's about it as far as terminals go
dmesg -T | curl -F 'format=url' -F 'content=<-' https://dpaste.org/api/
That'll put it all in a dpaste snippet and give you a URL
WOAH
Okay Mr Wizard
Hell I can't tell if it's posting these lines and breaking what i'm typing or just if that command didn't work
I wish I could paste into this terminal
ctrl+v, ctrl+shift+v and insert methods don't work in case that's the next suggestion bee tee dubs
Nah, you've got to re-type it if you're using the browser console client in Proxmox
Should be accessible just using the SSH addon though
Hmm. I did add the typical ssh addon in HA, should I try this there?
I mean, if you've got the browser console open already... this will work there, up to you, the output will be close
It looks like the addon doesn't have the -T option for dmesg, not the end of the world
Just wondering if anyone had any ideas about my issue? #installation-archived message
Feels like there's probably more to the story there... those are only 100 line snippets and on the second one, you can see there's an earlier error cut off at least
It can take a bit to upgrade some times as well... you mentioned you rebooted, you may have halted the progress
Oh, additionally - part of the issue I had with my old installation was that the DNS/network settings were cocked up, so the new installation also resolved those, and I've installed updates successfully to it. However, since the reoccurrence of this 'read-only fs' problem, I now appear not to be able to apply 2023.8.1 - it just drops back to the 'install' button again as if there's an error in starting the update (but it visibly throws none.) I was going to look into it after this problem.
Also also - my HA buttons all seem to shift upwards when this error is happening. All pictures attached.
https://imgur.com/XWCDROC
Maybe this is all part of the same problem? Not sure but thought I shold note it!
You ignored my command 😞
Hello. How to set up an old laptop for home assistant? Thankyou
You can start here: https://www.home-assistant.io/installation/generic-x86-64, Home Assistant OS would be recommended if you're new to Linux
I'm doing this on a 64GB card, I do not see the 'free-space' issue. But still stuck downloading.
No sorry that was the reboot as part of the install/update. How do I get more logs? That’s all it gives me when I hit download
What do you all host HA on?
Sensible people don't use a Pi
Sensible people also don't use a VM on Windows desktop
I'm surprised about the number of VM installs
ew
I might be setting up a proxmox cluster soon and was thinking of just hosting a vm on that for it
Yeah. Would prefer to go esxi for learning it better for work
That works too, not as many people do that, but it's a positive number
yep, free version doesn't have a lot of features I want to use though, like clustering. oh well, that is a separate conversation
thanks for the tips, i really appreciate
Maybe more of a correlation of people providing analytics on a VM vs people on a container?
Hi, I installed an update in HA and it became unavailable. I flashed my SD card with rpi imager and restored via backup (tar file) but home assistant just disconnects and is unreachable via WiFi. Has anybody ran into this issue?
Is there much space requirements for a kind of small home installation or is the space excessively growing over the time?
I would also use a 32GB Card for my install on an RPI.
Hard to say. Depends on configuration, addons, number of entities, how often they change their value and so on.
The short-term history/log data is only kept for 10 days by default: https://www.home-assistant.io/integrations/recorder/#purge_keep_days
SD cards are relatively cheap and it's not that hard to migrate. Maybe add some some disk related sensors to your dashboard and take action if needed
Thanks for the feedback! Would you directly install HA on the RPI or is there any deeper sense in deploying it via Docker or something?
I use the docker/container installation method for the flexibility/freedom and because I'm not a fan of HAOS but most people seem to be happy with it.
I'd suggest you try out HAOS and see if it fits your needs
I think I’ll try the direct install to keep it simple 😉
Has anyone tried to install HomeAssistant in Asahi Linux (Debian on Apple Silicon)?
I have it running but getting some errors popup in the cli - seems to mostly be around https (which is not currently confirgured)
"some errors"?
yeah a bunch of these: File "aiohttp/_http_parser.pyx", line 557, in aiohttp._http_parser.HttpParser.feed_data
aiohttp.http_exceptions.BadStatusLine: 400, message:
Bad status line "Invalid method encountered:\n\n b'\x16\x03\x01\x02'\n ^"
everything seems to be working (other than MFA) though
are you running core? or container?
You can run a virtual machine of HAOS. But it sounds like you're running a VM and trying to run HA Supervised in that VM
no I am not running a VM I am running Debian ARM64 native on Apple Silicon
because there is no container support for Debian on Apple Silicon yet
https://www.home-assistant.io/installation/macos
Installation on macOS is supported in a VM or through core, but not core inside another OS 😉
well first you talked about asahi and now its debian 😛
and I am not running supervised I am running core
I am not running MAcOS lol - not sure how many time sI need to say NATIVE debian ARM64
you didn't try what I suggested
I didn't you suggest anything - all I saw was you ask about docker... let me recheck
this all points to some python lib not liking apple silicon yet, so use a container and get rid of that problem
I cannot replace Debian with Redhat (podman) because my machine is not just for homeassistant it is also my 3d printer server
itis not a problem I can just run the native MacOS version on my Mac Studio instead - I was just trying to get some more utilsation out of my M1 Mac Mini (which is the one running debian)
no I am just checking out what the best options will be
I already discovered core is not the best idea because I can't run any addons so I will need to rethinnk things
nobody needs addons
I do for my merross bulbs they are nto being picked up by core
so?
I managed to get all my apple devices, my apple tv and my lg tv but none of my lights
well if I am going to use HA to replace Apple Home, I need to be able to get all my devices 🙂
every single addon is just a docker container or piece of software that you can install on your own and was usually available before there was a addon
I literally started looking at this 1 hour ago - so I haven't read everything (or even close to everything) yet
I was mostly just curious why I get that error
you'll probably find the problem deep down in some py http library, nothing you can change there unless you change the environment
yeah probably - I was expecting some issues running on Apple silicon - given how many issues I had getting klipper to run on it
I'd do a arm debian vm on macos, thats probably the best you can do rn
I will play around for now and see how things go - if I need to move it to another machine it is not an issue
I can probably run it in unraid if I need to - I expect there is a container for it
Sorry! I didn't, I was starting to work through it - but then time ran away from me last night and I didn't do any of the above. Re-attacking it tonight though so I'm sorry for the blue balling but at least I can jump back in now!
Okay, checking back in on this! So - should I just leave the -T out or replace it with something when using the in-browser client?
I ended up hard resetting the system and am now trying it in both
Added some new screenies: https://imgur.com/E0fQzz8
I'm not sure if either command worked correctly
I've added the top and tail of a dmesg as well which I ran in the ssh console in HA
so quick question - the installation instructions for core tell you to start the service with just: hass
Obviously this is not ideal as everything is dumped to stdout and seems standard debugging level is set quite high as well and catches warnings.
So what is the preferred method of starting it and can I change the debug level to only log errors?
usually people start it at boot...
and the debugging level? cos I am getting thousands of warnings currently - I don't want them logged
getting signature error warning for all my meross bulbs
but they all work fine so there clearly isn't an issue
logger
You forgot the second -F and they are ' not `s
ahhh Gandalf is here
systemd - and the forum has a sample unit file
Hey I'm super smart and capable and strong and capable and smart
Alright nabbed it, is this something I should be keeping private due to sensetive info in the output?
I wouldn't imagine there is much too concerning there, there's hardware identifying information and things like that... unsure if a public IP would wind up in it
My public IP should rotate soon anyway, so I'll take that gambit
Thank you again by the way
Looks like, with my very uneducated eye, either
[Mon Aug 7 12:19:11 2023] acpi PNP0A03:00: fail to add MMCONFIG information, can't access extended PCI configuration space under this bridge. or
[Mon Aug 7 12:19:11 2023] pci 0000:00:01.1: can't claim BAR 4 [io 0xd300-0xd30f]: address conflict with PCI Bus 0000:01 [io 0xd000-0xdfff]
[Mon Aug 7 12:19:11 2023] pci 0000:00:01.2: can't claim BAR 4 [io 0xd2e0-0xd2ff]: address conflict with PCI Bus 0000:01 [io 0xd000-0xdfff]
[Mon Aug 7 12:19:11 2023] pci 0000:00:03.0: can't claim BAR 0 [io 0xd280-0xd2bf]: address conflict with PCI Bus 0000:01 [io 0xd000-0xdfff]
[Mon Aug 7 12:19:11 2023] pci 0000:00:05.0: can't claim BAR 0 [io 0xd240-0xd27f]: address conflict with PCI Bus 0000:01 [io 0xd000-0xdfff]
[Mon Aug 7 12:19:11 2023] pci 0000:00:08.0: can't claim BAR 0 [io 0xd200-0xd23f]: address conflict with PCI Bus 0000:01 [io 0xd000-0xdfff]
[Mon Aug 7 12:19:11 2023] pci 0000:00:12.0: can't claim BAR 0 [io 0xd2c0-0xd2df]: address conflict with PCI Bus 0000:01 [io 0xd000-0xdfff]
might be a clue
That's only less than 2 minutes of information. Were you seeing the readonly errors between 12:19 and 12:20?
Not that I know of, no
I certainly didn't set anything myself, if I did, it was entirely accidental
I can retry the command?
Are you seeing the errors now?
You want to get the output after the errors start to occur, reviewing the dmesg output should hopefully provide more insight into the trigger of them
Ah nuts, yeah right I get it now. I restarted the machine because none of the consoles were reliably responding to me, which cleared the fault, which means this report is less useful. Hell!
Okay good to know though - I'll have the command ready to go for when they start again (probably by tomorrow night)
I do have a couple of light errors in HA's terminal but likely not a biggie - either way, adding to the ever-growing imgur link
Hello, I'm currently running version 2022.11.4. I'm trying to update but the web ui say there is no update available. The command ha available-updates return an error 403. I'm using hassos, updated some days ago at version 10.3. Any idea what's wrong?
Is anyone able to help? I just loaded virtual box on the new windows pc and tried to restore home assistant from a previous backup (.tar file) but it failed to fetch. Is there a chance the backup is too old? It’s from January 2021. How do I even log in to start over if I have to? Do I just create a new login username and password?
You can set it up with a new account and upload the backup through samba. It's recomended for bigger backups: https://www.home-assistant.io/common-tasks/os/#restoring-a-backup-on-a-new-install
There have been quite some breaking changes since January 2021. Chances are something might be incompatible
I just installed 10.4 on a pi4. How do I get past the cli so I can install ssh?
That's not how this works with HAOS. You probably want to go here and install the community SSH addon: https://my.home-assistant.io/redirect/supervisor_store/
What were you planning on doing via SSH?
ssh in so I can go headless. I tried the advanced ssh addon and it didn't work so I tried from putty and a linux account which also failed. I assumed because ssh wasn't installed so I hooked up a monitor and keyboard.
it didn't work
Can you elaborate on that?
Once HAOS is properly installed you usually only need a browser after that.
The host's SSH server is not started by default as it's meant for developers only: https://developers.home-assistant.io/docs/operating-system/debugging/
The SSH addon runs in a container and is meant for non-developers.
Give me a minute to get it running again. Ultimately I'm trying to get hacs installed which requires running wget from a shell.
ok, I get this followed by 502: bad gateway
Advanced SSH & Web Terminal
The add-on seems to not be ready, it might still be starting. Do you want to try again?
What do its logs say?
Ah, I didn't see the configuration tab.
[01:43:55] INFO: Starting the ttyd daemon...
[01:43:55] FATAL:
[01:43:55] FATAL: Configuration of this add-on is incomplete.
[01:43:55] FATAL:
[01:43:55] FATAL: Please be sure to set at least an SSH password
[01:43:55] FATAL: or at least one authorized key!
[01:43:55] FATAL:
[01:43:55] FATAL: You can configure this using the "ssh.password"
[01:43:55] FATAL: or the "ssh.authorized_keys" option in the
[01:43:55] FATAL: add-on configuration.
[01:43:55] FATAL:
do I use the user and pw I used to set up haos?
Use whatever you want.
Once configured you can try putty on the configured port again. The web terminal can be a bit infuriating to use
it seems to go to the cli again. Can I run wget from there? And I see what you mean.
it's working well from putty now. Thanks. Trying to get sonoff devices working. I thought it would be easier.
Hello everyone, i hope this is the right channel for my question.
So, my homeassistant is on a really old version and i wanna update it. My question is, do i have to update Homeassistant Core or OS first or does it not matter?
It doesn't matter, but
take a backup first
and download that backup before you start the updates
ok, thank you
I'm unable to install HomeAssistant Supervised in a Debian 12 running in a Raspberry Pi 4B. I am hitting the systemd-resolved error every time
Anyone was able to make this installation? I need Supervised on linux since I am planning to use the same RPi for other stuff
And ... there we have it
Doing that is very clearly against the Supervised requirements, and will result in an unhealthy and unsupported system
Things will break
Ok thanks
HI, there. I am running ha in a docker on a NAS. I got myself a PI4. Now I am searching for a link on a howto, how to transfer the running system from a docker to a dedicated Pi4. Maybe someone can help me to find an instruction to do that...
Take a backup, download the backup, restore it to the new install (assuming you're going to HAOS)
Thanks!
sooo, now it shows no internet connection for the supervisor, is there an easy fix for it or a way to see whats wrong with it?
i also got those here
[supervisor.resolution.evaluations.base] Systemd journal is not working correctly or inaccessible (more-info: https://www.home-assistant.io/more-info/unsupported/systemd_journal)
[supervisor.core] System running in a unsupported environment!
[supervisor.jobs] 'Updater.fetch_data' blocked from execution, no supervisor internet connection
[supervisor.utils.codenotary] Timeout while processing CodeNotary
ERROR (SyncWorker_3) [supervisor.docker.interface] Can't install ghcr.io/home-assistant/raspberrypi4-64-homeassistant:2023.2.3: 500 Server Error for http+docker://localhost/v1.41/images/create?tag=2023.2.3&fromImage=ghcr.io%2Fhome-assistant%2Fraspberrypi4-64-homeassistant&platform=linux%2Farm64: Internal Server Error ("Get "https://ghcr.io/v2/": context deadline exceeded")
I'd reboot the device
Somebody who uses and knows HAOS will have to help from here then I'm afraid
is there anyone i can directly dm or is there a channel for special help or anything?
DMing people isn't going to help, and this is that channel
ah ok
There's no hidden "special help" channel
ok, wasnt sure
i was more thinking of a ticket channel
No tickets
No formal support channel that you pay for
It's us - other users of HA - helping people
ok, thank you for your help
The problem I seem to have is that when these errors are being spouted out, HA becomes pretty unreliable. The Terminal plugin won't load now and I can't restart it, either. It's so strange, especially for a brand new setup (and even more so, one that I didn't have a hand in to mess up)
Added to the imgur, https://imgur.com/GUxYHue
Suuuuuuper weird. I feel like I just keep having dodgy luck with this. VirtualBox threw kernel hangups and froze, baremetal was too cumbersome to do, well, this with, and now this on proxmox which out of everything so far should be the most optimal setup
Is the console unresponsive as well? Not the SSH add-on
No it's responding (albeit spamming more read-only errors) and in fact by pure chance I typed the whole command in correctly last night and when I hit the up key just now it picked it up in full and seemed to work - how do I scroll up in the console window, again, please?
I saw it post the dpaste for a half second
read-only errors
?
Check me lil' ol' imgur Tinkerer
That sounds like an issue with the SD card/SSD/whatever
Full history of the faults seen so far are there
Hmm. I don't think it should be faulty - it's reused but not that old and certainly shouldn't be near it's nand r/w limit
Is there a fancy command you can use to check SMART or something similar?
I see one image at that link, nothing about read only
Oh it should have history too - hold on I don't use imgur much, let me grab the lot if I can
Read only file system
Google says you should be able to hit Shift+PageUp for the console; however, I'm not having luck w/ that myself
Apologies NSX I didn't realise linking you one image didn't show you others, I should have been linking you the post instead
Intel 10th gen NUC, 500 GB NVMe SSD
It's an older model but not heavily used, maybe 2-3 years old - it's all the same hardware I initially provisioned in it so about the same age
Well, that rules out the usual Pi problems
Yeah, I would have thought it would be quite passive and non-problematic
I mean, it could be bad NAND or perhaps a bad controller... I remember some of those gigabyte sticks being dodgy
but I doubt it, it'd be the first
Could I perhaps clone the NVMe drive to a new stick and plug the new one back in it to keep everything as it is but replace the drive?
Read only file system is 99% down to failing hardware or power issues
It could be other things, but it's damn rare
Sorry - yeah, nor am I
It doesn't respond to any 'control' commands, only regular text input
what a pain, I thought the whole point of setting up a VM host was to resolve these "can't plug into the device, can't see it's output direct on hardware" problems
Well shoot
I guess that's the go then... Clone the little bugger
Proxmox isn't likely to have, like, hardware/driver issues with a new disk?
Nope
I'm going to avoid the same unit, I think I have a spare Intel or WD drive around somewhere
Of course, if the disk is failing you may not be able to clone it
@runic venture how did you install HAOS?
Oh I'll clone the little sucker if I have to tear into the flash itself and-
Initially, by hand with the .iso, then I had this issue
THEN, I used the script on the HA page
Hold on, I'll find it again
ah
It's old but worked a treat and appears not to have much in the way of a version specific setup
Oh, also, hi tteck, good to chat to you
The image of the hard disk settings are not from the script.
I've seen your name about a bunch - I've been hoping to set up HA again proplerly, and then Frigate (until all this crap began)
Ah, some of those may be from my previous VM which I shut down due to this fault - let me take one of the current setup. Will just the hardware page work or should I go somewhere specific? [edit, yep vm 100 is the first VM which has my old setup on it, VM 101 is the current]
Added two to imgur the link (I hope they show up right away?)
I just did it again and looked really quickly with my quick looking eyeballs. Here you go when ready: https://dpaste.org/aLfu7
Hi how can I install this on my installation? https://github.com/wlcrs/huawei_solar
The #integrations-archived channel can help with that, but if you scroll down it tells you
It very well may have rolled over the actual issue, since the first line is after the issue already occurred
I'd give swapping out the HD a shot and see how that goes
Aw hell, yeah it looks that way
Alright guys, thank you both for all your help - I'll probably be back in another 6 weeks when I can get back to opening this little bugger up xd
I'll absolutely ping you both to let you know if a new, cloned disk does the trick
Anyone got any thoughts on the issues I'm having? or any tips on how I can capture more logs to help? #installation-archived message
Paste links are dead
I can't update HA since last version on 2023.8. What can I do?
What do you mean "I can't update"
I'm running home assistant container core-2023.7.3
I know 2023.8 has been released but it does not appear
🥹
I know... documentation... nobody bothers reading it 😛
Sorry Forgot about the paste links - before the upgrade http://pastie.org/p/3Jrt4N0GK9QYk7AM16To0A and after http://pastie.org/p/7HYraOOq44EGSIqEi3mVXE
Ah, I remember
Those logs aren't complete
Are you able to attempt another upgrade?
how do i get complete logs? and yes I can - it says 10.4 is available now
No... Hahahha I'm new in docker. I do not know too much how to operate with it
I think you mentioned you were also restarting the device, which isn't something you should necessarily need to do. How long were you waiting prior to restarting
I always had HA in a raspi
no sorry i ws talking about the restart that happens as part of the upgrade. I haven't had to force a restart
Got it
I can try the upgrade again now but how do I capture the full log? I have also enabled SSH into the host so can grab it from the filesystem if you can point me to it
Are the 100 lines even when you hit the Download Full Logs button?
I know you can get access to the full supervisor logs directly from the host, but I don't think you can via addon
what I pasted was what I got when hitting the download button. I have both the ssh addon and ssh host access. I think I should be able to find the full log via the host but not sure how to - my linux skills are only enough to be dangerous
via 22222?
yes via 22222
journalctl "CONTAINER_NAME=hassio_supervisor" will get you the full logs
aaah yes much more there. Do you want what I have since the last restart or should I attempt the upgrade again and then share what it produces?
You could scroll around to where your prior upgrade was and see what you find
You'll probably want some prior to the last restart, since it likely failed somewhere
hmmm i can't actually find those lines. I'll follow the log and attempt the upgrade and share the output
okay so this is something that sometimes happens - when the install reboots I've just tried to SSH back in and its telling me the host key has changed - is that expected?
logs I have so far are http://pastie.org/p/0QR7uLOgJGi7VPZbtGgiTk
Following the boot I'm still on HAOS9.5 new output from journalctl is http://pastie.org/p/1UeJpma6thNSfLFmKzJe8Q
I feel like there's a gap in the logs? I'm a bit lost
hi is there someone who can help me setting up my WiFi using nmcli like in https://community.home-assistant.io/t/guide-connecting-pi-with-home-assistant-os-to-wifi-or-other-networking-changes/98768/98
ah you're still following the guide and trying to do wifi...
if that's the case, then you DO need the community SSH & Terminal addon
It does gain you downsides
I will switch back to ethernet when I have played around with it, i will add multiple interfaces for sensors, like p1, zigbee etc, I want to play around with the devices on my desk. as well I want to get familiar with how flexible Home Assistant is.
You’re wanting to gimp yourself to test with?
Plug in a USB extension cable and then you can add your "interfaces"
USB extension should be around 5 meters long 😛
Seriously, there's nothing to gain by dropping it on your desk
literally
there's literally nothing that you're going to plug into your pi that requires it to be right next to it, zigbee included
5 meters is well within zigbee's range
the extension cable is to get the zigbee controller away from the noise of the PI
to extend it's range
(see the video pinned in #zigbee-archived )
there should be zero reasons you need to have your HA box next to you
even with DIY sensors.
maybe your right with it not being nescessary....
it's not, and you're wasting your time
and you're gimping your system
there's a reason the guide as the following note:
I am not wasting my time, I am seeing how it behaves when I want something for no reason
just because
@fallen lichen were my logs that bad?
From the logs, it looks like the upgrade was fine, but I'm wondering if the boot then failed. There's 2 slots w/ HA OS and if one slot fails it'll fall back to the prior install
that sounds familiar actually - I forget the command someone gave me but I did notice it would sometimes swap between slot A and slot B? I didnt understand the significance
Is there a way I can check that? or find out why the boot is failing?
Perhaps try ha host logs -b -1 Not terribly familiar with that, though
not seeing anything there
sorry for the text but it was the ha os info command that gave me the boot slot
ha os info
board: rpi4-64
boot: A
data_disk: KINGSTON-SA400S37240G-50026B7282D60EE6
update_available: true
version: "9.5"
version_latest: "10.4"
trying the update again and will see if the boot slot changes
got the complete before and after log this time and it didnt switch slots. http://pastie.org/p/1AXmv5sgFwTe2jkjGkdXMB Interesting it says upgrade to 10.4 worked but then after boot it is still on 9.5 and now I have apparmour issues again
I am going to have to call it a night for today but if you have any ideas happy to try them and let you know tomorrow
This seems similar, nothing promising for a workaround though: https://github.com/home-assistant/operating-system/issues/2522
Hi I have a mariaDB database for Homeassistant running on the Proxmox server. Only I forgot to limit the retention days?
How can I get the database now smaller without deleting the power consumption data with it?
By default it's 10 days
Haha yup that’s similar because I raised it. Sorry thought I had linked that in a post
Gotcha
Your comment of reflashing with 10.1 and then restoring sending you back to 9.5 doesn't seem to make too much sense though. To my knowledge, there wouldn't be anything in the backup that'd cause OS changes
I'd say maybe try that again w/ the latest OS
is it weird that i dont get a option to backup in the onboarding page of home assistant
Not really. Nothing to back up
i meant restoring a backup
poor wording on my end
You can restore after onboarding too https://www.home-assistant.io/common-tasks/os/#backups
Hey! I've got an Orange Pi 5 on the way (more available than a RPi and much more powerful)
I'd like to imagine I can follow this guide: https://www.home-assistant.io/installation/raspberrypi
But I have a feeling it may not be quite that easy as I have different hardware.
What is the best way to get Home Assistant running?
thanks
oh, add-ons are built for rpi only and not available for other platforms?
that's a really big shame, I guess I miss out on all this? https://community.home-assistant.io/tag/hassio-repository
maybe I wasted money on the orange pi :(
They are on other platforms, but for HAOS and one other unholy install method only
Not for Docker (but add-ons are just fancy Docker containers)
Why not mini pc
orange pi is silent / efficient like a rpi
mini pc like optiplex as far as I know has fans that spin up
given it's going to be running in my living room all day every day, figured SBC was the way to go
Yea that’s what mini pcs do
https://www.ebay.com/itm/385670237223 link i've shared previously for an $80 mini pc "m910q" there are many variants on ebay for reasonable prices with 8gb ram or higher and 128gb ssd/nvme or bigger
Runs circles around any sbc and cheaper
can you install HAOS on those?
I friends, I need help. I've been using Home Assistant for many years, and have gone thru many many updates and different setups. Started on on an Raspberry PI and now running on my larger file server.
A few months back I had a catastrophic failure of the drive that hosted my HA and all configs were lost (with no backups). I spent the better part of a week getting back up and running on the new VM OS option. All was working well until I got home from a vacation yesterday and my lights were not responding. yada yada, somehow the entire VM working directory was gone. The disk, everything. Ok, fine - fool me once shame on you - I have a full backup from a few days ago.
I just downloaded the vdi again, went to homeassistant.local (didn't work for some reason) and then the IP address. I created a dummy user, uploaded the backup and hit go (btw - sucks that there is no progress). I used "supervisor logs" over and over until it looked like everything was ready. Alas - no login.
Restarted HA, no login.
I've done the same procedure just restoring the "core" of HA, to the same result. I can't seem to see the full "core logs" because it's cutoff in my VM window and I'm not sure how to do a "more" type of view of that, nor how to ssh into the system without any access to HA
Morning all. trying to get started, looking for some help. Rpi3 installed from homeassistant website instructions. HA is up and working. Now, I'm trying to get a BME 280 running using an adafruit feather huzzah. Their install guide suggests using hass.io, but I don't see that on the install. Tried the Rpi install from Adafruit's Hass.io instructions but it failed. Back to the HA instructors which worked fine. Question, then: what the heck is hass.io, how is it different, shoudl I use it, can I add it to a vanilla HA install etc.
Hass.io was the old name for the appliance like install option, that uses Docker. It was renamed in January 2020 and is now called Home Assistant OS.
good bot
If you link to the instructions we can tell you which channel you should be asking in
(to get working with the AdaFruit stuff)
this is what I am trying to do but the instructions seem out of date: https://learn.adafruit.com/integrating-adafruit-io-with-home-assistant?view=all
Which bit are you stuck at?
Happy to offer a beer or two for anyone who can help me out with this. I am 100% lost 😢
The instructions say I should click on hass.io tab to start installing. that tab doesnt exist. I've skipped ahead to trying to get ESPhome runing on my feather, but I cant open the USB serial port from my mac.
AAAnnndddd I figured it out. Needed to connect via https after the restore. Still confused by homeassistant.local doesn't work (on either http or https) but I'll work on that next. Ooof I miss when I had full access to everything and it wasn't shielded by a VM
mDNS is fun
Ok, you're looking for #add-ons-archived
That menu moved years ago
Somewhere under Settings now
Yeah and "configurator" is now called "file editor". It's very frustrating trying to follow an online tutorial that hasn't been updated to the current state of things.
Is there a channel for adafruit stuff?
thx
Is there a way to fix this? I m running an openvpn server directly on the host os further its clean.... Im getting the message unhealty system not priveliged
Are you running Supervised or HAOS?
Supervised
On debian 12
Recently installed this openvpn server for outside network acces since its not available in the addon store sadly
It does not touch docker or other weird permissions thing so this is weird to me
Well, the requirements are quite clear that you can't run other software outside of add-ons
Is there a way to run custom docker containers from docker compose file?
If you're comfortable doing that, just make an addon
Would suck to lose the addons inside home assistant
@brisk willow made one so it can't be THAT hard
gottem'
But you also have to maintain it
You can't without making your installation unhealthy: #integrations-archived message
What do you want to run?
Pritunl openvpn server
They have a docker image available
I m not that familiar with custom images can get them running but more then that is magic 😅
Hey Guys, I am trying to install HA on a Synology NAS using docker following the official instructions. I am stuck in the setup where you specify your location. It cant find my location because the request fails because of CORS (Cross-Origin Request Blocked: The Same Origin Policy disallows reading the remote resource at https://nominatim.openstreetmap.org/search.php?q=berlin&format=jsonv2&limit=3&addressdetails=1&accept-language=en&email=abuse@home-assistant.io. (Reason: header ‘user-agent’ is not allowed according to header ‘Access-Control-Allow-Headers’ from CORS preflight response).
(I am accessing my NAS from my home Network (http://192.168.x.x:8123) could that be the Problem?
Any Idea how I can fix this or progress in the Setup?
Okay, that seemed to be a Firefox Problem. Chrome on Android had no Problems...
if you don't like the way Tediore drives, stay off the sidewalk!
what to do if homeassistant-superviced depends on systemd-resolved
and there is no package available for debian 11
You should be using Debian 12
wtf?
no you can't install !ANY! homeassistant-supervised on debian12
configure script tells me debian bookworm is not supported
https://github.com/home-assistant/architecture/blob/master/adr/0014-home-assistant-supervised.md
Debian Linux Debian 12 aka Bookworm (no derivatives)
installed homeassistant-supervised package post-installation script subprocess returned error exit status 1
Errors were encountered while processing:
homeassistant-supervised
There's a long forum thread on it
ok would you please share the url?
The assumption in the requirements though is that you're an expert if you're going to use it
You can find the community forum here
There, somewhere...
I don't memorise every one of the hundreds of thousands of topics
would be nice if the "devs" of the homeassistant-supervised.deb postinstall bash script would be too...
pullrequest incoming...
Brace for impact!
haste dir das teil mal angeguckt? das ist grausam
./giphy eats-popcorn
No, I haven't. But I haven't bothered with Supervised.
If I want it comfortable, I use HAOS and if I want to have more control, I use Container.
What OS are you using? Also see here: https://github.com/home-assistant/supervised-installer/issues/304
@raven forge after the release of bookwork, i upgraded all my machines (webservers, hypervisors, vms etc.) among them was also the homeassistant-supervised instance. which was not supported until then.
now i know the main issue is the "systemd-resolved" and the way the postins script exites after there is no systemd-resolved available
I went on vacations for a week and during that time I lost access to my HA installed in my Pi. I have access to it at this moment after I disconnect it from the plug but I'm getting a lot of errors everywhere. For example, I noticed that I had a problem with my Tapo camera and so Frigate was not working. When I try to restart the addon, I get this message: Failed to restart add-on
500 Server Error for http+docker://localhost/v1.42/containers/addon_ccab4aaf_frigate/json: Internal Server Error ("readlink /mnt/data/docker/overlay2/: invalid argument")
after i extracted the deb file and debugged the script, i noticed there is some problem with the systemd-journal-gatewayd.socket
another example of an error, there is an update to Zigbee2MQTT and when I try to install it I get this popup in the bottom left corner of HA: Failed to call service update/install. connection lost
so i purged the config and reinstalled the package! so the postinstall script exited with 0 and i was able to install the homeassistant-supervised in the newest version. AND tbh, it would be nice if the script would do something sensible with the exit codes, instead of pipe them to /dev/null... this would be save hundred of thousand forum topics!
There are two updates for HA Core and HA Operating System, the first is from 2023.7.3 to 2023.8.1 and the second one is from 10.3 to 10.5. When I try to install any of those, it says "nstalling" and after a moment it stops and the updates are still there
the same message shows up: Failed to call service update/install. connection lost
When I try to "use an existing virtual hard disk file" when creating a Virtual machine on Virtualbox, and I try to to find the zip file from the website, I get this error: Failed to open the disk image file C:\Users\musta\Downloads\haos_ova-10.4.vdi.zip.
Could not get the storage format of the medium 'C:\Users\musta\Downloads\haos_ova-10.4.vdi.zip' (VERR_NOT_SUPPORTED).
Result Code:VBOX_E_IPRT_ERROR (0X80BB0005)
Component:MediumWrap
Interface:IMedium {ad47ad09-787b-44ab-b343-a082a3f2dfb1}
Callee:IVirtualBox {7682d5eb-f00e-44f1-8ca2-99d08b1cd607}
Callee RC:VBOX_E_OBJECT_NOT_FOUND (0X80BB0001)
sounds like you need to extract the archive at first.
i think that is the reason why i have to listen to "you're an expert if you're going to use it"
any idea about the issues I'm having? It seems that nothing is working correctly. I didnt' made a single change to my HA for months
It might be a dying microSD card
hmm ok. Is there something I can do to confirm that?
@spiral stag check dmesg -T
Flash another SD with HAOS and restore a backup (you have a recent not corrupted one?) - if it runs without issues, the other SD is propably toast
Yet another pi victim
Eventually it becomes available
I understand that
on that topic, if I'm understanding correctly... should I change my Pi for which solution?
one of the things that I don't want is to have something that uses too much power. The idea of having a 24/7 device only works for me if it's not expensive
In my experience, you'll have great results with a used small-form-factor PC, like a dell/lenovo/hp "litre class" box.
They are <10W idle. See https://www.servethehome.com/introducing-project-tinyminimicro-home-lab-revolution/
Can get the from ebay for like $80
I've got two of the lenovo thinkcentre m910q, and they're great (including remote management), but I'm sure the other brands are similar.
https://www.ebay.com/itm/385670237223 link i've shared previously for an $80 mini pc "m910q" there are many variants on ebay for reasonable prices with 8gb ram or higher and 128gb ssd/nvme or bigger
Paste from SHS
I like running proxmox on it natively (it's a virtualisation platform, allows you to run multiple things on the box), and running HAOS in a VM. But you can also install it bare-metal if you like
I didn't have HAOS initially in my PI and I got many problems
but having it in the VM is the same right?
Basically yes. Tiny bit of overhead, but way more flexible, easier to manage, easier to backup. Can run lots of other things on your box...
yes, that is great
anyway, the Pi's are great but I see that I will run into these kind of issues again right?
maybe my card is not so good also
eg https://imgur.com/Nk1HDJP is my current proxmox1 box
Pis are fine, but they are very low CPU power (so things are slower, and there are things you can't do). SD cards are super unreliable, and pi's are also very sensitive to power supply issues.
ok thanks lyricnz
I still use a pi to drive my 3d-printer, but my HA is now on a VM on my proxmox machines
at the price the pi's are, I can make the switch without paying more
my only issue is really the power consumption
it's at least twice the consumption of the pi
I measured my box at about 8W, semi-idle (like HA).
do you have frigate?
It's less than a light-bulb. and about 1/400th of your AC
well yes, of course. I'm just saying that the low consumption of the pi was a plus
I don't have/use frigate, but it's pretty cpu intensive (unless you have accelerator). A SFF box can do a few cameras native, but uses lots of CPU
Yup, I understand. My internet connection (starlink) uses probably 5x my proxmox box in power.
If you want to stay with a Pi for now, you should at least consider a SSD and a powered USB hub for the SSD
Yup
ok, again, thanks for your support. I will try to get the backups and try on another card and go from there
(my HA is accesible now, ~15min later...)
btw @tender wren, these are the backups I have
hmm I can't put pictures here
anyway, I have zigbee2mqtt, core, and other addons
I don't think I have a complete backup of everything right?
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.
If you haven't created one, no. The backups on updates are partial ones
my HA is not slow, I'm not sure if it's an sd card issue
If you haven't got a downloaded recent full beckup (you should always have one), you would have to create one (and hope nothing has been corrupted) and download that one asap
the fact that it shows that popup "Failed to call service update/install" seems to be another thing
And/or backup off the rpi (to google drive, a local file-server, etc)
clicking on Grafana or Frigate on the left menu shows this: Unable to load the panel source: /api/hassio/app/entrypoint.js.
this is really messed up :S
Home Assistant observer
Supervisor: Disconnected
I'm not an expert, but have you made sure systemd-journal-remote and systemd-resolved are installed enabled and running, and that systemd-journal-gatewayd.socket is enabled and running? Also that the HA OS-agent is installed and running. Those need to be in place before running the HA supervisor installer on Debian 12.
systemctl start systemd-resolved.service> /dev/null 2>&1;``` 🤦♂️
that is like taping the check engine light...
people tape those? I thought you just unplugged em
unplugged is with & instead of the pipe
no u
You’ve already been got
yeah he's old news
he's been git got
and his discord tag on github is wrong, what a noob.
Working on moving over to a NUC. If I put my full back up on a USB stick will a new install HA be able to see it to pull the backup?
^plugged into the NUC
you restore it via webui (or smb if its larger)
no need to plug the stick into the new machine
Oh oh ok I think I get you. Plug it into the machine I'm using to get to the webui
yes
sorry I'm attempting to eat dinner, but plug the usb into the machine you're browsing from and do yourself a favor and copy it to the machine, don't upload from the usb directly
That makes sense. Migrating from a pi to a NUC. Ready to go just need to get up the nerve to shut down my PI to free it's IP address. I'll copy the backup to the workstation I will hit the webui from thanks
That makes more sense than a USB stick
welcome to the big boys club
Ya two years on pi and getting nervous for the SD card and my growing system
pfft, thems rookie numbers
I just know the figuring out what USB port my z-wave stick is on is prob gonna be the only pissmeoff thing
If it was configured correctly there will be no issues
It probably wasn’t tho
Dev/by-id not ttyusb0 etc
Oh damn. That's great
Ugh. It asks if I want to wipe my system and restore this back up I confirm restore and nothing happens
it takes some time, and the ui doesn't always refresh
click the blobby dot thing to see logging
The backup is a .tar file
Thank you. But I forgot to activate the commented out points again. How should I proceed to get the database smaller again?
Blobby dot......
She's alive!!!! Ok well mostly. My z-wave stick isn't being seen I'll have to find the port.
O o o o! Z-wave stick found!
Holy ... Thanks all! All good this is awesome!
Any thoughts on what I should do with my pi4? 🤣
birdnetpi, adsb receiver, camera, stream converter, oh so much
you can also just put your zwave stick on it and place it better if suitable
Ohh adsb... Nice.
Ima look and see that or lightning tracker or software defined radio stuff but if not PIs are $ right now
Thank you But I forgot to activate the
hello i have an issue can someone help me, i have hassos on a raspberry pi 4 and it was working fine until suddently after a unplug and replug the raspberry pi the hassos is stuck on boot on "Waiting for the home assistant CLI to be ready" and its for hours soo nothing happens and i cant reach the raspberry pi over the internet
any chance your sd-card didn't survive?
mhh it seems fine
am trying to mount it on a linux system to see
Am not sure about this but looks like the sdcard Is corrupted
in gparted
it says unknown for everything exept hassos-boot in the file system column
i guess i will just reinstall hassos on another sdcard and lost all my entire config
💀
Backups are important.
yeah i will remember now
Sometimes it takes great pain to forge a memory.
exactly
I backup my haos to a nas, and I make copies of the vm image to another NAS
but the thing is... what actually caused that ?
am pretty sure the sd card is not dead
SD cards gonna SD card.
jsut corrupted

I've had it happen to me at least 6 times since 0.69 or thereabouts
We are on what 10.4 now...
But the SD card corruption only bit me until version 6.0, then my rpi 1B+ was no longer supported by haos
oof yeah thats sucks
Corrupt SD cards happen all the time, a little power blip and away it goes
But I moved to a VM and now I hang with the big boys
Dead SD cards also happen all the time, one bit too many written and kaboom
mhh power you think thats possible
Yes power can cause it. Crappy PSU can cause it
some peoples told me to put hassos on ssd on raspberry pi usb but... a ssd for this ? seriously ? idk
A Pi .... power is the cause of 90% of problems
you have a recomendation ?
If you attach anything to the USB ports you must invest in a powered USB hub
no mobile usb power ?
If you use a rpi this is the way. Make sure to use tested powered USB hubs
yeah i know that thx
Yes, buy a sff PC for cheap. Eliminate the headache of SD cards.
They are great to get started on
But for longevity.... upgrade when you can
mhh ok
HA VENV on ubuntu server, just updated python to 3.11 and installed 2023.8.1 from 2023.7.3, and tried to restart. it immediately warns about the custom integrations, but after that nothing seems to be happening after that. Been waiting for half an hour at least. Is there something I have not noticed?
What the logs say?
Sadly we're not mind readers (any more anyway, not after the last time we tried). Please share the YAML and any errors so we can see what you've done.
Only about the integrations. Is it perhaps trying to download something and not logging it?
Started it again few minutes ago...
2023-08-09 16:19:31.855 WARNING (SyncWorker_2) [homeassistant.loader] We found a custom integration sonoff which has not been tested by Home Assistant. This component might cause stability problems, be sure to disable it if you experience issues with Home Assistant
nothing after that
What about host logs and core
this venv installation, no supervisor here
and no frontend is available since it hangs after those warnings
Oh I clearly need coffee I thought I read VM
Should I change my logger: in configuration.yaml?
What is it set to now?
ah. changed it to default: info
It seems to be installing packages. I wonder how it is this slow...
2023-08-09 16:32:55.168 INFO (SyncWorker_2) [homeassistant.util.package] Attempting install of SQLAlchemy==2.0.15
but this slowness is not normal. might some kind of a dns issue I guess
It seems that pip is veeery slow
Ah. seems to be ipv6 related issue
need to do sysctl net.ipv6.conf.all.disable_ipv6=1
was not an issue earlier
Seems I'm still getting update issues involving sporadic 404 or 500 failures with pulling images from github. But this is new, the errors now show up in the Core log and popup and error. Before hand they were silent failures and only showed up in the supervisor log. I'll keep at my attempts, I know it works if I try enough times now, but it can be days of trying sometimes 😦
It's DNS issues if you're getting 500
Indeed, when I had this issue on a similar non-HA project recently, it was resolved when I set my router to use the 1.1.1.1 dns servers from cloudflare. For some reason it doesn't seem to have resolved the issues on HA though, despite it reporting that it's using the correct dns server, so I've been assuming it is a different issue.
Trying to run docker pull ghcr.io/home-assistant/aarch64-hassio-supervisor:2023.08.1
First two attempts have insta resulted in a single line: Error response from daemon: Get "https://ghcr.io/v2/": EOF
Third attempt is downloading stuff with periodic failures. This is how I resolved it the last few times, just keep trying until it succeeds then hit the update button once I can see in terminal that it's downloaded.
You could try ha dns options --fallback=false
What is that doing exactly? 🙂
I can indeed see from ha dns info that fallback is currently set to true
Ah, second attempt at docker pull has succeeded, will try a quicky update in gui...
Seems to be good, supervisor updated. I'll have a look into what this fallback does.
Still not sure what the fallback does but leaving it on for now. One thing I have done though is added the cloudflare dns servers directly to home assistant, so now when I run ha dns info it reports - dns://1.1.1.1 and - dns://1.0.0.1 in both the locals and servers sections. I'll see if I see any improvement having done that now and if not I'll try turning off the fallback thingy 🙂
Whelp, just tried to update an addon and same problem occured, I suppose I'll try killing the fallback.
It forces cloudflare dns when the ones you have set aren't working properly
It 'falls back' to cloudflare
So your setting it to cloudflares dns servers means its the primary and the 'fall back' that's all
hi, idk if anyone can help me. i'm locked out of my home assistant because i was trying to setup duckdns. I'm not great at networking. I just following this tutorial https://www.youtube.com/watch?v=jkBcGl7Vq8s
basically i restarted my home assistant. Tried to go to that url i made with duckdns but it says site can not be reached
Can you access it using https:// and your LAN IP/port?
you mean my main ip, not my ha ip?
And https?
https://192.168.0.42:8123/
nothing
Did you at least set up SSH access?
shit lol
and when using http:// it's also nothing?
i don't think so
yes
im not even sure what ssh is tbh.. i really shouldn't have dabled into something i wasn't sure about
I am pretty new to ha i apologize!
i installed duckdns and did everything in that tutorial. Once i restarted home assistant i couldn't access it again
on my phone though it select your home assistant server and it has the address put for duckdns
welp i guess im screwed. What do i do at this point? as my automations are still running
am i better off just reinstalling home assistant? This whole thing happened because all i was trying to do was add smartthings into home assistant. I guess i'm better off just leaving it.
Do you have a keyboard and monitor you can connect?
Yes
like a spare keyboard and mouse you mean or?
Well, the web UI doesn't work, you said
You didn't set up SSH either
So .... connect a keyboard and monitor
i'm already on a computer
@chilly basin 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).
right haha 😂
You're not locked out while you can access the console
Interesting. Do you think I should log into my synology nas and see if I can access the virtual machine?
Before I plug the keyboard into the nas that is?
@placid sage
And ... your NAS should give you a way of accessing the VM's console
Right... now go looking for the console
so basically create a new virtual machine with home assistant and delete the existing one?
i'm gonna be honest i have no idea what i'm doing
i did this
came up with like ipv4 address and home assistant url
http:/homeassistant.local:8123
Well, I know nothing about your VM platform, so all I can do is point you at the docs and hope you can follow them
no worries
i understand
i wish i wasnt so bad at networking
things just dont work. seems you have to be almost a network engineer to understand some of this stuff
When you open a port on your router are you supposed do something else after that?
This isn't networking 😉
Depends on what you're doing, probably not
literally all i did was open port to internet. assign static ip in ha. Installed duckdns and added external/internal urls in config.yaml then just restarted ha. next thing you know i can't connect
To check your configuration use the command for your install method:
with my low networking knowledge i probably have to just delete the virtual machine and start all over.
You should always do that after touching YAML
Also, the URLs can be set in the UI - and should be
The only bit you have to edit YAML for is for SSL
oh
Good news, YouTube videos are often horribly outdated and misleading
However, HA should have come back up, even with broken YAML
This is where the console would help you, a lot
As would SSH
i apologize for my ignorance, but what is ssh exactly? is it similar to command prompt on windows??
is it just a command line?
so basically just a networking command line essensially
right that would help lol
also ha will come back up, but doesnt mean i can access it
If it came back up you'd be able to access it
You can't access it with either http or https so it didn't come back up
How did you create the VM?
NAS
synology
@placid sage You probably entered something bad enough to break it
configuration.yaml
probably
(which takes some doing)
Then the IP address you think it has might not be the one that really is?
Can't you click connect in synology's GUI to access the "console" of the VM?
i did
but i try to connect to that ip
it wont let me for some reason
unless its something on my router i did wrong
i'm using ubiquiti and it's extremely confusing
note to self - don't install duckdns without knowing what im doing 😆
Well, knowing how to access your host with the web UI down is the key
The console will work, if you can work out how
that i have no clue how to do. I'd assume id connect a keyboard to my nas run something in a command line or something lol
No. You have access to the Synology UI, right?
i do yes
The VM manager should offer a console for the vm
Maybe it says terminal instead of console? I don't know
It would be in the details of the VM
Highlight the vm. Click connect?
yeah i've done this
And...
black screen. says home assistant at the top in large letters ... home assistant url http://homeasistant.local:8123
Yeah, thats the host console
Sorry core logs
error unknown command logs for ha
okay
bunch of errors
i'm not sure why its showing errors for esphome and everything presence one sensors
never trusting youtube again 😭
Esphome dashboard data: cannot connect to host 127.0.0.1:64878
core check
command completed successfully
videos you youtube are usually out of date by the time they are uploaded
didn't realize that part 😅
next time i should just ask on the discord lol
infact lots of guides are as well
it's hard enough for the guys to keep the HA docs up to date in line with changes
you might not think so right now, but you did it the right way
you read, you tried, you failed, you asked for help
oh for sure these guys know a lot more than i do thats for sure 😂
i tried 😆
yes, and that's a good thing
but failed
would have been nice if i didn't fail tho 😂
yes but then you wouldn't know how not to configure duckdns
i still don't know how to configure it LOL
yes but you know at least one way that doesn't work
true
beats trying the same guide over and over and getting the same results
😉
which I have done
😛
i assume at this point my only option is to delete the vm
i wouldn't say your only option
and reinstall home assistant since i screwed up the yaml file
but it might be faster
o
well i mean what other option is there at this point LOL
but that being said I've reinstalled many times when I could have just figured it out with some patience
when there is an angry lady who can't turn the lights on, reinstall is faster
go for it
